Aurum: Special Issue of ArcheoSciences 33 (Guerra &amp; Rehren, Eds, 2009) - full volume (16 MB)

Thilo Rehren

In this introduction to volume 33 of ArcheoSciences, we provide a brief overview of the use and abuse of gold over time, and its different aspects, from the mine to the objects, their use, analysis, and restoration. For this purpose, we focus on the papers presented in this volume, which originate from the Workshop AURUM: authentication and analysis of goldwork, organised under the auspices of the EU-DG Research funded project AUTHENTICO. The main aim of this project was to develop tools and expertise for law enforcement agencies to combat illicit trade in antiquities and to fight fraud and forgeries; to do so requires an understanding of the diversity of the cultural, technical and material manifestations of gold artefacts, and their very specific combinations and expressions. Some of the scholarly foundations of this endeavour are illustrated by the selection of the 55 papers, arranged in five topical sections, which are introduced in this text.

2009 n° 33 AUTHENTICATION AND ANALYSIS OF GOLDWORK Guest Editors Maria Filomena Guerra and Thilo Rehren P re s s e s U n i ve r s i ta i re s d e R e n n e s ArcheoSciences, éditée annuellement par le Groupe des Méthodes Pluridisciplinaires Contribuant à l’Archéologie (GMPCA), est ouverte à toute communication, en français ou en anglais, sur toute recherche originale et inédite traitant de l’application de diverses techniques scientifiques (sciences physiques, chimiques, mathématiques et sciences de la terre et de la vie) à la résolution de problématiques archéologiques ou de la mise au point de nouvelles méthodes. Les articles peuvent porter sur les méthodologies et leurs limites, sur des recherches scientifiques fondamentales ou des techniques spécialisées, sous réserve que leur justification archéologique soit clairement explicitée. ARCHEOSCIENCES, a scientific journal published by the “Groupe des Méthodes Pluridisciplinaires Contribuant à l’Archéologie” (GMPCA), covers the entire spectrum of disciplines in archeometry. Specifically, ARCHEOSCIENCES publishes, in English or in French, original work in physics, chemistry, mathematics, geology, biology, paleoecology applied to archaeological research. ArcheoSciences est publiée avec le concours du Centre National de la Recherche Scientifique et du ministère de la Culture et de la Communication. Elle est indexée dans FRANCIS (Vandœuvre-les-Nancy, France), BRITISH ARCHAEOLOGICAL BIBLIOGRAPHY (Londres, GB), ANTHROPOLOGICAL LITTERATURE (Cambridge, États-Unis), ART AND ARCHAEOLOGY TECHNICAL ABSTRACTS (États-Unis), GETTY CONSERVATION INSTITUTE PROJECT BIBLIOGRAPHIES (États-Unis). ARCHEOSCIENCES is published with the financial support of the Centre National de la Recherche Scientifique and the french Ministère de la Culture et de la Communication. This journal is indexed in FRANCIS (Vandœuvre-les-Nancy, France), BRITISH ARCHAEOLOGICAL BIBLIOGRAPHY (London, GrandeBretagne), ANTHROPOLOGICAL LITTERATURE (Cambridge, USA), ART AND ARCHAEOLOGY TECHNICAL ABSTRACTS (USA), GETTY CONSERVATION INSTITUTE PROJECT BIBLIOGRAPHIES (USA). Rédacteurs/Editors Philippe Lanos (CNRS, Rennes/Bordeaux), Benoît Devillers (université de Montpellier) Rédacteurs adjoints/Associate Editors Claire Delhon (CNRS, Valbonne), Stéphan Dubernet (Université de Bordeaux), Matthieu Le Bailly (Université de Reims), Joséphine LesurGebremariam (MNHN, Paris), Ina Reiche (CNRS, Paris), Vincent Serneels (Université de Fribourg), Jacques Thiriot (CNRS, Aix-en-Provence), Yona Waksman (CNRS, Lyon) Le Comité de Rédaction, qui se réunit en moyenne tous les trimestres, est constitué par le Conseil élu du GMPCA (Cf. ci-dessus, rédacteurs et rédacteurs adjoints). Chaque proposition d’article est soumise à deux rapporteurs choisis au sein d’un Comité de lecture constitué du Comité de Rédaction et d’un Comité Scientifique comprenant notamment les personnalités suivantes : The Editor committee, which brings together every quarter, is made up of the board of directors of the GMPCA (editors and associate editors above). Each manuscript is under the responsibility of an editor and is evaluated by two scientists of known competence in the relevant field, notably members of the Scientific committee: A. Bouquillon (Paris), J.-P. Bravard (Lyon), O. Buchsenschutz (Paris), N. Cantin (Bordeaux), L. Chaix (Genève), Ph. Colomban (Paris), B. Cunliffe (Oxford, GB), M. Dabas (Paris), F. Delamare (Valbonne), Ph. Dillmann (Paris), C. Domergue (Toulouse), P. Fluzin (Belfort), P. Gassmann (Neuchâtel, CH), B. Gratuze (Orléans), M.-F. Guerra (Paris), P. Guibert (Bordeaux), N. Huet (Nantes), C. Leroyer (Rennes), I. Liritzis (Rhodes, GR), M. Magetti (Fribourg, CH), Ph. Marinval (Toulouse), M. Martinon-Torres (Londres, GB), P. Meyers (Los Angeles, USA), C. Oberlin (Lyon), M. Pernot (Bordeaux), M. Picon (Lyon), G. Querre (Rennes), T. Rehren (Londres, GB), I. Reiche (Paris), H. Richard (Besançon), L. Robiola (Paris), A. Schmitt (Lyon), A. Tabbagh (Paris), M. Vendrell-Saz (Barcelone, ES), J.-D. Vigne (Paris), Ph. Walter (Paris). Cover / Couverture Top left / Haut à gauche : The inverted bow-shaped gold fibula from the Phoenician tomb of Kition, Cyprus. Reference MLA1742/20, National Museum of Archaeology of Nicosia / La fibule en or à l’arc inverse de la tombe phénicienne de Kition, Chypre. Référence MLA1742/20, Musée national d’Archéologie de Nicosie. Top right / Haut à droite : X-radiography of a 19th century fibula with a lion produced in the Etruscan style. Reference 85037, National Museum of Archaeology of Florence / Radiographie X de la fibule au lion produite au XIXe siècle dans le style étrusque. Référence 85037, Musée national d’Archéologie de Florence. Bottom left / Bas à gauche : Myanmar jeweller producing gold beads, which stylistically replicate the ancient Asiatic beads / Orfèvre à Myanmar fabricant des perles en or stylistiquement identiques aux anciennes perles asiatiques. Bottom right / Bas à droite : PIXE analysis at the AGLAE accelerator of the C2RMF of a Kushan gold pendant. Reference 274, National Museum of Antiquities of Tajikistan / Analyse PIXE à l’accélérateur AGLAE du C2RMF d’un pendentif Kushan en or. Référence 274, Musée national des Antiquités du Tadjikistan. Sauf opposition formelle des auteurs ou de leurs ayant-droit, les articles parus dans ArcheoSciences pourront être mis en ligne par l’éditeur pour être consultés et imprimés, à titre gratuit ou onéreux. www.pur-editions.fr © Presses Universitaires de Rennes et ArcheoSciences Campus La Harpe 2, rue du Doyen-Denis-Leroy 35044 RENNES cedex Dépôt légal : mars 2009 ISSN 1960-1360 ISBN 978-2-7535-1181-1 This volume is dedicated to the memory of our dear colleague and friend Maurizio Donati, who attended the Workshop and unexpectedly passed away on the 20th of May 2010 when he was working on his contribution to these Proceedings. He was a respected Master goldsmith and a dedicated Emeritus Professor at the Istituto Statale d’Arte in Rome. Besides the quality of his work, we were all in many ways touched by the quality of the Man. He will be missed. Ce volume est dédié à la mémoire de Maurizio Donati, notre cher collègue et ami qui a participé au Workshop et qui est subitement décédé le 20 mai 2010 pendant la rédaction de sa contribution à ces actes. Il était un Maître orfèvre respecté et un Professeur Émérite dévoué de l’Istituto Statale d’Arte de Rome. Au-delà de la qualité de son travail, nous avons toujours pu apprécier ses qualités humaines. Nous le regretterons beaucoup. Publié avec le soutien de AS, 33, 2009 Sommaire / Contents GUERRA M. F., REHREN Th., AURUM: Archaeometry and authenticity of gold AURUM : Archéométrie et authenticité de l’or 13 MATERIALS AND METHODS: THE FOUNDATIONS MATÉRIAUX ET MÉTHODES : LES BASES PEREA A., Wands in the hand! Or Potter’s powers Une baguette magique dans la main! Ou les pouvoirs de Potter 21 DEMORTIER G., Targeting Ion Beam Analysis techniques for gold artefacts Techniques de faisceaux d’ions ciblés pour l’analyse d’objets en or 29 RADTKE M., REINHOLZ U., RIESEMEIER H., Synchrotron Radiation Induced X-Ray Fluorescence for the characterization of ancient Gold objects Fluorescence de rayons X induite par radiation de synchrotron pour la caractérisation d’objets anciens en or YANAKIEVA D., TONKOVA M., SPIRIDONOV E., VERGILOV Z., PENKOVA P., X-ray Diffraction method for determination of crystallite sizes of gold and silver items – New opportunities for archaeology and for protection against forgery Méthode de diffraction de rayons X pour la détermination des tailles de cristallites des objets d’or et d’argent – Nouvelles opportunités pour l’archéologie et pour la protection contre la production de faux FERRO D., VIRGILI V., CARRARO A., FORMIGLI E., COSTANTINI L., A multi-analytical approach for the identification of technological processes in ancient jewellery Approche multi-analytique pour l’identification des procédés technologiques de fabrication de bijoux anciens EUGSTER O., PERNICKA E., BRAUNS M., SHUKOLYUKOV A., OLIVE V., ROELLIN S., Helium, uranium and thorium analyses of ancient and modern gold objects: estimates of their time of manufacturing Analyse de l’hélium, de l’uranium et du thorium dans des objets anciens et modernes en or : estimation de leur date de fabrication 39 45 51 59 SPIRIDONOV E., YANAKIEVA D., Modern mineralogy of gold: overview and new data Minéralogie moderne de l’or : bilan et nouvelles données 67 HAUPTMANN A., KLEIN S., Bronze Age gold in southern Georgia L’or en Géorgie du sud à l’Âge du Bronze 75 TĂMAŞ C. G., BARON S., CAUUET B., Minéralogie et signature isotopique du plomb des minerais auro-argentifères exploités durant l’époque romaine à Alburnus Maior (Roşia Montană, Roumanie) Mineralogy and lead isotope signature of the gold-silver ores exploited during the Roman period at Alburnus Maior (Roşia Montană, Romania) 83 BERGONZI G., The earliest gold objects in Italy: a review of the archaeological evidence Les plus anciens objets d’or découverts en Italie : bilan des donnés archéologiques 91 BENNETT A. T. N., Gold in early Southeast Asia L’or dans le Sud-Est asiatique ancien 99 STUDIES OF OBJECTS: MANUFACTURING SKILLS AND ALLOY SELECTION ÉTUDES D’OBJETS : COMPÉTENCES TECHNIQUES ET CHOIX DES ALLIAGES TROALEN L. G., GUERRA M. F., TATE J., MANLEY B., Technological study of gold jewellery pieces dating from the Middle Kingdom to the New Kingdom in Egypt Étude technologique de pièces d’orfèvrerie datées du Moyen Empire au Nouvel Empire en Égypte 111 TATE J, EREMIN K., TROALEN L. G., GUERRA M. F., GORING E., MANLEY B., The 17th Dynasty gold necklace from Qurneh, Egypt Le collier en or de Qurneh daté de la XVIIe Dynastie, Égypte 121 PRÉVALET R., Preliminary observation on three Late Bronze Age gold items from Ras Shamra-Ugarit (Syria) Observations préliminaires de trois objets en or de l’Âge du Bronze récent de Ras Shamra-Ougarit (Syrie) 129 ADRIMI-SISMANI V., GUERRA M. F., WALTER P., La tombe mycénienne de Kazanaki (Volos) et le mythe de la Toison d’or The Mycenaean tomb of Kazanaki (Volos) and the myth of the Golden Fleece 135 FLOURENTZOS P., VITOBELLO M. L., The Phoenician gold jewellery from Kition, Cyprus L’orfèvrerie phénicienne en or de Kition, Chypre 143 GUERRA M. F., REHREN Th., In-situ examination and analysis of the gold jewellery from the Phoenician tomb of Kition (Cyprus) Examen et analyse in situ d’orfèvrerie en or de la tombe phénicienne de Kition (Chypre) PEREA A., HUNT-ORTIZ M. A., New finds from an old treasure: the archaeometric study of new gold objects from the Phoenician sanctuary of El Carambolo (Camas, Seville, Spain) Nouvelles trouvailles pour un ancien trésor : l’étude archéométrique de nouveaux objets en or du sanctuaire phénicien de El Carambolo (Camas, Séville, Espagne) 151 159 FERRO D., BEDINI A., RAPINESI I. A., Two small orientalising spirals (Rome, 10th-9th century BC): common objects – precious jewels Deux petites nattes en spirale orientalisantes (Rome, Xe-IXe siècle avant J.-C.) : objects du quotidien – bijoux précieux 165 MELCHER M., SCHREINER M., BÜHLER B., PÜLZ A. M., MUSS U., Investigation of ancient gold objects from Artemision at Ephesus using portable μ-XRF Enquête sur les objets anciens en or de l’Artemision à Ephèse au moyen d’un équipement de μFX portable 169 GUERRA M. F., DEMORTIER G., VITOBELLO M. L., BOBOMULLOEV S., BAGAULT D., BOREL T., MIRSAIDOV I., Analytical study of the manufacturing techniques of Kushan gold jewellery (National Museum of Antiquities of Tajikistan) Étude analytique des techniques d’orfèvrerie Kouchane (musée national des Antiquités du Tadjikistan) 177 ARMBRUSTER B., Gold technology of the ancient Scythians – gold from the kurgan Arzhan 2, Tuva La technologie de l’or chez les Scythes anciens – l’or du kurgan Arzhan 2, Touva 187 ILIEVA P., PENKOVA P., Funeral golden mask and hand with a ring. The necropolis of Trebeniste Le masque funéraire en or et la main baguée – la nécropole de Trebeniste 195 TONKOVA M., PENKOVA P., Les parures en or de la nécropole thrace de Duvanli du ve siècle av. J.-C. : le cas du tumulus de Kukova The gold jewellery from the Thracian necropolis of Duvanli (5th century BC): the case of the Kukova mound SHEMAKHANSKAYA M., TREISTER M., YABLONSKY L., The technique of gold inlaid decoration on the 5th-4th centuries BC: silver and iron finds from the early Sarmatian barrows of Filippovka, Southern Urals La technique de décoration par inclusion d’or aux Ve-IVe siècles av. J-C : les trouvailles en argent et fer des tumulus de la période Sarmate ancienne de Filippovka au sud de l’Oural 201 211 CONSTANTINESCU B., BUGOI R., COJOCARU V., RADTKE M., CALLIGARO T., SALOMON J., PICHON L., RÖHRS S., CECCATO D., OBERLÄNDER-TÂRNOVEANU E., IONESCU C., POP D., Dacian bracelets and Transylvanian gold: ancient history and modern analyses Bracelets de Dacie et or de Transylvanie : histoire ancienne et analyses modernes OANŢĂ-MARGHITU R., NICULESCU G., ŞECLĂMAN D., BUGOI R., GEORGESCU M., The gold belt buckle from Apahida III (Romania), 5th century AD La boucle de ceinturon en or d’Apahida III (Roumanie), Ve siècle apr. J.-C. WHITFIELD N., ‘More like the work of fairies than of human beings’: the filigree on the ‘Tara’ brooch, a masterpiece of late Celtic metalwork « Œuvre des fées plus qu’œuvre des hommes » : le filigrane de la fibule de « Tara », un chef d’œuvre de l’orfèvrerie celtique tardive AUFDERHAAR I., From the goldsmith’s point of view: gilding on metals during the first millennium AD – techniques and their development in the Germanic area Du point de vue de l’orfèvre : dorure sur métal au premier millénaire apr. J.-C. – les techniques et leur évolution dans l’aire germanique SAPRYKINA I., TETERIN A., MITOYAN R., Gold foil covering of the handle of an iron knife from burial 2 of the Hunnic Period cemetery at Mukhino, in the Upper Don area L’ornement en or du manche d’un poignard en fer de la tombe 2 de la période des Huns de la nécropole de Moukhino, dans le Haut Don DONATI M., The wire ‘at astragals’, or beaded wire, from medieval tradition to the technique and tools used by the Roman goldsmiths Castellani in the 19th century Le fil « à astragales », ou fil perlé, de la tradition médiévale à la technique et aux outils utilisés par les orfèvres romains Castellani au XIXe siècle OLIVEIRA M. J., MARANHAS T., SERUYA A. I., MAGRO F. A., BOREL T., GUERRA M. F., The jewellery from the casket of Maria Pia of Savoy, Queen of Portugal, produced at Castellani’s workshop Les bijoux du coffret de Maria Pia de Savoie, Reine du Portugal, fabriqués à l’atelier Castellani 221 227 235 243 255 259 265 SOUTH AMERICA: GOLD STUDIES IN THE NEW WORLD AMÉRIQUE DU SUD : ÉTUDE DES ORS DU NOUVEAU MONDE BOUCHARD J.-F., GUERRA M. F., Archéologie précolombienne et analyses scientifiques : la figurine d’El Angel, une œuvre composite d’orfèvrerie de la culture La Tolita Tumaco (Équateur-Colombie) Pre-Columbian archaeology and science-based analysis: the figurine of El Angel, a composite goldwork from La Tolita Tumaco culture (Ecuador – Colombia) CESAREO R., BUSTAMANTE D. A., FABIAN S. J., CALZA C., DOS ANJOS M., LOPES R. T., ALVA W., CHERO Z. L., GUTIERREZ V. F., ESPINOZA C. M., RODRIGUEZ R. R., SECLEN F. M., CURAY V., ELERA C., SHIMADA I., Pre-Colombian alloys from the royal tombs of Sipán and from the Museum of Sicán: non-destructive XRF analysis with a portable equipment Alliages précolombiens des tombes royales de Sipán et du musée de Sicán : analyse non-destructive avec un système FX portable 273 281 RUVALCABA SIL J. L., PEÑUELAS GUERRERO G., CONTRERAS VARGAS J., ORTÍZ DÍAZ E., HERNÁNDEZ VÁZQUEZ E., Technological and material features of the gold work of Mesoamerica Caractéristiques technologiques et matérielles du travail de l’or en Mésoamérique 289 PEÑUELAS-GUERRERO G., CONTRERAS-VARGAS J., RUVALCABA-SIL J. L. GARCÍA-ABAJO A., Technological study of pre-Columbian bimetallic discs from Monte Alban, Oaxaca, Mexico Étude technologique des disques bimétalliques précolombiens de Monte Alban, Oaxaca, Mexique 299 ORTIZ DÍAZ E., Location of gold placers in Oaxaca during the late pre-Hispanic period and early Colonial times: 1250 to 1550 AD Localisation de placers aurifères à Oaxaca pendant la période préhispanique tardive et la période coloniale récente : 1250 à 1550 apr. J.-C. 303 DUTTINE M., GUERRA M. F., LOBO VIEIRA R. M., SCORZELLI R. B., PEREIRA C. E. B., PEREZ C. A., The first gold coins struck in Brazil: myth or reality? Les premières monnaies en or frappées au Brésil : mythe ou réalité ? 309 PARREIRA P. S., APPOLONI C. R., LOBO VIEIRA R. M., SCORZELLI R. B., LE CORRE L., GUERRA M. F., Precious metals determination in ancient coins by portable ED-XRF Spectroscopy with a 238Pu source Détermination des métaux précieux dans les monnaies anciennes par spectroscopie ED-FX portable avec une source de 238Pu 313 Eva BRITO E., CHAIRE F., Gold jewellery of devotional images in Campeche State, Mexico Bijoux en or des images de dévotion dans l’état de Campeche, Mexique 319 AUTHENTICATION: APPLYING EXPERT KNOWLEDGE AUTHENTIFICATION : APPLICATION DES CONNAISSANCES DU SPÉCIALISTE LA NIECE S., Forgeries and public collections Les contrefaçons et les collections publiques 329 FORMIGLI E., Micro rayures et signes d’usure : authentification d’orfèvrerie archéologique The analysis of micro-scratches and trace-wear in the authentication of archaeological gold 335 BOBIN O., GUEGAN H., A new approach to the authentication of goldwork using combined Scanning Electron Microscopy and External-beam PIXE Une nouvelle approche pour l’authentification des objets en or en couplant la microscopie électronique à balayage et le PIXE en faisceau extrait 341 VELLA D., LICARI J., VELLA N., SULTANA S., CIANTAR V., Copy or authentic? Analysis of a Phoenician gold ring from the National Museum of Archaeology, Valletta, Malta Copie ou authentique ? Analyse d’une bague phénicienne en or du Musée national d’archéologie, La Vallette, Malte 349 RASTRELLI A., MICCIO M., TROALEN L., MARTINÓN-TORRES M., GUERRA M. F., SIANO S., SUMBERA A., VITOBELLO M. L., Modern and ancient gold jewellery attributed to the Etruscans: a science-based study Orfèvrerie moderne et ancienne attribuée aux Étrusques: une étude scientifique 357 KULEFF I., STOYANOV T., TONKOVA M., Gold Thracian appliqués: authentic or fake? Appliqués thraces en or: authentiques ou faux? ENIOSOVA N., A unique 10th century AD gold-plated brooch from south-east Russia : technical and stylistic authentication Une broche unique plaquée or du Xe siècle apr. J.-C. provenant du Sud-Est de la Russie : authentification technique et stylistique 365 375 RESTORATION AND CONSERVATION RESTAURATION ET CONSERVATION BERGER O., Observations and questions on gold artefacts from underwater excavations The view of an archaeologist-restorer working in situ Observations et interrogations sur des objets en or provenant de fouilles archéologiques sous-marines Le point de vue d’un archéologue-restaurateur travaillant in situ TISSOT I., TISSOT M., PEDROSO P., RAPOSO L., Treasures of Portuguese archaeology. Notes towards a preventive conservation project Trésors de l’archéologie portugaise – notes pour un projet de conservation préventive 383 389 PACINI A., Ancient gold patinas: experimental reconstruction Les patines anciennes de l’or : recréation expérimentale 393 MOUNIER A., DANIEL F., BECHTEL F., L’illusion de l’or. Imitation de dorures dans les peintures murales médiévales The illusion of gold. Imitation of gilding in mediaeval mural paintings 397 BUCCOLIERI G., BUCCOLIERI A., BRACCI S., CARNEVALE F., FALLETTI F., PALAMÀ G., CESAREO R., CASTELLANO A., Gold leafs in 14th century Florentine painting Feuilles d’or dans la peinture florentine du XIVe siècle 405 CAVALLO G., VERDA M., Gilding on wall paintings from a period between the 14th-16th centuries AD in the Lombard lake region La dorure sur peintures murales entre le XIVe et le XVIe siècles dans la région des lacs lombards 409 BIDARRA A., COROADO J., ROCHA F., Gold leaf analysis of three baroque altarpieces from Porto Analyse de feuilles d’or de trois retables baroques de la ville de Porto 417 LE GAC A., SERUYA A. I., LEFFTZ M., ALARCÃO A., The main altarpiece of the Old Cathedral of Coimbra (Portugal): characterization of gold alloys used for gilding from 1500 to 1900 Le retable majeur de l’ancienne cathédrale de Coimbra (Portugal) : caractérisation des alliages d’or employés dans la dorure de 1500 à 1900 423 AURUM: Archaeometry and authenticity of gold AURUM : Archéométrie et Authenticité de l’or Maria Filomena Guerra* and Thilo Rehren** Abstract: In this introduction to volume 33 of ArcheoSciences, we provide a brief overview of the use and abuse of gold over time, and its different aspects, from the mine to the objects, their use, analysis, and restoration. For this purpose, we focus on the papers presented in this volume, which originate from the Workshop AURUM: authentication and analysis of goldwork, organised under the auspices of the EU-DG Research funded project AUTHENTICO. The main aim of this project was to develop tools and expertise for law enforcement agencies to combat illicit trade in antiquities and to fight fraud and forgeries; to do so requires an understanding of the diversity of the cultural, technical and material manifestations of gold artefacts, and their very specific combinations and expressions. Some of the scholarly foundations of this endeavour are illustrated by the selection of the 55 papers, arranged in five topical sections, which are introduced in this text. Résumé : Dans cette introduction au volume 33 d’ArcheoSciences, nous faisons un bref bilan du us et abus de l’or au long du temps, des divers aspects qu’il peut prendre de la mine à l’objet, et de l’utilisation des objets d’or, de leur analyse et de leur restauration. Ce bilan repose sur les articles présentés dans ce volume, qui émanent du Workshop AURUM : authentification et analyse d’objets en or, organisé sous les auspices du projet AUTHENTICO, financé par le DG-Recherche de la Commission Européenne. Le but principal de ce projet a été le développement d’outils et de compétences pour combattre le trafic illicite d’antiquités et pour lutter contre la fraude et le commerce de faux. Pour atteindre cette fin, il est nécessaire de comprendre la diversité des manifestations culturelles, techniques et matérielles des objets en or et leurs combinaisons et expressions très spécifiques. Quelques uns de ces fondements sont illustrés par les 55 articles, organisés en cinq sections thématiques, qui sont introduits par ce texte. Keywords: AURUM, introduction, gold, analysis, forgeries. Mots-clés : AURUM, introduction, or, analyse, faux. The appeal of gold is near-universal, and throughout history almost all metal-using cultures have put gold at the top of their list of desirables. Even today, with other metals being more expensive and more versatile to use, gold holds a particular fascination for most of us. The unique aesthetic qualities of gold and its alloys – their subtle range of colours – and its smooth untarnished surface quality even after years and, as we see with the many objects studied in this volume of ArcheoSciences, millennia of use – like a representation of eternal youth – are certainly the reasons of that attraction. The workshop from which the papers presented in this volume of ArcheoSciences originate was entitled AURUM: authentication and analysis of goldwork. It was organised under the auspices of the EU-DG Research funded project AUTHENTICO (Authentication methodologies for metal artefacts based on material composition and manufacturing techniques) n. 044480 under the 6th Framework Programme. The main aim of this project, including ten partners from eight different countries and co-ordinated by Maria Luisa Vitobello, director of the European Jewellery * Laboratoire du Centre de Recherche et de Restauration des Musées de France, UMR171 CNRS, 14, quai François-Mitterrand, 75001 Paris, France. (maria.guerra@culture.gouv.fr) ** UCL Institute of Archaeology, 31-34 Gordon Square, London WC1H 0PY, United Kingdom. (th.rehren@ucl.ac.uk) rec. Oct. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 13-18 14 Technology Network, was to develop tools and expertise for law enforcement agencies to combat illicit trade in antiquities and to fight fraud and forgeries. Access to the results of AUTHENTICO is necessarily restricted; AURUM acts as the public face of AUTHENTICO, within the academic ethos of open discussion, free flow of information and sharing knowledge. During the three days of the workshop in May 2009, more than 115 participants from over 28 countries (Austria, Belgium, Brazil, Bulgaria, Canada, Colombia, Cyprus, Czech Republic, Egypt, France, Greece, Germany, Hungary, Iran, Ireland, Italy, Malta, Mexico, The Nedherlands, Portugal, Romania, Russia, Spain, Switzerland, Tajikistan, Tunisia, United Kingdom, and the United States of America) participated in wide-ranging discussions, spanning the entire spectrum of aspects indicated above (Fig. 1). Many more wanted to come; but the room available simply did not allow more people to attend. This overwhelming interest, not only from scholars but also the wider public, represented by government officials, a member of the European Parliament, representatives of two European Law Enforcement Agencies – the Command Office Carabinieri, Department for the Protection of Cultural Heritage in Rome, and the Central Office for the Fight against Traffic in Cultural Goods of the French Criminal Investigation Department – and the press, encouraged us to prepare the presentations for swift publication. This collection of papers offers but a glimpse of some of the main themes played out by gold, both in the past, but also in today’s world. Even though wonderful gold objects can be easily produced by simple casting and hammering, such as the Scythian ornaments studied by Armbruster and the finds from the royal tombs of Sipán analysed by Cesareo et al., the most delicate objects have been made from gold, by the most skilful artisans of their times, multiplying the worth of the metal with the value of their craft. The elbow fibula from the Phoenician tomb of Kition in Cyprus, studied by Vitobello and Flourentzos and Guerra and Rehren, the Celtic ‘Tara’ brooch described by Whitfield, the belt buckle found in a princely grave from Apahida in Romania dissected by Oanta-Marghitu, the adult’s necklace of the Qurneh burial near Luxor studied by Tate et al., and the bimetallic discs from Monte Alban, Oaxaca presented by Peñuelas-Guerrero et al. are just a few examples of the artisans’ skills at different periods treated in this volume. The functions of the objects fabricated with gold were manifold: they include everyday jewellery from antiquity (such as in Ras Shamra, as reported by Prévalet, and from the Phoenicians, see Ferro et al.) to modern times (as in the 19th century, described by Donati), funerary objects (studied by Ilieva and Penkova and Tonkova and Penkova), offerings ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? Maria Filomena GUERRA, Thilo REHREN and worship items (illustrated in the paper by Brito and Chaire), coins (analysed by Duttine et al. and Parreira et al.), insignia of power, and even intermediate shapes such as ingots for transporting the precious metal and for use in trade. By using different gold alloys or/and by adding gems, glass, enamel, and other more or less precious materials to gold objects, polychromy was achieved by the artisans. The Egyptians (as shown by Troalen et al.), the Minoans and the Mycenaeans (as illustrated in the paper by Adrimi et al.) used that technique already in the Bronze Age. Other techniques, such as gold inlay (described by Shemakhanskaya et al.), gilding (an overview of this technique in Germany is given by Aufderhaar), plating (examples are given by Saprykina et al. and by Eniosova), and patination (Pacini shows how to produce ancient patinas) can be used to make entirely or partially golden objects. The production of those objects can change over time, responding to religious, political, or other needs, or yet to the taste and style of each period, as exemplified in the ‘archaeological style’ of objects produced in the 19th century by renowned goldsmiths such as Castellani; the study of the jewellery from a royal casket produced in Castellani’s workshop is presented by Oliveira et al. But gold has other resonances, too. Gold is power. Whoever controls the gold sources controls the economy. Many myths, like those of Jason and the Golden Fleece, of the El Dorado, and of King Midas, are linked to the quest for gold and its recovery from alluvial deposits, the first to be exploited. However, such mythic sources of gold are as difficult to locate in the present as they were in the past (see Hauptmann et al. and Adrimi et al. for the myth of Jason and the Golden Fleece), which is true also for other sources used more recently (as exposed by Ortiz Díaz for Mesoamerica, by Duttine et al. for South America and by Constantinescu et al. for Eastern Europe). Throughout history, battles and wars, migrations and shifts in population were triggered by the search for gold sources or the desire to control them. The gold rushes in, for example, Alaska, Klondike, California, Witwatersrand, and Serra Pelada are recent examples of this. In their time, the Romans knew it when they went to Spain; the Arab rulers applied it in their quest for North Africa in the early Islamic period, the Spanish and the Portuguese when they went to South America, and the British Empire in its more recent occupation of South Africa. Gold has been at the heart of money for millennia. Even after putting an end to the traditional gold standard in their meeting in Bretton Woods in 1944 and the subsequent unilateral decision by the US Government to break the fixed link between the dollar and gold metal in 1971, governments around the world still stockpile vast quantities of the metal and exert political and economic pressure with their tons of gold. Archaeometry and authenticity of gold Of course, in cultural heritage contexts the material value of gold is further exacerbated by the exclusive appeal of pieces of art and the very limited and non-renewable nature of the resource. This rarity in itself drives desire. Archaeological gold objects combine in them all these multiple connotations, further enhancing each other in complex ways. As a result, archaeological gold objects attract particular attention; some of it good, some less so. As early as 1237, King Henry III of England issued a Royal Charter which required Bronze Age tumuli to be dug up for treasure, not for their cultural value but to boost the royal coffers. Similarly, the European desire for untold riches represented the driving force behind much of the American conquest, with devastating consequences for the gold objects produced by several cultures, such as the Incas and the La TolitaTumaco (Bouchard and Guerra present the analytical study of a gold figure from the latter civilisation). If hoarding is the starting point of collecting, more enlightened approaches to antiquities in the Renaissance and in the age of Enlightenment led to the formation of important collections, many of which are now in public ownership and freely accessible for all to enjoy (some including some non-genuine objects, as illustrated by La Niece for The British Museum collections). However, the urge to own gold, archaeological gold in particular, is undiminished, and many private collectors compete for a very limited pool of genuine and legitimately available objects. There are many reasons for this: ranging from the psychological, almost primordial desire for gold; to the rise of an affluent middle class and nouveaux riches in many countries with their need to showcase their wealth and ‘cultured’ manners; and genuine interest in aesthetic beauty, art history and cultural heritage. Taken together, they result in a market demand that by far outstrips supply. The consequences are many: prices rise, even in times of global economic crisis; tomb robbery, illicit excavation and museum theft feed the market at the cost of wholesale destruction of our common cultural heritage; and forgers produce tirelessly new objects to satisfy the ballooning demand with copies, reproductions, imitations, and heavy restorations re-interpreting an object, by producing pastiches and more or less fantastic fraudulent inventions (Rastrelli et al.). One might argue that forging archaeological objects is a lesser evil, as compared to the looting of entire sites. However, the unchecked demand leads to an entire industry producing forgeries, thus causing cultural damage, by diluting, debasing and distorting the genuine archaeological record, the artistic and technological achievements, the aesthetic, religious and personal expressions of countless past people. 15 The absence of discussions on illicit traffic and trade in forgeries from most conferences in the field of archaeological sciences led us to dedicate an entire day of AURUM to a special session entitled Forgeries, Trade and Authentication. After three invited talks by Pieter Meyers (Los Angeles County Museum of Art Conservation Center), Susan La Niece (Department of Conservation and Scientific Research, The British Museum) and Colonel Alberto Deregibus (Head of Command Office Carabinieri, Department for the Protection of Cultural Heritage, Rome), a roundtable discussion was dedicated to these questions (see Fig. 2). We won’t be able to stop these illegal activities from happening; but it is important to highlight their corrosive effect on so many different aspects of society. We mentioned the destruction and debasement of our common cultural heritage already, but there are also other major issues for current societies; from loss of tax receipts for governments to the countless individual acts of fraud committed in the process, to the damage done to the potential for the development of an archaeological site for sustainable tourism and education, robbing the local population of their long-term livelihood. This volume of ArcheoSciences is devoted to the scholarly study of archaeological gold objects, not from the art historical or aesthetic point of view typical of traditional collections and catalogues, but with the aim of unravelling the technical skills of the craftspeople who made these objects. Characterising their tools and materials as a way to understand their actions, their gestures allow us to place them within their unique geological, economic and cultural contexts and traditions (Perea illustrates in her paper the big advances reached by coupling different fields of research in the study of gold work, and gives an example in the paper by Perea and Hunt). Studying the tangible finds thus reveals the intangible wealth of the cultural heritage of craftspeople and artisans. This, in turn, will inform the study of those objects which have no proven pedigree, no documented historical or archaeological origin and whose authenticity is therefore in many cases open to dispute. Exploring authenticity combines art history and connoisseurship, an array of methods of scientific examination under many lights and radiations. It involves scientific analysis with portable instruments, as for example in the study of the Artemision objects by Melcher et al., and fixed equipment based on X-rays, mass spectrometry, synchrotrons (as developed by Radtke et al.), particle accelerators (as illustrated by Bobin and Guégan and by Demortier), and so on, and can be combined with detailed visual and microscopic examination techniques (examples of studies combining techniques are provided by Ferro et al. and Guerra et al.). In all cases, the skill of the detective is needed in the search for the wrong aesthetic, ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? 16 technical and material elements, which should not be there: it is never routine. Following a rigorous selection and peer review process, we have arranged the accepted contributions in five overarching sections. The first, containing eleven papers, serves as an introduction and offers examples of the material and methodological foundations relevant for the study of archaeological gold. This covers the archaeological record of early gold objects as much as the geology of gold, and ancient gold mining. Examples here include the papers by Bergonzi and by Bennett on the occurrence of gold artefacts in Italian and South-East Asian archaeology; by Spiridonov & Yanakieva on the mineralogy of gold, by Hauptmann et al. on early gold mining in Georgia, and by Tamas et al. on the geology of the gold exploited in Romania under Roman control. Other fundamental papers concern modern analytical approaches to characterise, provenance, and even date gold; papers by Eugster et al., Radtke et al. and Demortier are amongst those addressing these issues. The second section contains the main ‘meat’ of the conference; twenty-one papers present individual case studies concerning the material characterisation and manufacturing skills of the ancient artisans, from Late Bronze Age Egypt right through to the 19th century AD. The flow of these papers follows broadly chronological and cultural trends, from the origins in the Near East to the classical Mediterranean cultures and on to the early medieval and later European centres of developments in metal crafts. An important and impressive observation that emerges from reading these papers is the continuing manifestation of manufacturing skills at the highest levels, seemingly uninterrupted by the broad political and economic convulsions that tormented the Old World during those four thousand years. Individual cultures ebbed and flowed, flourished and perished; but overall humankind persevered, developing and refining its technical achievements in one way or another. Remarkable also within this broad advance is the exceptional, and often unique and unsurpassed, achievement of individual artists, stimulated by their own imagination, driven by their devotion, and maintained by their communities, religious orders, or wealthy patrons. The previous section has shown how coherent and at the same time divers cultural and technological developments were across time and space in the Old World. The New World, in contrast, approached gold in its own and independent manner; the paper by Ruvalcaba Sil et al. illustrates this by providing an overview of the artisans’ gold work of Mesoamerica. This third section combines eight papers on gold work from this micro-cosmos of its own, with its unique gold deposits which are often geochemiArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? Maria Filomena GUERRA, Thilo REHREN cally quite distinct from gold sources in the Old World, but also blessed with a plethora of unique archaeological styles, and burdened by unique archaeological problems of looting, forging and wholesale plunder, intricately linked to the political and economic realities of the last five hundred years. This section thus covers the entire sequence of gold work and its investigation, from studies of gold deposits to analytical methods and the investigation of iconic objects, and finally to the origin and use of gold for coinage and Christian decorations following the European dominance. To some extent, all papers up to this point can be seen as a prelude to the fourth section, on authentication. Here, the accumulated expert knowledge is applied to address specific case studies, giving examples of the more fundamental concerns and showcasing the results of established and new methods. Of course, these papers only represent the tip of an iceberg, and here is not the place to repeat the discussion from the round table, and the many reasons why this is the case. Suffice it to say that if anything emerges from these papers, it is the recognition that there is no ‘golden bullet’ that delivers the final answer to all our problems; in some cases it is a specific chemical tracer (Bobin & Guégan), in others a particular tool mark or wear mark (Formigli and Vella et al.), and in any case the scientific and technological investigation (Kuleff et al.) has to be fully integrated into an art historical and cultural study. The fifth section finally combines papers on the restoration and conservation of gold artefacts, sometimes in known public collections (such as in the Portuguese National Archaeological Museum, Tissot et al.), sometimes discovered in particular environments, such as underwater excavations (in this volume, Berger describes those held in Alexandria and Aboukir Bay, Egypt). Given the inherent durability of gold itself, it is not surprising that the majority of these papers concern gold in its most delicate state: as leaf and foil covering other materials, such as wood, paint or gesso (see Mounier et al., Buccolieri et al. and Cavallo & Verda). Some of these papers raise further interesting questions about the concept of authenticity and conservation of ‘living objects’. It may be easy to decide how to best preserve a museum piece, kept under controlled conditions and frozen in time in the condition in which it was last found. But what about churches and their centrepieces, the altars (discussed by Le Gac et al. and Bidarra et al.)? These have important functions to fulfil for their current communities and need to be fit for purpose, but they are also pieces of art and part of our heritage and need to be preserved in the way they were first made to look like. Which authenticity does one preserve here: the authenticity of the original maker of the altar, or the authenticity of the current congregation using Archaeometry and authenticity of gold it, adapting it to their needs and changing artistic tastes and preferences? There are important and divers questions to be asked in the study of archaeological gold objects. Should we simply adore and cherish their absolute beauty removed from any distraction, or should we see them embedded in a past culture, as expressions of beliefs and priorities? Should we see primarily their social significance, read their materiality as mere projections of something bigger, intangible, or should we deconstruct them, if not literally then metaphorically, to understand them as products of mundane acts of crafts and industry? Whose authenticity do we prioritise? The artist’s, who made the object? The patron’s, who facilitated its production? The current owner’s, who holds it dear? What does the present state of the object tell us about its live history as an artefact, not just its creation but its use and repair, its adaptation and modification over time, and what does its present condition and whereabouts tell us about the changing fortunes and functions of the object? This collection of papers can not give answers, but it can still tell us a lot. Factual information, of course. Case studies inspiring one’s own work, enabling comparisons to be drawn. The sheer range and diversity of approaches, the depth of individual scholarly knowledge that underpins these papers is impressive and heartening to see. On the other hand, it tells us also where the gaps are, where we still know far too little, and where we may have lost our heritage already. It is the nature of Archaeology that only a minute amount of material survives the ravages of time, even in the best of circumstances. Gold is no exception, and we may ever only see a tiny fraction of the true wealth and richness of our cultural heritage. What gold benefits from its durability once buried in the ground it suffers from its propensity for recycling before it gets there. Thus, it is our responsibility towards the peoples who made these objects, as well as towards our children and grand children, to preserve what we have, to enhance our understanding of what there is, to safeguard it against the greed of the individual and the ignorance of the uninformed. Analysing these objects scientifically and technologically reveals entirely new dimensions of knowledge about the artefacts themselves, their creators and the cultures from which they emanate; it is a profoundly constructive analysis, even if minute changes are made to the objects in the process. Acknowledgements We are very grateful for the many participants of the workshop AURUM who have together made it such a success, and who have submitted their manuscripts in time to be included in this volume. The support of many colleagues who acted as peer reviewers is much appreciated, as is the cooperation of the editors of ArcheoSciences in facilitating this volume. We are most 17 Figure 1: (See colour plate) The participants of Workshop AURUM: authentication and analysis of goldwork at the C2RMF laboratories (photograph by Elsa Lambert, C2RMF). Figure 1 : (Voir planche couleur) Les participants du Workshop AURUM : authentification et analyse d’objets en or dans les laboratoires du C2RMF (photographie par Elsa Lambert, C2RMF). grateful to Raul Carstocea for his excellent work in editing all the final manuscripts, ensuring where necessary the proper use of English, and who checked many references, and to Antoine Mattei who provided all the website development and support for the Workshop and the peer review process of this volume. The European Commission, DG Research under FP6 contract 044480 (AUTHENTICO), the Research and Restoration Centre of the French Museums (C2RMF), the Direction of the French Museums of the Ministry of Culture and Communication (DMF-MCC), the French National Centre for Scientific Research (CNRS) – Institute of Chemistry, GdR3174-ChimArc, UMR171-LC2RMF, UMR6115LPC2E and Paris A Delegation –, the Institute for ArchaeoMetallurgical Studies of the University College London, and Bruker AXS Microanalysis GmbH, are gratefully acknowledged for funding and website hosting. ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? 18 Figure 2: Workshop AURUM roundtable on Forgeries, Trade and Authentication. From left to right: Thilo Rehren (UCL – Institute of Archaeology), Maria Alicia Uribe (Gold Museum of the Banco de la República in Bogotá), Captain Dominique Lambert (Central Office for the Fight against Traffic in Cultural Goods of the French Criminal Investigation Department), Colonel Alberto Deregibus (Command Office Carabinieri, Department for the Protection of Cultural Heritage, Rome), Astrid BrandtGrau (EU-DG Research-I3, Environmental Technologies and Pollution Prevention, Protection, Conservation and Enhancement of Cultural Heritage), Barbara Armbruster (CNRS, Université de Toulouse – Le Mirail), Peter Northover (Department of Materials, Oxford University), Maria Luisa Vitobello (Coordinator of AUTHENTICO). Figure 2 : Table ronde du Workshop AURUM intitulé Faux, Commerce et Authentification. De gauche à droite : Thilo Rehren (UCL-Institut d’Archéologie), Maria Alicia Uribe (muse de l’Or du Banco de la República à Bogotá), Capitaine Dominique Lambert (Office Central de lutte contre le trafic des Biens Culturels, France), Colonel Alberto Deregibus (Office Carabinieri, Département de Protection du Patrimoine Culturel, Rome), Astrid Brandt-Grau (EU-DG Recherche-I3, Environnement Technologies et Prévention de Pollution, Protection, Conservation et Amélioration du Patrimoine Culturel), Barbara Armbruster (CNRS, Université de Toulouse- Le Mirail), Peter Northover (Département de Matériaux, Université d’Oxford), Maria Luisa Vitobello (Coordinateur d’AUTHENTICO). Maria Filomena GUERRA, Thilo REHREN Figure 3b: The workshop reception at C2RMF. Figure 3b : La réception du workshop au C2RMF Figure 3c and d: A lunch break at C2RMF. Figure 3c et d : Une pause déjeuner au C2RMF. Figure 3a: The workshop dinner at a French bistro. Figure 3a : Le diner du workshop dans un bistrot. ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? M  M:   Wands in the hand! Or Potter’s powers Une baguette magique dans la main ! Ou les pouvoirs de Potter Alicia Perea* Abstract: Some relections on the history of archaeometry addressing more speciically gold metallurgy will help us understand recent trends and achievements in this broad ield of research. Interdisciplinarity as a peer dialogue between scientists and historians has encountered some serious diiculties, which, however, are almost completely overcome today. he wide choice of non-destructive analytical techniques available is perhaps the main feature of actual research, and this also means that a great variety of problems can be dealt with and solved. Finally, in the present paper I propose some case studies selected from Project Au, the research programme I have been coordinating for more than ten years, and from other researchers and teams with whom I collaborate, in order to illustrate how we can conduct goldwork research. hese include the two principal, and unique, Visigothic treasures, from Guarrazar and Torredonjimeno (Spain), and the Braganza brooch, a controversial Iberian ibula, masterpiece of ancient jewellery now on display at the British Museum and presented in Madrid during an exhibition in 2007. Résumé : En ce qui concerne la métallurgie de l’or, quelques rélexions sur l’histoire de l’Archéométrie aideront à comprendre les orientations et les succès de ce large domaine de recherche. L’interdisciplinarité, en tant que dialogue scruté entre scientiiques et historiens, a enduré de sérieuses diicultés qui sont actuellement quasiment surmontées. Le vaste choix de techniques non-destructives d’analyse est peut-être la dominante principale de la recherche, ce qui signiie ainsi qu’une large variété de problèmes peut être traitée et résolue. Enin, de façon à illustrer comment les recherches sur l’or sont menées, j’exposerai quelques études soit prises dans le cadre du Project Au, programme de recherche que je coordonne depuis plus de dix ans, soit réalisées en collaboration avec d’autres chercheurs et groupes de recherche. J’invoquerai les deux principaux et uniques trésors Visigoths, Guarrazar et Torredonjimeno (Espagne), et la broche de Braganza, une ibule ibérique controversée, chef-d’œuvre de l’orfèvrerie antique, conservée actuellement au British Museum et présentée à Madrid lors de l’exposition de 2007. Keywords: Archaeometry, archaeometallurgy, goldwork, analytical techniques, Europe. Mots-clés : Archéométrie, archéométallurgie, orfèvrerie, techniques d’analyse, Europe. 1. INTRODUCTION A memorable exhibition directed by Mark Jones (1990) at the British Museum in 1990, entitled Fake? he Art of Deception, brought together the worst or best errors made in over two centuries of collecting precious art and archaeological objects, not only by the museum’s curators, but by other museums all over the world and private collectors as well. In the catalogue, Jones stated that experts saw what they wanted to see. his does not mean we should get rid of all experts, but of arrogance and most of all… of emotions. Expertise and validation were traditionally based on three aspects: perception-observation, knowledge-experience, and expectations-emotions. he connoisseur’s irst contact with the object involved an emotional response, positive or negative, based on his knowledge and experience, but conditioned by his or others’ expectations. * Arqueometal Research Group. Centro de Ciencias Humanas y Sociales. CSIC – Albasanz 26-28, 28037 Madrid – Spain. (alicia.perea@cchs.csic.es) rec. Aug 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 21-28 22 Since the advent of instrumental methods for quantitative elemental analysis and powerful observation devices in art and archaeology, the work of experts has relied much more on physical sciences and less on emotions and experience. Moreover, and more importantly, these are diicult times for forgers, or at least for non-expert. Concerning ancient gold, I have always thought that knowledge, not expectations, and the use of binoculars are suicient to reach a sensible opinion on most of the issues an object presents regarding its authenticity. If we want to proceed further and are interested in the skills of ancient craftsmen, in how they dealt with temperature, tools and workshops; in how they managed know-how, transmission and innovation; if we want to know where they obtained their raw materials, who paid for them and who paid for their work; to sum up, if we are interested in the society behind the object, rather than the object itself, then we must rely on scientiic methods of analysis. Today, the scientist is presented with a wide choice of non-destructive analytical techniques, depending on the questions he/she wants to answer. Two of them are powerful wands in our hands for gold technology research: electron beam microscope/microprobe (SEM), and ion beam analysis (IBA), which includes PIXE, PIGE and RBS methods (Ferro et al., 2003; 2008; Adriaens et al., 2005; Demortier and Adriaens, 2000). heir power consists of, in the irst case, the SEM capability for providing high resolution, high magniication images with spot elemental microanalysis; and, in the second, the IBA’s ability to detect light and heavy elements, even in large objects, at atmospheric pressure during irradiation. Reaching this point has not been an easy task. It represents the result of much efort from many people who sometimes failed, but never gave up. Let us have a look back at this story and throw a quick glance at the future as well. 2. IN THE BEGINNING … THERE IS FAILING Actual research into ancient goldwork is one of the specialisations within the ield of archaeometallurgy and its orientation is primarily archaeometric (Rehren and Pernicka, 2008), that is, based on scientiic methods of observation and analysis. his has not been achieved in a straightforward way, or without problems and tensions (Pollard and Bray, 2007). he irst major project aimed at the analytical and systematic study of gold production during European Prehistory was initiated more than half a century ago in the context of the programme Studien zu den Anfängen der Metallurgie undertaken by the ArcheoSciences, revue d’archéométrie, 33, 2009, p. 21-28 Alicia PEREA Württembergisches Landesmuseum Laboratories, Stuttgart. he results of sampling almost 5,000 gold artefacts from museums all over Europe were published by Hartmann in 1970 and 1982. However, the results did not live up to the expectations generated by the project, either in terms of technology or from the social and economic point of view. he disappointment was followed by a period in which the elemental chemical composition did not form part of the central approach adopted for research into gold (Perea and Armbruster, 2008). Much has been said about the reasons why this European and international project failed, but they boil down to two major ones: a) a statistical approach unsuitable for analytical data, and b) the absence of an archaeological theoretical basis that would provide a historical frame of reference for the vast quantity of numeric results thrown into disarray by the interpretive statistics themselves (Waterbolk and Butler, 1965; Taylor, 1980). In brief, the selected methodology could not answer the questions raised about the origin of metallurgy and metal exchange in Prehistoric Europe. However, the resulting data are still there, available to any researcher who may be interested and able to interpret them more skilfully (Warner, 2004). At the same time, as the analytic-scientiic approach was running out of steam, the traditional methodology based on typology or style was being lost in treatises on form, aesthetic valuations, and other subjective approaches that betrayed the intellectual legacy of Worsae and Montelius as much as that of Winckelmann. Unrelated with this situation, in a diferent context, somewhere between archaeology and art, midway between the academic world and that of museums, in restoration laboratories and gold workshops, archaeologists and historians were building up a corpus of knowledge on traditional gold working processes based on the observation of artefacts, craftsmen’s practices, and knowledge regarding the technical conditions of the historical period under consideration. One of the pioneers of this line of research was Herbert Maryon, who published his book Metalwork and Enamelling in 1912, an enormous work which remained in print until 1971. his line of work requires experimentation as a method of research into ancient technology. One great such experimenter was Fortunato Pio Castellani (1794-1865), founder of the Archaeological School of Jewellery in the mid19th century, followed by his sons Augusto and Alessandro (Castellani, 1861), who were astonished by the skill and dificulty of ancient gold working techniques, although some would prefer to describe them as great tricksters. However, that is another story (Bury, 1975; Munn, 1981; 1983). 23 Wands in the hand! Or Potter’s powers 3. THE PRODIGIOUS DECADE Between 1980 and 1990, the panorama changed radically in almost all European countries, where research on metalwork incorporated archaeometry, that is the application of scientiic methods of identiication, measurement and quantiication of the archaeological and historical heritage, with the aim of interpreting, dating, observing, restoring and displaying that heritage to a public that was becoming increasingly interested in its past. In my opinion, two fundamental causes were responsible for the new situation. Firstly, it was the empiricist, or, better said, objectualist trend that favoured archaeological practices whose paradigm was to achieve objectivity: archaeometric data would thus be the only unquestionable data as far as postmodern research is concerned. Secondly, new, non-destructive analytical methods were reined and became increasingly powerful, precise, inexpensive and accessible, and others were adapted to the needs of a material with a high intrinsic and museum value. For example, the development of portable X-ray luorescence (XRF) equipment led to an exponential growth of the number of metal objects analysed before and after 1980. During this same stage, lead isotope ratio analysis began to be ofered for copper-based alloys, which, together with the analysis of trace element patterns, made it possible to begin investigating the mineral’s provenance (Gale and Stos-Gale, 1982), opening up the possibility of proposing interpretive models for the circulation and exchange of raw materials. It has been a long and strange trip indeed… (Pollard, 2009). As with all aspects of research, funding is the determining factor for ensuring the health and future of a line of research. Along these lines, the policy adopted by the United Kingdom since the 1990s in endowing archaeometry with special funding is signiicant, and explains its current world leadership in this ield (Killick, 2008). Recent bibliometric studies (López-Romero and MonteroRuíz, 2006; Costa Caramé, 2008) have shown that the elemental analysis technique most used in archaeometallurgical research since the mid-1980s has been electron microscopy (SEM), accounting for 25.9% of the ten techniques considered (Fig. 1). Occupying the second place is X-ray luorescence (XRF) with 20.0%. Another technique that particularly interests us is particle induced X-ray emission (PIXE): the percentage of its use among the techniques considered is only 5.5% over the same period, but, considering its high cost and the complex infrastructure required, this is not an unimpressive igure. In the case of gold, the current trend has led, as in other areas of archaeometallurgy, to the combination of various analytical techniques and modes of observation not only Figure 1: he most frequently used elemental analysis techniques (based on López-Romero and Montero-Ruíz, 2006). Figure 1 : Les techniques d’analyse les plus rependues (d’après LópezRomero et Montero-Ruíz, 2006). as methods of control and evaluation, but as ways of responding to the many questions raised by an ancient artefact: when it was made, how, why, by whom and for whom: in short, its life history and present destiny (Gosden and Marshall, 1999). It thus appears necessary to go back to the artefact, because it contains all the technological information, and, in its context, all the symbolic and ideological information pertaining to it. However, this process is not without problems. he greatest diiculty that has been detected in the course of these years is the dialogue between the archaeological scientist, on one hand, and the archaeological historian or anthropologist, on the other (Jones, 2004). However, this is, in my opinion, a question of perspective rather than a real problem, which is expressed by resorting to the old duality of materiality versus meaning, object versus subject. 4. TODAY he study of a gold artefact should not be an isolated and anecdotal event. here is no point in using the most powerful and sophisticated analytical technique available if we are unable to set the artefact in its socio-technological context – society explains technology as much as the technology explains society. In order to achieve this, planning and method are important, because materiality cannot be investigated stripped of its meaning, at least not in historical terms. At present, we can no longer conceive of research without a prior methodological approach that does not fulil one of the following conditions: systematic / interdisciplinary / ArcheoSciences, revue d’archéométrie, 33, 2009, p. 21-28 24 Alicia PEREA experimental. hese tools ensure a greater prospect of success when confronting the problems presented by gold, even with its exceptional character, because in our ield of study the exceptional is usually the norm. I should like to present two cases that have posed very different problems for research. In both these cases, inding the solution has involved turning to the knowledge and experience of other researchers and experts, and to instrumental techniques of observation and analysis. he human factor can never be replaced by technology. 5. THE CASE OF VISIGOTH GOLDWORK: GUARRAZAR AND TORREDONJIMENO he Visigoths reached the Iberian Peninsula in 418 and founded a Hispanic Kingdom that lasted until 711. At that time, within a year, the devastating Islamic invasion launched from Jebel al-Tarik (Gibraltar) and took over the major cities, including Toledo, the capital of the Visigoth Kingdom. his led to destruction and plunder. Only two treasures, hidden before the imminent Arab invasion, remain of the ancient splendour of the Visigoth court and church, which had adopted the Byzantine rite. One was found in Guarrazar, near Toledo, and the other in Torredonjimeno, near Jaén. Both consisted of crowns and crosses that kings, courtiers and eminent dignitaries ofered to certain churches in order to obtain divine favour and demonstrate their temporal power to human subjects – among them, the famous crown of King Reccesuinth (653-672). he Guarrazar treasure has been divided and is currently on display in various museums in Madrid and Paris, which adds to the diiculty of examining a valuable and unique material. Nevertheless, the assemblage has been studied in a systematic and interdisciplinary way. he entire collection, consisting of ten crowns and eight crosses (Fig. 2), was analysed using PIXE, since we thought this efort was justiied in view of our lack of knowledge of the processes of Visigoth manufacture and raw materials. At that time, there was no particle accelerator with an external beam in Spain, so we had to obtain European funding for this project. With the help of two COST Actions, in 1997 and 1999, we gained access irst to the LARN (Laboratoire d’Analyses par Réactions Nucleaires) in Namur (Belgium), to analyse a series of samples that had been taken from the part of the treasure kept in the Museo Arqueológico Nacional, Madrid, and second, to the AGLAE accelerator of the Centre de Recherche et de Restauration des Musées de France, in the Louvre complex, where all the pieces kept in the Musée Nationale du Moyen Âge de Cluny (Paris) were analysed using PIXE and PIGE techniques. An interArcheoSciences, revue d’archéométrie, 33, 2009, p. 21-28 Figure 2: (See colour plate) 19th century illustration of the Guarrazar Treasure (de Lasteyrie du Saillant, 1860). Figure 2 : (Voir planche couleur) Illustration du XIXe siècle du trésor de Guarrazar (de Lasteyrie du Saillant, 1860). disciplinary team subsequently studied the metals, gems and glass, and also the various technical, social, economic and historical aspects pertaining to these objects. he results were published in a monograph (Perea, 2001). he Guarrazar project was a successful experience of international and interdisciplinary collaboration. It is still opening up new avenues of research: for example, the micro-samples that were extracted from the pieces in the Museo Arqueológico Nacional de Madrid to be analysed in Namur were subsequently re-analysed in Paris, modifying the previous conditions. he aim was to detect the trace elements in order to determine the provenance of the gold (Guerra et al., 2004; 2007). he second phase of this research project, in 2004, consisted of analysing the Torredonjimeno treasure, which is also distributed among three museums in diferent Spanish cities: 25 Wands in the hand! Or Potter’s powers Barcelona, Cordoba and Madrid. his treasure contained royal crowns and votive crosses, like the previous one, but it was in very poor condition. Most of the pieces had been broken up, the royal crowns had disappeared, and we only knew they had existed from the letters that had hung from the diadem, as in the case of the Reccesuinth crown. At irst glance, the gold appeared to consist of alloys with very low purity, and containing pieces made directly from silver. In addition to the terrible fragmentation of the crosses, some restoration work had damaged the pieces even more (Fig. 3a, b). In view of the impossibility of analysing almost two hundred pieces or fragments, a sampling of 29 objects was carried out, and these were analysed by PIXE and PIGE in the Tandetron accelerator of the Centro de Microanálisis de Materiales of the Universidad Autónoma de Madrid, which by this time had an external beam. On this occasion, we found to our surprise that most of the pieces were not made of gold, but of silver gilded with a mercury amalgam, an aspect that had not been detected in the entire history of research on this collection since it was discovered in 1926. Once again, an interdisciplinary team studied the various archaeometric and historical aspects pertaining to the artefacts. he result was a second monographic volume, recently published (Perea, 2009), which completes the series on Visigoth goldwork in Spain. 6. THE CASE OF THE BRAGANZA BROOCH, OR AN EXCEPTIONAL PIECE Let us leave aside the great treasures in order to discuss the small, exceptional and unique. he case of the Braganza Brooch is one of them (Fig. 4). his ibula, 14 cm long, made in gold and enamel, has a long history behind it, not just because it dates to a period around the 4th-3rd century BC, but also because of its bizarre contemporary history, once being in the possession of the Portuguese Royal House of Braganza at some point in the 19th century. But that is another story, for which there is no space in the present paper. he period that concerns us begins in 2001, when the British Museum purchased this jewel in the Christie auction of 25 April. Although perhaps it is better to begin the story around 1956, when one of its for- Figure 3a, b: (See colour plate) he cross MAC 25093 from the Torredonjimeno Treasure, made up of at least two diferent fragmented objects (a), and the arrangement as seen from the back (b) (Photograph by Archivo Au, O. García-Vuelta). Figure 3a, b : (Voir planche couleur) La croix MAC 25093 du trésor de Torredonjimeno, constituée de fragments d’au moins deux objets diférents (a), et leur disposition à l’arrière (b). (Photographie par l’Archivo Au, O. García-Vuelta). Figure 4: (See colour plate) he Braganza Brooch (he British Museum. Photograph by Archivo Au, A. Perea). Figure 4 : (Voir planche couleur) La broche de Braganza (he British Museum. Photographie par l’Archivo Au, A. Perea). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 21-28 26 mer owners sent some photographs to the British Museum in order for them to get an idea of what he actually had in his collection… and it is at this point that a controversy, which for some people has still not been resolved, ensued. At that time, Bernard Ashmole did not consider it to be an original ancient work. For his part, Paul Jacobsthal completely disagreed and thought it was “easily one of the most exciting antiques I have come across”. From that time onward, the most varied opinions concerning this piece succeeded each other, limited however to the academic world. In spite of this, in the Iberian Peninsula, its place of origin, these debates passed completely unnoticed. We then had to wait until 1993, when the Braganza Brooch was deposited in the British Museum on loan and was submitted to an exhaustive study that conirmed, irstly, its age, and secondly, that nothing, ancient or modern, had been added to it. his model study, carried out by Ian Stead and Nigel Meeks (1996), is based on the observation of work traces using an optical microscope and a scanning electron microscope (SEM). In order to characterise the material, an elemental analysis was undertaken using a combination of X-ray luorescence and EDX analysis in the SEM. he enamel was analysed by X-ray difraction in order to identify the opaciier and by SEM/EDX for composition and colorants. he results brought to an end an absurd controversy based solely on ignorance and prejudice in response to an exceptional object. In 2007, with the help of a collaboration with the British Museum in the person of Dyfri Williams, the Braganza Brooch came to Spain – most probably its place of origin – to be presented to the Spanish public, who never before had the opportunity to admire it (Perea et al., 2007). his seemed to me a unique opportunity to organise a small symposium for scientiic discussion among specialists (Perea, in press). We can now state with certainty that the brooch belongs to a very special type of Iberian ‘bent-back foot’ ibulae made in gilded silver, which usually depict hunting scenes. It was probably ofered to an Iberian sanctuary by the prince who ordered it for a special occasion. It also became clear that the goldsmith who made it was an exceptional artist, whose eye was familiar with the Hellenistic aesthetics of the time, and was very probably of Greek origin. However, uncertainties and disagreements regarding this object still remain, mainly related to the iconographic interpretation of the scene depicted on it. 7. AND THE FUTURE?… I do not think the immediate future holds major surprises, although new avenues of research are still being inaugurated, ArcheoSciences, revue d’archéométrie, 33, 2009, p. 21-28 Alicia PEREA mainly in two ields: studies looking at the provenance of the gold, and methods of dating. Concerning the provenance of gold, much work has been carried out in this direction and good results produced by characterising trace element patterns (Guerra et al., 1999). he most frequently used techniques are inductively coupled plasma mass spectrometry (ICP-MS) and proton activation analysis. However, there is still a need to characterise natural gold, despite the work that has been undertaken, particularly in Ireland, where the study of gold has traditionally been a topic of ‘national interest’ (Chapman et al., 2006). Other recent research has centred on the analysis of osmium isotope ratios for those samples of gold that display metal inclusions belonging to the platinum group (Junk and Pernicka, 2003); lead isotope signatures have been used for Au-Ag-Cu alloys with a gold content of more than 70% (Bendall et al., 2009). Both systems have been used to study coins. his approach sufers from the same aforementioned problem of a lack of natural gold references. With regard to the methods of dating, experiments are currently being undertaken with the Uranium/horiumHelium content of gold in order to detect fakes, although the technique is still in the experimental stage (Eugster et al., 2008; 2009). Scientiic methods for investigating gold, like any other metal, are now essential for a serious examination of its history and meaning. However, we cannot solve all the problems or answer all the questions with analytical techniques alone. As in the story of Harry Potter, the magic wand never works if it is not in the hands of the right person. he one Potter chose measured eleven inches and was made of holly and phoenix feather (Rowling, 1998: 65). Note his work is part of a research project entitled “Grounds for an Archaeometric and Technomic Research on Metallurgy during Prehistory and Antiquity. he Iberian Peninsula” (Ref.: HUM2006—06250/HIST) within the Programme CONSOLIDER INGENIO 2010 (CSD-TCP), funded by the Ministry of Science and Innovation, Spain. References ADRIAENS, A., DEGRIGNY, C. and CASSAR, J. (eds.), 2005. Beneits of non-destructive analytical techniques for conservation. Papers from a COST Action G8 workshop held in Kalkara, Malta, on 8 January 2004. Luxembourg, Oice for Oicial Publications of the European Communities. Wands in the hand! Or Potter’s powers BENDALL, C., WIGG-WOLF, D., LAHAYE, Y., VON KAENEL, H.M. and BREY, G.P., 2009. Detecting changes of Celtic gold sources through the application of trace element and Pb isotope laser ablation analysis of Celtic gold coins. Archaeometry 51(4): 598-625. BURY, S., 1975. Alessandro Castellani and the revival of granulation. Burlington Magazine 117(871): 664-668. CASTELLANI, A., 1861. A memoir on the art of the goldsmith in Ancient Times. he Archaeological Journal 18: 365-368. CHAPMAN, R.J., LEAKE, R.C., WARNER, R., CAHILL, M., MOLES, N.ÇR., SHELL, C. and TAYLOR, J.J., 2006. Microchemical characterisation of natural gold and artifact gold as a tool for provenancing prehistoric gold artefacts: a case study in Ireland. Applied Geochemistry 21: 904-918. COSTA CARAMÉ, M.E., 2008. Estado actual de la investigación arqueometalúrgica en España: una aproximación bibliométrica, in S. Rovira Llorens, J. García-Heras, M. Gener Moret, I. Montero-Ruiz (eds.), Actas VII Congreso Ibérico de Arqueometría (Madrid, 8-10 October 2007). Madrid, CSIC, 398-409. DEMORTIER, G. and ADRIAENS, A. (eds.), 2000. Ion beam study of art and archaeological objects. A contribution by members of the COST G1 Action. European Community (EUR 19218). Luxembourg, Oice for Oicial Publications of the European Communities. EUGSTER, O., KRAMERS, J. and KRÄHENBÜHL, U., 2008. Detecting forgeries among ancient gold objects using the U, h-4He dating method. Archaeometry 50: 1-10. EUGSTER, O., PERNICKA, E., BRAUNS, M., SHUKOLYUKOV, A., OLIVE, V. and ROELLIN, S., 2009. Helium, uranium and thorium analyses of ancient and modern gold objects: estimates of their time of manufacturing. ArcheoSciences 33. FERRO, D., FORMIGLI, E., PACINI, A. and TOSSINI, D., 2003. Experimental Archaeology approach to the study of copper salts weldings, in Archaeometallurgy in Europe. Proceedings: International Conference, 24-26 September, 2003, vol. I. Milano, Associazione Italiana di Metallurgia, 305-314. FERRO, D., FORMIGLI, E., PACINI, A., TOSSINI, D., 2008. La Saldatura nell’Oreiceria antica. Indagini archeometriche e archeologia sperimentale. Roma, CNR. Edizioni Kappa. GALE, N.H. and STOS-GALE, Z.A., 1982. Bronze Age copper sources in the Mediterranean. Science 216: 11-19. GOSDEN, C. and MARSHALL, Y., 1999. he cultural biography of objects. World Archaeology 31(2): 169-178. GUERRA, F. and SARTHRE, C.O., GONDONNEAU, A. and BARRANDON, J.N., 1999. Precious metals and provenance inquiries using LA-ICP-MS. Journal of Archaeological Science 26: 1101-1110. GUERRA, F., PEREA, A. and CALLIGARO, T., 2004. Coins and crowns. A irst approach to the gold characterization of the treasure of Guarrazar (Toledo, Spain), in A. Perea, I. Montero-Ruiz, 27 O. García-Vuelta (eds.), Tecnología del Oro Antiguo: Europa y América. Madrid, CSIC, 417-422. GUERRA, F., CALLIGARO, T. and PEREA, A., 2007. he treasure of Guarrazar: tracing the gold supplies in the Visigothic Iberian Peninsula. Archaeometry 49(1): 53-74. HARTMANN, A., 1970. Prähistorische Goldfunde aus Europa I. Studien zu den Anfängen der Metallurgie, Band 3. Berlin, Mann. HARTMANN, A., 1982. Prähistorische Goldfunde aus Europa II. Studien zu den Anfängen der Metallurgie, Band 5. Berlin, Mann. JONES, A., 2004. Archaeometry and materiality: materials based analysis in theory and practice. Archaeometry 46(3): 327-338. JONES, M. (ed.), 1990. Fake? he Art of Deception. London: Published for the Trustees of the British Museum by British Museum Publications. JUNK, S.A. and PERNICKA, E., 2003. An assessment of osmium isotope ratios as a new tool to determine the provenance of gold with platinum-group metal inclusions. Archaeometry 45(2): 313-331. KILLICK, D., 2008. Archaeological science in the USA and Britain, in A. Sullivan (ed.), Archaeological Concepts for the Study of the Cultural Past. Salt Lake City, University of Utah Press, 40-64. DE LASTEYRIE DU SAILLANT, F.-C.-L., 1860. Description du Trésor de Guarrazar. Paris, Gide Éditeur. LÓPEZ-ROMERO, E. and MONTERO-RUÍZ, I., 2006. Archaeometry and the international evolution of studies on metallurgy: a bibliometrical perspective, in 34th International Symposium on Archaeometry, Zaragoza (2004). Zaragoza, CSIC, 195-200. MARYON, H., 1971. Metalwork and Enamelling (5th revised edition). New York, Dover Publications. MUNN, G.C., 1981. Jewels by Castellani. Some sources and techniques examined. he Connoissseur 206(828): 126-131. MUNN, G.C., 1983. Les Bijoutiers Castellani et Giuliano. Retour à l’Antique au XIXe siècle. Fribourg, Oice du Livre, Éditions Vilo. PEREA, A. (ed.), 2001. El Tesoro Visigodo de Guarrazar. Madrid, CSIC. PEREA, A. (ed.), 2009. El Tesoro visigodo de Torredonjimeno. Madrid, CSIC, Ed. Polifemo. PEREA, A. (ed.), in press. La Fíbula Braganza. Diálogos bilingües en torno a la excepcionalidad. Madrid, Ed. Polifemo PEREA, A. and ARMBRUSTER, B., 2008. L’Archéologie de l’or en Europe. Perspective. La revue de l’INHA 1: 29-48. PEREA, A., WILLIAMS, D. and OLMOS, R., 2007. El Héroe y el Monstruo. Exhibition Catalogue (Museo Arqueológico Nacional, 27 April-29 July). Madrid, Ministerio de Cultura. POLLARD, A.M., 2009. What a long strange trip it’s been: Lead isotopes and Archaeology, in A.J. Shortland, I.C. Freestone, h. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 21-28 28 Rehren (eds.), From Mine to Microscope. Advances in Ancient Technology. Oxford, Oxbow Books. POLLARD, A.M. and BRAY, P., 2007. A bicycle made for two? he integration of scientiic techniques into archaeological interpretation. Annual Review of Anthropology 36: 245-259. REHREN, TH. and PERNICKA, E., 2008. Coins, artefacts and isotopes. Archaeometallurgy and Archaeometry. Archaeometry 50(2): 232-248. ROWLING, J.K., 1998. Harry Potter and the Philosopher’s Stone. London, Bloomsbury. STEAD, I.M. and MEEKS, N.D., 1996. he Celtic Warrior Fibula. he Antiquaries Journal 76: 1-16. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 21-28 Alicia PEREA TAYLOR, J.J., 1980. Bronze Age Goldwork of the British Isles. Cambridge, Cambridge University Press. WARNER, R., 2004. Irish gold artifacts: observations from Hartmann’s analytical data, in H. Roche, E. Grogan, J. Bradley, J. Coles, B. Raftery (eds.), From Megaliths to Metal. Essays in Honour of George Eogan. Oxford, Oxbow Books: 72-82. WATERBOLK, H.T. and BUTLER, J.J., 1965. Comments on the use of metallurgical analysis in prehistoric studies. Helinium 5: 228-234. Targeting Ion Beam Analysis techniques for gold artefacts Techniques de faisceaux d’ions ciblés pour l’analyse d’objets en or Guy Demortier* Abstract: he present study discusses the best experimental conditions for the quantitative analysis of gold jewellery artefacts by ion beam techniques (PIXE, RBS, PIGE and NRA). Special attention is given to the detection of enhancement or depletion below the surface, down to 10 microns, without any sampling or destruction. PIXE is certainly the most interesting technique for this purpose and the optimal geometrical arrangement of the experiment is described: orientation of the incident beam relative to sample surface, selection of the characteristic signals, use of reference material, choice of the detector and incident particle energy and type, computation of absorption and secondary luorescence efects. Résumé : Nous discutons des conditions idéales d’utilisation des faisceaux de particules pour l’étude d’objets en or (PIXE, RBS, PIGE, et NRA). On attache une attention particulière au choix des paramètres expérimentaux pour détecter la présence éventuelle d’efets de surface sur une profondeur de l’ordre de 10 microns, sans prélèvement ni destruction. Le PIXE s’avère la technique idéale pour autant qu’on prenne en compte une série de paramètres comme l’orientation du faisceau incident relativement à la surface, choix des signaux pour la caratérisation, choix des échantillons de référence, choix du détecteur, choix de l’énergie et du type des particules incidentes, calcul des efets d’absorption et de luorescence secondaire. Keywords: PIXE, PIGE, RBS, NRA, surface enrichment, luorescence efects, solders Mots-clés : PIXE, PIGE, RBS, NRA, enrichissement supericiel, efets de luorescence, soudures. 1. INTRODUCTION Gold artefacts are often considered as un-corroded items which can be analysed for their bulk composition by surface techniques probing thicknesses of one micron or more. All of these methods involve ion beam techniques using accelerators delivering particles in the MeV range. We have demonstrated that this statement is not absolute (Ruvalcaba and Demortier, 1996; Demortier and Ruvalcaba, 2005), and that surface enhancement of gold (and consequently copper depletion) is often veriied in pre-Columbian jewellery items. he presence of this enhancement can be rapidly identiied by PIXE with incident protons of at least two diferent energies. In the present study, we discuss the best experimental conditions required to quantitatively proile the elemental depth distribution of Au vs. Cu and Ag by PIXE and RBS, but also to analyse low concentrations of elements like Fe, Zn, Cd by PIXE, and low Z elements like Al, Si and S, which are often present as dust or remains of the alloying procedure, but at low concentration, in the surface grain boundary of the basic Au-Ag-Cu alloy. * LARN (University of Namur – Belgium) and CEDAD (University of Salento – Italy), 24, rue Sergent Delisse B-5004 Bouge. (guy.demortier@ tvcablenet.be) rec. Aug. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 29-38 30 Guy DEMORTIER 2. AVAILABLE ION BEAM TECHNIQUES WITH LOW ENERGY ACCELERATORS FOR THE ANALYSIS OF GOLD JEWELLERY ARTEFACTS When a material is irradiated with charged particles like protons, deuterons, α-particles or other heavier ions, with energy ranging from a few hundred keV to a few MeV, the emission of photons, charged particles and neutrons makes available various spectroscopic methods of elemental analysis. he great majority of the interactions of the incident charged particles with the atoms of the irradiated material take place with atomic electrons: the incident particle progressively loses its energy and stops after crossing a distance R, known as the range. By interacting mainly with atomic electrons and since interaction with an atomic nucleus occurs only rarely, this incident particle does not experience an appreciable deviation. he range R is then nearly the total distance that the particle travels following a straight line in the material. R increases with the incident projectile energy (R ∼ E1.8), and decreases with the atomic number Z and its mass M. his property of a linear trajectory of protons in the materials allows us to use them to irradiate the material under analysis outside the vacuum of the accelerator. he incident beam crosses a thin foil of material (mostly aluminium, kapton or silicon nitride), as shown in Figure 1, where we have compared the trajectories of incident protons and electrons on a metallic target. he simple linear trajectory of the incident protons contrasts with the complicated one of electrons. One immediately sees the advantage of using an external proton beam for the easy positioning of the artefact outside the vacuum chamber of the accelerator. Since the pioneering work of the LARN team in 1972 (Deconninck, 1972), constant improvements of the external beam technology have been made in many laboratories involved in archaeometry. he interaction of the incident protons with the atomic electrons of the target (i.e., the sample under investigation) induces ionisation. When ionisation takes place in an inner electronic shell, a subsequent electronic rearrangement to return to an atomic equilibrium leads to the emission of characteristic X-rays: the analytical technique based on this atomic process is called PIXE (Particle Induced X-ray Emission). PIXE is a very sensitive method for the bulk and trace analysis of elements with an atomic number Z greater than 20. PIXE may also be used for the study of elements from Na to K, but the absorption of the low energy characteristic X-rays (1-3 keV) of these light elements into thick samples generally restricts the potentiality of the method for quantitative determination. When an incident particle passes very close to the atomic nucleus, it may be elastically scattered. At very low impact parameters, an incident particle of mass m, lower than the mass M of the collided nucleus, can be scattered at an angle close to 180°. he spectroscopy of these backscattered particles (RBS) is often recommended for depth proiling of elements with thicknesses up to 30% of their own range. he energy of the backscattered particle is related to the mass of the collided nucleus Figure 1: (a-b) Comparison of the interaction of electrons and protons with copper in vacuum. (c-d) Use of the trajectory of protons for non-vacuum analysis. Figure 1 : (a-b) Comparaison de l’interaction des électrons et des protons avec le cuivre dans le vide. (c-d) Utilisation des trajectoires des protons pour l’analyse en dehors de l’enceinte de l’accélérateur. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 29-38 Targeting Ion Beam Analysis techniques for gold artefacts and to the depth in the material at which the scattering took place. When protons are used as incident particles, K (called the kinematic factor) is very close to 1 for all the components of Au-Ag-Cu alloys (K = 0.94 for Cu, 0.96 for Ag and 0.98 for Au), and the depth proile is very limited. he use of heavier projectiles like α−particles (with K = 0.78, 0.86, 0.92 for Cu, Ag and Au, respectively) is not recommended, due to their low range in materials (about 12 times less than for protons), and consequently their important energy loss in the exit foil of the accelerator, which prevents the use of an external beam, apart from using a very narrow beam crossing a thin Si3N4 membrane of 0.1 µm or so. Furthermore, the linear energy transfer of heavy incident ions in the irradiated sample would produce an undesired heating of the precious samples when the jewellery item is bombarded in the vacuum. As the cross-section for such elastic scattering is proportional to the Z2 of the target atoms, RBS is better suited for the study of low concentrations of high Z elements in light matrices, and is then mostly incompatible with the study of low concentrations of copper and silver in gold-rich jewellery items. he probability for a Coulomb excitation of a target nucleus leading to the emission of a characteristic γ-ray is much lower than both the probability of backscattering leading to RBS and the probability of inner shell ionisation phenomena leading to PIXE. Consequently, PIGE (Particle Induced γ-ray Emission) is of particular interest for the study of light elements, where PIXE and RBS sufer from a lack of sensitivity and accuracy. For these light target elements, the energy of the characteristic X-rays is so low that they are absorbed into the material itself and RBS ofers a lower cross section. If the energy of the incident particle is maintained below 3 MeV, no delayed radioactivity may be induced as in CPAA (Charged Particle Activation Analysis), typically performed at incident energies greater than 10 MeV. he interaction of a loosely bound compound projectile (like a deuteron) gives rise to the emission of energetic protons after the stripping of the incident ion. his is known as NRA. For instance, the breakdown of the incident deuteron may be written as: 32S (d, p0)33S. he stripping of the incident deuteron by a 32S nucleus leads to the capture of a neutron in 32S and leaves the 33S nucleus in its fundamental state, with an emitted proton with a maximum energy. Various other procedures could be employed to leave the residual nucleus of 33S in one of its (i = 1, 2, 3) excited states. he corresponding energies of the pi protons are then 31 lower than those for p0. he stored energy in the excited 33S* nucleus will shortly give rise to a cascade of characteristic γ-rays. Several groups of protons of diferent energies, as well as γ-rays, may be used for analytical purposes. As deuterons of energies in the MeV range cannot approach heavy nuclei like Cu, Ag and Au due to the Coulomb barrier, this type of (d, p) reaction is particularly useful to complement PIXE and RBS. Sulphides are indeed expected to be sometimes present in solders of gold artefacts if ores were directly alloyed with a basic Au-Ag-Cu to obtain a low melting point alloy suitable for soldering. By using simultaneously (or sequentially) all these analytical techniques, one may achieve a very powerful characterisation of gold jewellery items. We can then summarise our argument by this inal recommendation regarding ion beam techniques: PIXE is the most appropriate method for the elemental analysis of the gold items, achieving the necessary accuracy for a complete quantitative determination of medium and heavy elements; PIGE would be used only if it is compulsory to analyse light elements. RBS, usually recommended to provide depth information on medium and heavy elements, is generally not suiciently eicient for quantitatively separating information on Cu, Ag and Au components, due to the high content of Au, but would be interesting to use as a complement to the PIXE results on items which are non-homogeneous in their depth. NRA, inducing the transmutation of light elements, would complement PIGE in determining their depth proile. Below 3 MeV, no delayed radioactivity is induced. Another popular, quantitative and non-destructive analytical technique whose performances are similar to PIXE is XRF. he methodology used to detect elements is exactly the same for PIXE and XRF. he main diference between the two lies in the ionisation process of the inner shells of atoms. In XRF, the incident radiation is a photon whose energy is meant to be higher than the ionisation energy of the atom to be analysed, and also higher than the energy of the characteristic X-ray line to be detected. he depth from which the characteristic L lines of Au, and K lines of Ag and Cu are emitted is then deeper than for PIXE, due to the high energy of the Ag lines. Incident X-rays of about 35 keV are then necessary, which would involve an important absorption of emitted signals from Cu and Au. Working with lower incident photon energy would require to analyse Ag with their characteristic L lines which would concern a too low bulk information. Nevertheless, XRF is a powerful technique, much cheaper than PIXE, and is likely to be more extensively used than PIXE due to its easiness of transportation to museums and excavation sites. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 29-38 32 Guy DEMORTIER 3. TARGETING PIXE FOR GOLD JEWELLERY ITEMS With the exception of trace elements, PIXE is a very appropriate technique for the quantitative analysis of major and minor elements in gold jewellery artefacts. To increase the sensitivity for the study of objects with low concentrations of Cu and Ag, it is often compulsory to selectively decrease the intensity of the AuL X-ray lines by inserting a thin foil (20 to 30 microns) of Zn (or Cu) between the target and the detector. he huge increase of the absorption at 9.66 keV for Zn and at 8.98 keV for Cu justiies their use in the iltering of gold L lines. A 20 microns-thick foil of Zn reduces the intensity of the Lα line by a factor of 45 and that of the Lβ line of Au by a factor of 11.5, but only by factors of 2.4 and 1.5 the Kα lines of Cu and Ag, respectively. One can see (Table 1) that, for an interference-free detection, the Kα lines of Cu and Ag, and the Lβ line of Au are the most appropriate. he calculation of the concentration of Cu would nevertheless include the computation of the escape peak of the Lα line of Au in the region of the Kα line of Cu. he Zn concentration would be impossible due to the complete interference of both Zn K-lines with the Ll and Lα lines of Au. he concentration of Cd (often useful for the detection of fakes or modern repairs) is not easy, due to the interference with the sum peak (twice 11.5 keV) of the very intense Lβ line of Au. he complete computation of the elemental concentrations is carried out using a home-made programme taking into account all the necessary physical processes, including the secondary luorescence induced mainly by the Au L lines in Cu (Van Oystaeyen and Demortier, 1983; Demortier and Morciaux, 1994), or by using other commercial computer programmes. Element Fe Cu Zn Ag Cd Au X-ray Energy (keV) Kα 6.4 Κβ 7.06 Kα 8.04 Κβ 8.9 Kα 8.64 Κβ 9.6 Kα 22,1 Κβ 24.9 Kα 23.1 Κβ 26.1 Lα 9.6 Lβ 11.5 Ll 8.4 Most of the PIXE users working in the ield of archaeometry select protons as incident particles in a non-vacuum geometry and tune their accelerator above 3 MeV (Calvo del Castillo et al., 2008; Bugoi et al., 2008; Migliori et al., 2008; Quarta et al., 2008; Mathis et al., 2008). heir main arguments for this are: – the higher the incident energy, the lower is the energy deposited in the exit foil; – the higher the incident energy, the higher is the crosssection for X-ray emission for all the elements of the target (see Fig. 2). Let us comment on this general attitude. In order to probe the maximum distance in the irradiated material, the maximum incident energy seems to be the most appropriate, and the sample is to be irradiated with its surface normal to the beam direction, as illustrated in Figure 1d. In this geometry, the X-ray detector (made with a silicon wafer) is situated at an angle of about 135° relative to the incident beam and the outgoing X-rays path is longer by a factor of 1.42 relative to the incident particle path. In Table 2 we provide the proportion of outgoing characteristic X-rays of Cu, Ag and Au for a series of typical gold alloys. One can observe (irst line in the table) that only 22.5% of the entire intensity of the Au Lα produced by 3 MeV protons may escape from a homogeneous alloy (50% Au, 25% Ag and 25% Cu sample), but this proportion increases up to 67.8% when the incident energy is lowered down to 1 MeV. he detected Kα line of Cu follows the same trend, but the Ag Kα is much less afected by this absorption efect. At proton incident energies above 2 MeV, most of the emitted X-rays of Cu and Au are absorbed in the target itself. For low concentrations of copper in gold rich Selection yes Remarks traces of ores Interference escape of CuKα yes main or minor escape of AuLα traces ? with AuLl with AuLα yes yes large absorption in ilter sum twice AuLβ low intensity with ZnKβ yes ArcheoSciences, revue d’archéométrie, 33, 2009, p. 29-38 with ZnKα Table 1: Characteristic X-ray lines used for the analysis of gold jewellery items. Tableau 1 : Raies X caractéristiques utilisées pour l’analyse des objets d’orfèvrerie. 33 Targeting Ion Beam Analysis techniques for gold artefacts parison with a homogeneous reference material containing all the elements of interest. Incident beam intensity, detector eiciency, detector solid angle, duration of the irradiation, computation of absorption of X-rays in the material between the sample and the detector would not be measured (Demortier et al., 1990). 4. DIFFERENTIAL PIXE FOR DEPTH PROFILING OF GILDED ARTEFACTS Figure 2: Variation of the ionization cross-section for K and L shells with protons. Figure 2 : Variation de la section eicace d’ionisation des couches K et L avec des protons. artefacts, this absorption efect reaches 85% at 3 MeV (last line in Table 2). he quantitative determination is then very highly afected by irregularities in the artefact’s surface or by the incorrect reproducibility of the sample surface orientation relative to the direction of the incident beam. We also observe that the absorption of the Lα line of gold is much higher than the absorption of the Lβ line, justifying the reason for selecting this Lβ line for the calculation of gold concentration. Due to its high energy, the Kα line of silver is less sensitive to any absorption efect. A careful choice of the position and of the size of the zinc ilter is compulsory in order to avoid additional detection of zinc K X-ray lines in the detector. his luorescence efect has been extensively discussed in previous works [Demortier et al., 1993; Demortier and Morciaux, 1994). When all the parameters of the geometry of detection are known, PIXE is an absolute method of analysis, but the technology is even more straightforward when the measurement is made by com- A layered structure cannot be detected by PIXE using incident protons at a single energy. As explained above, RBS is not able to easily separate scattered particles on Cu, Ag and Au. Diferential PIXE is able to achieve this goal even for thicknesses up to 10 µm (Demortier and Ruvalcaba, 2005). he CuKα/AuLβ intensity ratio is very diferent for a homogeneous material or a layered structure, and this ratio varies in opposite ways if the material is irradiated at various incident proton energies. he explanation is given in Figure 4. In full lines we give the variation of the CuKα/AuLβ ratio for a layered structure and in doted lines the behaviour for homogeneous alloys. hus, if a sample made of pure Cu covered with a 4 µm thick layer of gold is irradiated with 1.8 MeV protons, the ratio is 0.86 (point A), but increases up to 2.05 if the incident proton energy is 2.6 MeV (point B), and decreases down to 0.17 (point D) if the proton energy is 1.2 MeV. At 1.8 MeV, this ratio of 0.86 would correspond to a homogeneous alloy containing 92% Au and 8% Cu, but this ratio would decrease down to 0.71 at 2.6 MeV and increase up to 0.96 at 1.2 ΜeV. he variations are then in the opposite directions. Furthermore, the variation is very sharp for a layered structure, and smooth for a homogeneous alloy. he slope of the variation of the CuKα/AuLβ ratio is increasing when the incident proton energy is decreased. he choice of an incident energy in the region of 2 MeV is then optimal. In addition to a lower absorption of the X-ray lines at lower proton energy, we add now a better capability to detect surface enhancement or depletion. In order to apply this diferential PIXE technique to materials with a depth proile down to several microns, measurements of the CuKα/AuLβ ratio are to be made at n diferent proton energies to obtain the Cu/Au intensity ratio in n-1 layers and in the bulk. For instance, if we assume that the anthropo-zoomorphic pendant (Quimbaya Peru – 1000 to 1600 AD) shown in Figure 5 is homogeneous, conventional PIXE data obtained with protons of energies ranging from 0.6 to 2.6 MeV would give contradictory results (Table 3) for the analysis at the same impact on the sample but at diferent incident beam energies. hus, the Cu content ArcheoSciences, revue d’archéométrie, 33, 2009, p. 29-38 34 Guy DEMORTIER Au % 50 Ag Cu Au 25 25 50 Ag Cu Au 5 45 75 Ag Cu Au 15 10 80 Ag Cu Au 15 5 96 Ag Cu 3.5 1.5 X-ray Lα Lβ Kα Kα Lα Lβ Kα Kα Lα Lβ Kα Kα Lα Lβ Kα Kα Lα Lβ Kα Kα 3.0 MeV 22.5 26.3 60.8 21.3 21.2 30.0 60.7 26.7 22.2 31.1 51.5 17.5 22.3 31.2 49.7 16.5 21.6 30.2 44.2 15.1 2.5 MeV 28.9 39.4 68.2 27.7 27.4 37.6 68.5 34.2 28.4 38.6 59.5 23.0 28.4 38.6 57.7 21.8 27.6 37.5 52.3 19.8 2.0 MeV 39.9 49.3 75.8 36.9 36.3 47.6 76.8 44.4 37.1 48.1 68.5 31.1 37.2 48.4 66.5 29.4 35.5 46.5 62.0 26.9 1.5 MeV 50.6 61.9 83.9 49.4 49.1 60.5 82.0 56.8 41.8 60.9 77.9 42.8 49.6 60.8 76.4 40.8 47.4 58.7 69.6 27.9 1.0 MeV 67.8 76.6 93.0 65.2 66.7 75.8 95.0 71.8 68.8 75.9 90.0 58.3 66.7 75.7 90.0 56.2 63.9 73.4 100.0 53.8 Table 2: Percentage of outgoing X-rays of Cu, Au and Ag used in the PIXE analysis of jewellery items with protons of various energies. Tableau 2 : Pourcentage des rayons X émergeant du Cu, de l’Au et de l’Ag utilisés pour l’analyse des objets d’orfèvrerie par PIXE avec des protons d’énergies variées. Depth below surface (µm) 0.28 0.28 0.28 0.28 0.28 0.28 bulk Cu % 0 4 8 12 16 24 28 Au % 94 90 86 82 74 70 68 Ag % 6 6 6 6 6 6 6 Table 3: Depth proile of elements in the Quimbaya pendant obtained by diferential PIXE. Tableau 3 : Proils en profondeur dans le pendentif Quimbaya par PIXE diférentiel. appears to be 8.6% at 0.6 MeV incident proton energy (in that case, most of the signal comes from the irst 2 microns below the surface), but increases up to 18.7% at 2.6 MeV proton energy (most of the X-rays are coming in this case ArcheoSciences, revue d’archéométrie, 33, 2009, p. 29-38 from a thicker layer, of about ten microns). he sample is then certainly not homogeneous in depth, but is highly enriched in gold on the surface. Repeating the measurement at 7 diferent proton energies, we have computed by an iteration method (similar to the one used for RBS spectra) that the depth distribution of Cu varied from 0 to 28% from the surface to a depth of 1.7 micron below the surface. he identiication of nearly no copper on the surface was veriied using PIXE with incident low energy α particles. Diferential PIXE has also been applied to other materials (Smit et al., 2008). 5. DETERMINATION OF TRACES OF SULPHUR AND ZINC IN ANCIENT GOLD ARTEFACTS PIXE has been successfully applied for the purposes of topographical elemental analysis of Cu, Ag, and Au in jewellery items. he composition of the material in narrow regions of minute solders on antique gold artefacts may provide valuable information regarding the workmanship Targeting Ion Beam Analysis techniques for gold artefacts 35 Figure 3: Mass absorption coeficient of X-rays in Cu, Zn, Ag and Au, to be used in selective ilters and for XRF induced by PIXE. Figure 3 : Coeicient d’absorption massique des rayons X dans le Cu, le Zn, l’Ag et l’Au à utiliser pour les iltres sélectifs et pour la FX induite par PIXE. of goldsmiths from the Ancient World. We have shown above that Zn is diicult to analyse by PIXE due to the high interference of its K-lines with the L-lines of gold, and that signals of Cd are very often neglected due to the high probability of interference with pile-up signals: the Kα energy of Cd is exactly at the position of twice the Lβ line of gold. In several previous papers, we have discussed the hypothesis that, in ancient times, cadmium sulphide (a yellow mineral with a colour close to that of gold) may have been ‘alloyed’ with gold to obtain a material suitable for brazing. To test this hypothesis, an experimental procedure for the analysis of traces of sulphur in gold artefacts was developed. It is indeed expected that during the ancient alloying procedure, traces of sulphur from the cadmium sulphide were introduced in the brazing alloy. he detection of energetic protons induced by a deuteron bombardment of the specimen under investigation was achieved. he best experimental conditions for sulphur determination are obtained by (d,p) reactions at Ed = 1.9 MeV and a detection angle of protons around 165° (Demortier and Gilson, 1987). S signals are well separated Figure 4: Variation of the CuKα/AuLβ intensity ratio for layered and homogeneous Cu-Au alloys under proton bombardment. Figure 4 : Variation du rapport d’intensités CuKα/AuLβ pour des alliages Cu-Au stratiiés et homogènes sous bombardement de protons. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 29-38 36 Figure 5: (See colour plate) he Quimbaya pendant (Colombia, 1000-1600 AD). Figure 5 : (Voir planche couleur) Le pendentif Quimbaya (Colombie 1000-1600 A.D.) Guy DEMORTIER from those of Si, Mg, Cl, Ca (for p0) and Al, P, Cl, K, Mg (for p1). he complete discussion is provided elsewhere. Sulphur and silicon (always introduced in metals when heating ores in ancient metallurgy) are of crucial importance for the understanding of ancient technologies. Sulphur (in blende and greenockite) and silicon (in a number of minerals) may be expected to be present in narrow parts of solders made during the brazing of ancient gold artefacts. he detected concentrations of S and Si were in the range of 50 to 120 ppm. he concentration of Cd was simultaneously determined by the detection of characteristic X-rays induced by the deuteron beam. his presence of residual sulphur (only at trace level) indicates that at least 95% of S was lost during the alloying procedure. he presence of traces of Si and often also traces of Al can be also attributed to dust inclusions in narrow regions of solders (Demortier and Gilson, 1987). Cadmium ores (and mainly CdS) always contain zinc as impurities. Modern brazing alloys containing Cd are Zn free or have a Zn concentration of about 20% of the Cd concentration. he measurement of the Zn/Cd ratio would therefore be of great interest for diferentiating brazing alloys containing Cd of ancient or modern origin. As the measurement of Zn in gold rich alloys is impossible by PIXE due to Figure 6: he experimental setup for the measurement of Zn in gold-rich samples using XRF induced by PIXE (Ge primary target). Figure 6 : Le montage expérimental pour la mesure du Zn dans les alliages riches en or par FX induite par PIXE (cible primaire en Ge). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 29-38 Targeting Ion Beam Analysis techniques for gold artefacts the superposition of the K lines of Zn with the L lines of Au, we have developed a technique involving XRF induced by PIXE in order to selectively excite the K lines of Zn without exciting the L lines of Au (Demortier, 1987). he irradiation of a pure Ge target with protons (and not with electrons, in order to avoid any contribution of Bremtrahlung) produces X-rays of 9.89 and 10.98 keV, which are suicient to extract K-shell electrons from Zn but not the L-shell of gold (see Figure 3). he same arrangement, using As as primary target, has been used by the team of M. Guerra to detect traces of Pt in gold (Guerra and Calligaro, 2003; 2004). he experimental assembly is shown in Figure 6. Our personal view on the possible use of Cd-Cu-Ag-Au alloys in antiquity has been extensively discussed, but our conclusions are far from being generally accepted (Demortier, 1987; 1989; 1992; Meeks and Craddock, 1991). 6. CONCLUSIONS Of the diferent ion beam analysis techniques (PIXE, RBS, PIGE and NRA), PIXE is the most appropriate technique using low energy accelerators for the analysis of gold jewellery artefacts. A careful choice of the characteristic X-ray lines may help one avoid interferences in the identiication of signals of selected elements, such as Zn or Pt. XRF induced by PIXE ofers additional possibilities of interferencefree analyses. Irradiation at various incident proton energies ofers the possibility to detect and to quantitatively and nondestructively proile major elements up to 10 microns below the surface. he choice of an incident proton energy around 2 MeV ofers the advantage of a sharp variation of the X-ray emission cross, useful in order to easily distinguish a layered structure from one with a homogeneous distribution of the elements. References BUGOI, R., COJOCARU, V., CONSTANTINESCU, B., CALLIGARO, T., PICHON, L., RÖHRS, St. and SALOMON, J., 2008. Compositional studies on Transylvanian gold nuggets: advantages and limitations of PIXE-PIGE analysis. Nuclear Instruments and Methods B 266: 2316-2319. CALVO DEL CASTILLO , H., RUVALCABA , J.L., CALDERON , T., SALINAS NOLASCO, M.F., MEJIA, L.M. and PERDIGON, K., 2008. Characterisation of an enamelled metallic object found in Guerrero Negro (Baja California) by PIXE and RBS techniques. Nuclear Instruments and Methods B 266: 2311-2315. 37 DECONNINCK, G., 1972. Quantitative analysis by (p,g) and (p,X) reactions at low energies. Journal of Radioanalytical Chemistry 12: 157-169. DEMORTIER, G., 1987. Trace analysis of medium Z elements in narrow regions of a heavy matrix by XRF induced by PIXE, in J.D. Brown, R.H. Packwood (eds.), 11th International Congress on X-ray optics and microanalysis. London: University of Western Ontario, 147. DEMORTIER, G., 1987. La chrysocolle des orfèvres est-elle jaune? Archaeometry 29(2): 275-288. DEMORTIER, G., 1989. Ancient gold solders: what was chrysocolla?, in R.O. Allen (ed.), Archaeological Chemistry IV, Advances in Chemistry Series 220. Washington, DC: American Chemical Society: 249-263. DEMORTIER, G., 1992. Comments on Meeks and Craddock’s position on cadmium in solders. Archaeometry 34(2): 305-319. DEMORTIER, G. and GILSON, A., 1987. Determination of traces of light elements in gold artefacts using nuclear reactions. Nuclear Instruments and Methods B 18: 286-290. DEMORTIER, G. and MORCIAUX, Y., 1994. PIXE gadgets. Nuclear Instruments and Methods B 85: 112-117. DEMORTIER, G. and RUVALCABA, J.L., 2005. Quantitative ion beam analysis of complex gold based artefacts. Nuclear Instruments and Methods B 239: 1-15. DEMORTIER, G., MATHOT, S. and VAN OYSTAEYEN, B., 1990. Complementarity of RBS,PIGE and PIXE for the determination of surface layers of thicknesses up to 25 microns. Nuclear Instruments and Methods B 49: 46-51. DEMORTIER, G., MATHOT, S. and STEUKERS, C., 1993. Secondary efects in PIXE analysis of binary alloys and thick surface layers. Nuclear Instruments and Methods B 75: 347-354. GUERRA, M.F. and CALLIGARO, T., 2003. Gold cultural heritage objects: a review of studies of provenance and manufacturing technology. Measurement Science and Technology 14: 15271537. GUERRA, M.F. and CALLIGARO, T., 2004. Gold traces to trace gold. Journal of Archaeological Science 31( 9): 1199-1208. MATHIS, F., VTIELYNCK, O., LACLAVETINE, K., CHÊNE, G. and STRIVAY, D., 2008. Study of the provenance of Belgian Merovingian garnets by PIXE at IPNAS cyclotron. Nuclear Instruments and Methods B 266: 2348-2352. MEEKS, N.D. and CRADDOCK, P.T., 1991. he detection of cadmium in gold/silver alloys and its alleged occurrence in ancient gold solders. Archaeometry 33(1): 95-107. MIGLIORI, M., GRASSI, N. and MANDO, P., 2008. Scanning PIXE analysis of gold lace embroideries in a relic of St. Francis. Nuclear Instruments and Methods B 266: 2339-2342. QUARTA, G., BUTALAG, K., CALCAGNILE, L., D’ELIA, M., ARTHUR, P., TINELLI, M. and CARAMIA, A., 2008. IBA analyses and lead contamination measurements of AMS-14C dated bones from ArcheoSciences, revue d’archéométrie, 33, 2009, p. 29-38 38 two medieval sites in Italy. Nuclear Instruments and Methods B 266: 2343-2347. RUVALCABA, J.L. and DEMORTIER, G., 1996. Elemental concentration proile in ancient artefacts by ion beam scattering. Nuclear Instruments and Methods B 113: 275-278. SMIT, Z., ISTENIC, J. and KNIFIC, T., 2008. Plating of archaeological objects: study by diferential PIXE. Nuclear Instruments and Methods B 266: 2328-2333. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 29-38 Guy DEMORTIER VAN OYSTAEYEN, B. and DEMORTIER, G., 1983. Matrix efects in PIXE evaluation of the major components in thick homogeneous samples. Nuclear Instruments and Methods 215: 299-313. Synchrotron Radiation Induced X-Ray Fluorescence for the characterization of ancient gold objects Fluorescence de rayons X induite par radiation de synchrotron pour la caractérisation d’objets anciens en or Martin Radtke*, Uwe Reinholz* and Heinrich Riesemeier* Abstract: Due to their non-destructive character, X-ray analytical methods are widespread in the ield of archaeometry. In the present study, the use of synchrotron radiation induced X-ray luorescence analysis (SR-XRF) is illustrated on three examples: a silver point drawing of Jan van Eyck, the Hiddensee Gold Treasure, and the Sky Disc of Nebra. he possibility to analyse large objects in normal atmosphere results in information regarding their origin, uniformity, and the manufacturing process. Résumé : Grâce à leur non-destructivité, les techniques analytiques qui utilisent les rayons X sont très répandues dans le domaine de l’archéométrie. L’utilisation de la luorescence de rayons X induite par radiation de synchrotron (SR-XRF) est illustrée à partir de trois exemples : les dessins à pointe d’argent de Jan van Eyck, le trésor de Hiddensee et le disque céleste de Nebra. L’analyse d’objets, même de grandes tailles, à l’air permet d’obtenir des informations sur leur origine, leur homogénéité et leurs techniques de fabrication. Keywords: Gold, Hiddensee Treasure, Nebra Sky Disc, non-destructive testing, silver point drawings, synchrotron, X-ray luorescence. Mots-clés : Or, trésor d’Hiddensee, disque céleste de Nebra, analyse non-destructive, dessins à pointe d’argent, synchrotron, luorescence à rayons X. 1. INTRODUCTION Synchrotron radiation (SR) is one of the most versatile tools for the characterization of materials. It combines high spatial resolution with good detection limits, the possibility to work in normal atmosphere with the capacity to examine large objects, which do not it into sample chambers. SR measurements on metals are non-destructive. Technically, a synchrotron is a machine which generates brilliant beams of light by moving electrons through a strong magnetic ield. his light, called synchrotron radiation, ranges from the infrared to hard X-rays, with some extremely useful properties. Additionally, a synchrotron provides the infrastructure for a number of diferent measuring techniques, and the facilities for sample treatment and preparation. Last but not least, synchrotrons are located in government funded science centre, where measurements can be carried out for free, provided that a beam time proposal was evaluated successful by an independent beam time committee. One of the disadvantages is that, even if there are now around 40 synchrotron light sources worldwide and the number is still increasing signiicantly, not every country has a synchrotron. Due to limitations related to moving precious objects and their crossing national boundaries, access * BAM Federal Institute for Materials Research and Testing – Richard-Willstätter-Straße 11, 12489 Berlin, Germany. (martin.radtke@bam.de), (uwe. reinholz@bam.de), (heinrich.riesemeier@bam.de) rec. Oct. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 39-44 40 to this methodology is limited. In any case, because of its high lux, SR-based techniques can usually be performed on minute samples. 2. METHODS SR is useful for all X-ray techniques; however, for the analysis of ancient metal objects, X-ray luorescence is the predominant method. For XRF, the use of SR presents especially the following advantages: he radiation is linearly polarized, and therefore the scattering in the plane of polarization is minimized and detection limits are improved. he excitation energy can be selected and adapted to the problem. It is possible to excite an element of interest just above the absorption edge in order to optimize excitation, or to go under an absorption edge to avoid excitation. SR has high brilliance. his means that the photons are emitted in a small solid angle, and this allows focusing the radiation efectively in a small excitation spot. All examples referred to in this publication have been selected from measurements carried out with the SR-XRF setup at the BAMline at BESSY II, Berlin. At this beam line, a super conducting wavelength shifter with a maximum ield of 7 Tesla is the X-ray source, with a useable energy range from 5 keV up to 80 keV. herefore, nearly all elements can be detected by measurement of their K-shell luorescence. A W/Si double-multilayer-monochromator (DMM) and/or a Si [111] double-crystal-monochromator (DCM) are used to select the desired excitation energy. In order to achieve higher luxes, it is possible to focus the beam to a size of approximately (300x300) μm² by bending one multilayer and/or one crystal (Görner et al., 2006). If smaller beam sizes are necessary, compound refractive lenses (CRL) can be used to achieve spot sizes of less than (2x2) μm² (Strub et al., 2008). For the detection of the luorescence signal, different types of solid state detectors (Si(Li) HPGe and SDD) are available. For data evaluation, various program packages are freely available. At the BAMline, QXAS from the IAEA, a Monte Carlo code, capable of simulating photon-matter interaction (Vincze et al., 1995) and pyMCA (Sole et al., 2007) are implemented. Details about the SR-XRF technique can be found elsewhere (Adams et al., 1998). SR-XRF is a non-destructive multi-element method. However, not all elements can be analyzed with the same sensibility. For gold, an excitation energy of around 33-38 keV, depending on the expected trace elements, has been found to be most useful. In this case, the minimum detection limits (MDL) of trace elements in gold are in the ArcheoSciences, revue d’archéométrie, 33, 2009, p. 39-44 Martin RADTKE, Uwe REINHOLZ, Heinrich RIESEMEIER order of 1 ppm. he detection of platinum requires special procedures because of the closeness of these two elements in the periodic table. By a very sophisticated approach (Guerra et al., 2005; 2008), the MDL of Pt can be decreased as low as 10-20 ppm. 3. EXAMPLES If an archaeologist or art historian comes to use a synchrotron, typically one of these questions is asked: – What material was used to manufacture this object? – How was it made? – Is it possible to discriminate diferent compositions within one object or between diferent members of a group of objects? Examples of questions that are more diicult to answer are the following: – Where does it come from? – When was it made? – Is the object authentic or a forgery? Unfortunately, there is no possibility to measure the age of a gold object with SR. herefore, the last two questions cannot be answered with our methods in a straightforward manner. In the following, examples of answers to the other categories of questions are presented. At irst, the apparently trivial task of the identiication of the material is illustrated by the non-destructive analysis of silver point drawings of Renaissance artists. For this purpose, extremely sensitive synchrotron – or accelerator-based techniques (PIXE – Particle Induced X-Ray Analysis) are necessary, because only small quantities of metal were deposited on the paper (Reiche et al., 2004a). On Van Eyck’s drawing, the ‘Portrait of Cardinal Albergati’, at least three diferent metal points have been identiied (Fig. 1). Two diferent types of silver marks have been found in the examined portrait. Silver containing gold marks has been detected in the inscriptions and over-subscriptions. his discovery has represented the irst evidence of the use of gold points in metal point drawings from the Middle Ages. he fact that a gold stylus diferent from the other silver points was used represents therefore an important result. An explanation of the exceptional use of a gold stylus could be the diference in mechanical properties of a gold point, which is much smoother than pure or especially Cu-containing silverpoints. hus, it seems easier to write with a gold point than with a silver one. Furthermore, it is possible to obtain diferent colour aspects of the marks when using another metal than silver. A gold stylus leaves a dark grey mark, whereas a silver mark becomes brownish quite rapidly. hus, it can be postulated that Van Eyck con- Synchrotron Radiation Induced X-Ray Fluorescence for the characterization of ancient gold objects 41 Figure 1: he drawing “Portrait of Cardinal Albergati” by Jan van Eyck (Kupferstichkabinett, Dresden, Germany, inv. no. C775). he plot shows a comparison of synchrotron-induced X-ray luorescence spectra of a silver mark (stylus 1), a gold mark (stylus 2), and the paper with preparation backing measured with an excitation energy of 30 keV. Figure 1 : Le dessin Portrait du Cardinal Albergati par Jan van Eyck (Kupferstichkabinett, Dresde, Allemagne, inv. no. C775). Le graphique montre la comparaison des spectres obtenus par luorescence de rayons X induite par radiation de synchrotron de marques en argent (stylus 1) et en or (stylus 2) et du papier avec la préparation d’entoilage mesurés avec une énergie d’excitation de 30 keV. sciously used diverse metal points in order to make the most of the changing properties of diferent metal styli, and to achieve special graphic efects, such as the particular threedimensionality or profundity of this portrait. It is thus possible to draw a conclusion regarding the genesis of the drawing. Van Eyck probably used irst the pure silver stylus for drawing the face, the robe, and the hair of the igure. he cross line was also made with this stylus. he hatchings on the left and on the right of the head were made with another copper-containing silver point, which emphasizes the plasticity of the portrait. hen, the eyes and other parts of the portrait were intensiied using the gold stylus. With the same stylus, the letters on the face were inscribed (Reiche et al., 2004b). he next example provides an insight into the fabrication techniques of Viking gold objects. he so-called ‘Hiddensee Gold’ objects (Fig. 2) are said to have been washed ashore after two great storm tides in 1872 and 1874, and saved by ishermen from Neuendorf, Germany. From there, they came to their present whereabouts, the Cultural and Historical Museum of Stralsund, under somewhat adventurous circumstances. he 16 objects, with a total weight of 600 g, have been manufactured around 1000 AD in a Viking workshop. he irst results showed a great homogeneity of the diferent gold alloy of all sixteen objects, except for the brooch, the neck ring, and one of the cross-shaped pendants (no. 12). his inding was not surprising for the neck ring, which was processed with the help of a diferent technique (twisted wire instead of soldered sheet metal, wire, and granules), and not even for the brooch, which is a unique piece among the iligree decorated objects. However, there is no answer yet to the question of why the crossshaped pendant no. 12 has a signiicantly lower gold content than the other nine cross-shaped pendants. Perhaps it was produced separately and later added to the set? Another important result concerns the composition of the solder. Remnants of metallic solder distributed on the sheets could clearly be seen under the microscope. For the solder, gold had been alloyed with other types of metal so that the melting point was signiicantly lower than that of the gold alloy of sheets and decorations. Upon heating, the sheets and decorations were connected by the solder without melting themselves. Normally, the melting point of pure gold is 1084 °C. By adding silver or copper to the gold, the melting point can be reduced to a temperature between 850 °C and 975 °C, depending on the exact proportions of the diferent metals. Our investigations of pendant no. 13, for example, conirm the presence of a high amount of copper in the solder (Armbruster et al., 2004). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 39-44 42 Martin RADTKE, Uwe REINHOLZ, Heinrich RIESEMEIER b a he provenance of gold objects based on elemental analysis is problematic, due to the frequent recycling of this precious material. However, for very old objects, it can be assumed that no recycling took place, and that the ingerprint of trace elements in the gold object is similar to that of the original gold. he third example presented in this publication, the ‘Sky Disc of Nebra’ (Fig. 3), was manufactured around 2000 BC and is featuring the earliest record of Central European astronomy known to date. he question that needed to be answered with SR-XRF was whether the composition of the diferent gold parts was the same, or if it had changed during diferent periods of completion according to the purpose of the disc. he analysis of minor components such as silver and tin provided the basis for a cluster analysis resulting in the separation of three distinctly diferent groups: 1. sun, moon and the majority of the stars; 2. horizon and one single-shifted star; 3. the so-called bark. Using these results, we were able to support the assumption of the archaeologists, that the disc was manufactured ArcheoSciences, revue d’archéométrie, 33, 2009, p. 39-44 Figure 2: (See colour plate) The Hiddensee Gold treasure (Kulturhistorisches Museum Stralsund, Germany, inv. no. 1873: 499a et cetera). he scatter plot (a) shows the diferent compositions of the brooch, the neck-ring, and pendant no. 12. In the plot (b), the increased copper content of the solder can be seen. Figure 2 : (Voir planche couleur) Le trésor de Hiddensee (Kulturhistorisches Museum Stralsund, Allemagne, inv n° 1873 : 499a). Le graphique ci-contre (a) montre les diférentes compositions de la broche, du ras de cou et du pendentif n° 12. Le graphique ci-dessus (b) met en évidence l’augmentation de la teneur en cuivre pour la soudure. in several phases, representing progress in knowledge and a change of religion and purpose. Additionally, we were able to identify the Carpathian basin as the most likely source of the gold. Using copper from Austria, and a ‘technology transfer’ from Mycenae (Greece), the manufacturing of the Nebra Sky Disk can be considered as a very early endeavour based on a joint European experience, possibly with contributions indebted to the symbolism of Egypt or the Levant (Pernicka et al., 2003). 4. SUMMARY he special requirements for the use of a synchrotron and the advantages of synchrotron radiation have been discussed above. SR-XRF made it possible to characterize small amounts of metal, as in the case of the silverpoint drawings, analyse big objects such as the Sky Disc of Nebra, and to measure the composition of objects on a small scale, as in the case of the solder used for the Hiddensee gold. It has been shown that knowledge about the composition of the material can provide signiicant information about the origin and insights into the manufacturing process of an object. While the examples presented above are covering the main topics related to the analysis of ancient Synchrotron Radiation Induced X-Ray Fluorescence for the characterization of ancient gold objects 43 Figure 3: (See colour plate) he Sky Disk of Nebra (Landesmuseum für Vorgeschichte, Halle, Germany, inv. no. HK2002:1649a). he measured silver and tin contents of the gold applications are shown in the scatter plot. he three groups are clearly recognisable: (a) bark; (b) star no. 23 and horizon; (c) the remaining stars with moon and sun. Figure 3 : (Voir planche couleur) Le disque céleste de Nebra (Landesmuseum für Vorgeschichte, Halle, Allemagne, inv. no. HK2002 : 1649a). Les teneurs en argent et en étain mesurées pour les applications en or sont indiquées sur le graphique. Les trois groupes sont clairement reconnaissables : (a) écorce ; (b) étoile n° 23 et horizon ; (c) les autres étoiles avec la lune et le soleil. gold with SR-XRF, it should be mentioned that there are also other techniques able to ofer useful information. For example, X-Ray Difraction (XRD) was used to investigate the weathering of gilding decorations (Lluveras et al., 2008). X-ray Absorption Spectroscopy (XANES and EXAFS) and Small Angle X-ray Scattering (SAXS) are useful tools for characterizing nano particles in gold ruby glass (Eichelbaum et al., 2007). In the future, the use of SR-XRF to characterize the Berlin Gold Hat (Museum für Vor- und Frühgeschichte, Staatliche Museen zu Berlin, Germany) and the Skythian Gold Treasures from Vettersfelde and Maikop (Antikensammlung, Staatliche Museen zu Berlin, Germany) will bring exciting new insights into the past. Acknowledgements he authors would like to thank our colleagues I. Reiche, M.F. Guerra, B. Armbruster, H. Eilbracht, A. Grüger, C.H. Wunderlich and E. Pernicka for the good collaboration. A special compliment for all the art historians and archaeologists involved, for the conidence to bring these unique objects to our beam line. he BESSY team is acknowledged for their good cooperation. References ADAMS, F., JANSSENS, K. and SNIGIREV, A., 1998. Microscopic X-ray luorescence analysis and related methods with laboratory and synchrotron radiation sources – Plenary lecture. Journal of Analytical Atomic Spectrometry 13(5): 319-331. ARMBRUSTER, B., EILBRACHT, H., GRÜGER, A., RADTKE, M., RIESEMEIER, H. and REICHE, I., 2004. he Vikings in Berlin: SR-XRF analyses of the Hiddensee gold jewellery. BESSY Scientiic Highlights 2004: 32-33. EICHELBAUM, M., RADEMANN, K., WEIGEL, W., LOCHEL, B., RADTKE, M. and MULLER, R., 2007. Gold-ruby glass in a new light: On the microstructuring of optical glasses with synchrotron radiation. Gold Bulletin 40(4): 278-282. GÖRNER, W., EICHELBAUM, M., MATSCHAT, R., RADEMANN, K., RADTKE, M., REINHOLZ, U. and RIESEMEIER, H., 2006. Nondestructive investigation of composition, chemical properties and structure of materials by synchrotron radiation. Insight 48(9): 540-544. GUERRA, M.F., CALLIGARO, T., RADTKE, M., REICHE, I. and RIESEMEIER, H., 2005. Fingerprinting ancient gold by measuring Pt with spatially resolved high energy Sy-XRF. Nuclear Instruments and Methods B 240(1-2): 505-511. GUERRA, M.F., RADTKE, M., REICHE, I., RIESEMEIER, H. and STRUB, E., 2008. Analysis of trace elements in gold alloys by SR-XRF at high energy at the BAMline. Nuclear Instruments and Methods B 266(10): 2334-2338. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 39-44 44 LLUVERAS, A., BOULARAND, S., ROQUE, J., COTTE, M., GIRALDEZ, P. and VENDRELL-SAZ, M., 2008. Weathering of gilding decorations investigated by SR: Development and distribution of calcium oxalates in the case of Sant Benet de Bages (Barcelona, Spain). Applied Physics A: Materials Science and Processing 90(1): 23-33. PERNICKA, E., RADTKE, M., RIESEMEIER, H. and WUNDERLICH, C-H., 2003. European Network of Competence at 1600 BC. BESSY Scientiic Highlights 2003: 8-9. REICHE, I., BERGER, A., GORNER, W., MERCHEL, S., RADTKE, M., RIEDERER, J., RIESEMEIER, H. and ROTH, M., 2004a. Following the traces of Albrecht Dürer: analysis of silverpoint drawings by spatially resolved synchrotron-induced X-ray luorescence analysis. Nuclear Instruments and Methods B 226(1-2): 83-91. REICHE, I., RADTKE, M., BERGER, A., GORNER, W., KETELSEN, T., MERCHEL, S., RIEDERER, J., RIESEMEIER, H. and ROTH, M., 2004B. Spatially resolved synchrotron-induced X-ray luorescence analyses of metal point drawings and their mysterious ArcheoSciences, revue d’archéométrie, 33, 2009, p. 39-44 Martin RADTKE, Uwe REINHOLZ, Heinrich RIESEMEIER inscriptions. Spectrochimica Acta B: Atomic Spectroscopy 59(1011): 1657-1662. SOLE, V.A., PAPILLON, E., COTTE, M., WALTER, P. and SUSINI, J., 2007. A multiplatform code for the analysis of energy-dispersive X-ray luorescence spectra. Spectrochimica Acta B: Atomic Spectroscopy 62(1): 63-68. STRUB, E., RADTKE, M., REINHOLZ, U., RIESEMEIER, H. and REMIKOVA, E., 2008. Measurements with compound refractive lenses at the “BAMline”. Nuclear Instruments and Methods B 266(10): 2165-2168. VINCZE, L., JANSSENS, K., ADAMS, F. and JONES, K.W., 1995. A general Monte Carlo simulation of energy-dispersive X-ray luorescence spectrometers. Part 3. Polarized polychromatic radiation, homogeneous samples. Spectrochimica Acta B: Atomic Spectroscopy 50(12): 1481-1500. X-ray Diffraction method for determination of crystallite sizes of gold and silver items – New opportunities for archaeology and for protection against forgery Méthode de diﬀraction de rayons X pour la détermination des tailles de cristallites des objets d’or et d’argent – Nouvelles opportunités pour l’archéologie et pour la protection contre la production de faux Denka Yanakieva*, Milena Tonkova**, Ernst Spiridonov***, Zlatil Vergilov**** and Petia Penkova** Abstract: This work presents the results of X-ray studies of the nanostructure of gold and silver items, using the obtained data for purposes of identification and comparison. The study covers gold and silver archaeological items, as well as gold and silver nuggets (both exposed and not to thermal and mechanical actions). The items were studied by means of XRD analysis and emphasis was placed on line-profile analysis, aimed at obtaining information about the microstructural properties of materials: size of coherently diffracting domains in crystals (crystallite size). Crystallite sizes, as a constant numerical feature of an item, can be used successfully for: ascertaining an item’s authenticity and thus providing protection against forging; studying issues relevant to the treatment of forgeable metals in the ancient ages, when man started using them; providing exact time and place correlations (included in complex researches of archaeological items and combined with other methods), etc. The present work is part of a detailed X-ray study of gold and silver, aimed at specifying the nanostructure characteristics of items: crystallite sizes, micro-strains at unit cell level, unit cell parameters. This work is based on crystallite size characteristics and offers the first examples demonstrating the potential of the method, while also suggesting possible future actions that would further evaluate this potential. Résumé : Ce travail présente les résultats d’une étude des rayons-X de la nanostructure d’objets en or et en argent ; les données obtenues sont utilisées pour l’identification et la comparaison. Cette étude concerne des objets archéologiques en or et en argent ainsi que des pépites d’or et argent (n’ayant subi aucune action thermique ou mécanique). Les objets ont été étudiés au moyen de l’analyse DRX, l’accent ayant été mis sur l’analyse des profils de lignes, de façon à obtenir des informations sur les propriétés microstructurales : taille des domaines diffractant de façon cohérente (taille des cristaux). La taille des cristallites, caractéristique numérique constante d’un objet, peut être utilisée avec succès pour : vérifier l’authenticité d’un objet et ainsi fournir protection contre la contrefaçon ; étudier des questions pertinentes liées au traitement de métaux utilisés par les faussaires dans le passé, aux débuts de leur utilisation par l’homme ; fournir des corrélations exactes de temps et localisation (inclues dans des recherches complexes sur les objets archéologiques et combinés à d’autres méthodes), etc. * Earth and Man National Museum – 4 Cherni Vrach Blvd., Sofia 1421, Bulgaria. (denkay@abv.bg) ** Institute of Archaeology with Museum, Bulgarian Academy of Sciences – 2 Saborna Street, Sofia 1000, (Bulgaria. milenatonkova@hotmail.com) (petiapenkova@yahoo.com) *** Department of Mineralogy, Lomonosov State University of Moscow – Vorobiovy Gory, 1192334, Russia. (mineral@geol.msu.ru) **** St. Kliment Ohridski Sofia University, Department of Archaeology, Petrology and Mineral Resources – 15 Tsar Osvoboditel Blvd., Sofia 1504. (zver@gea.uni-sofia.bg) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 45-50 46 Denka YANAKIEVA et al. Ce travail fait partie d’une étude détaillée des rayons X de l’or et de l’argent, ayant pour but de spécifier les caractéristiques nano-structurelles des objets : tailles de cristallites, micro-tensions au niveau de cellules d’unité, paramètres de la cellule d’unité. Ce travail se base sur les caractéristiques cristallites et offre les premiers exemples démontrant le potentiel de la méthode ainsi que suggérant des actions futures possibles permettant d’évaluer ce potentiel. Keywords: crystallite sizes, gold, line profile analysis, peak broadening, silver, X-ray diffraction. Mots-clés : tailles des cristallites, or, analyse de lignes de profil, élargissement de pic, argent, diffraction de rayons X. 1. INTRODUCTION X-ray diffraction is the first exact method for the study of the structure of matter. In archaeology, this analysis is used successfully to study stone constructions, ceramics, pigments, etc. The present study of gold and silver items outlines the first steps in a new field – analyzing the line profile of objects in order to obtain information about the microstructural properties of materials: sizes of coherently diffracting domains in crystals (crystallite sizes) and micro-strains in the lattice. These parameters depend on the forming conditions of the material and on various influences (mechanical and thermal) the material was exposed to. The present work is aimed at determining: the sizes of crystallites making up gold and silver objects of natural and artificial origin; the variability of crystallite sizes (D) due to the origin of the material and technological processing of the object; the applicability of such an approach to the characterization and study of archaeological artefacts. The main objects of the study are Thracian gold and silver jewellery items, dated to the first millennium BC, and belonging to the collection of the National Institute of Archaeology with Museum of the Bulgarian Academy of Sciences. Gold and silver objects belonging to different ages and regions were also studied for purposes of comparison. Natural gold and silver samples, both exposed and not to thermal and mechanic influences, were studied as well. The studied alloys belong to the systems Au-Ag (Ag-Au) and Ag-Au-Cu (Au-Ag-Cu). They represent cubic solid solutions with no decay and decomposition. Every item was submitted to 2÷8 XRD analyses. A total of 49 items were submitted to 136 XRD analyses. ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? 2. METHODOLOGY Theoretical Basis – crystallite sizes and their relation with profile line broadening Every grain of the microstructure of a crystal in an object is, in fact, a mosaic of small blocks showing angular discordance ranging from 20’ to 1-5°. The dimensions of these blocks are comparable with those of the so-called Areas of Coherent Scattering (ACS). That is why small angular discordances cause a widening of the diffraction angle (θ) with some extra value (±Δα) (Howard and Preston, 1989). This effect is observed as a broadening of the diffraction peak width. Crystallite size determination entails a study of the nanostructure of the object. For comparison: size of Au atom = 0.142 nm; size of Ag atom = 0.143 nm; Au unit cell parameter a = 0.4078 nm; Ag unit cell parameter a = 0.4086 nm (Strunz 2001); crystallites measured in the present work: D(Au) = 11.2 ÷ 31.8 nm, D(Ag) = 8.7 ÷ 28.1 nm. Apart from crystallite sizes, peak broadening is also influenced by the instrumental broadening caused by the equipment, and by micro-strains (Balzar, 1993). The instrumental broadening measured in this case was 0.05 (FMHM ~ 30° 2θ, CuKα), which is in line with the values quoted in the literature for this type of equipment (Balzar 1992). In the present research, the authors’ interest was focused exclusively on studying crystallite sizes. Measurements X-ray diffraction data were collected by means of a TUR M 62 diffractometer (Germany) with a standard two-circle goniometer in Bragg-Brentano geometry with secondary graphite monochromator. The surface of the studied object was perpendicular to the goniometer plane. We used CuKα radiation (λ = 0.15418 nm) under the following measu- X-ray Diffraction method for determination of crystallite sizes of gold and silver items… rement conditions: tube voltage = 32 kV; tube current = 15 mA; step-scan mode with step size = 0.02° 2θ; and counting time = 2 s per step for standard researches for phase analysis, and 0.01°/15-30 s for researches in the interval 35-41° 2θ for experiments performed in order to study the broadening. The gold items’ indexed diffractogram in this mode is shown in Figure 1. The instrumental broadening was determined through standard Si and La B6 powder. 47 Balzar and Popovic 1996; Balzar et al., 2004). The profile fitting of the 111 reflection of an Au item is shown in Figure 2 as an example. 3. RESULTS BASE GROUP – Thracian jewellery from the Collection of the National Institute of Archaeology with Museum. Data Analysis The diffractometer was controlled by a computer and all measurements were stored on the hard disk. Data were transferred to a personal computer for processing. We used WinFit (version 1.2.1-1997) freeware (Krumm, 1994) to fit profiles and calculate crystallite sizes. WinFit calculated crystallite sizes according to the Warrren-Averbach method. The Pearson VII function was used to fit profiles (Mittemeijer and Scardi, 2004; Uvarov and Popov, 2007). Profile Analysis The diffraction peak width (DPW) is sensitive to the mean value of crystallite thickness (size) in a direction normal to the diffracting crystal planes. DPW increases when crystallites diminish and conversely becomes narrower when the nanostructure is coarser. The crystallite sizes were determined by the one-order (single line) method. The profile of the reflex with indexes 111 at Bragg angle 2θ ~ 38.3°, corrected for instrumental broadening, was used for the determination of crystallite sizes (Scardi et al., 1994; 1st group of objects: 2nd century BC – 2nd century AD: The four silver bracelets from Rouzhintsi (Belogradchik region), Nos. 2858, 2859, 2860 and 2861, are decorated with snakes’ heads. The rings of the bracelets are cast. Nos. 2858 and 2859 have almost identical D (16.4 nm; 16.7 nm), i.e., they can be considered as a pair. The other two samples, Nos. 2860 and 2861, possess similar crystallite characteristics, with some insignificant diversion of D (D = 16.0 nm; 16.8 nm), which makes us accept them as products of the same atelier. The D values measured in the areas of the snakes’ heads diminish, an aspect which can be interpreted as a result of forging (D = 15.4 nm). A different technology was used for modelling the snakes’ heads. The silver torque from Zelenigrad (No. 2996) is formed in a similar manner as the four bracelets from Rouzhintsi – each end is shaped as an onion with a small granule. Archaeologists consider them as synchronous, dating to a period between the 2nd century BC and the 2nd century AD. The D (10.54 nm) of the torque, if compared to those of the four bracelets, appears to be quite different, which suggests a different origin and/or technology. Figure 1: Indexed diffractogram of gold items. Figure 1 : Diffractogramme indexé des objects d’or. ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? 48 Denka YANAKIEVA et al. Figure 2: Fitted diffraction line of studied peak 111. Figure 2 : Ligne de diffraction ajustée au pic 111. 2nd group of objects – 2nd half of the 4th century BC, items from western Bulgaria: The pair of fibulae from Penkovtsi (Tran region), Nos. 3015a and 3015b, have exactly identical D (15.9 nm) as a result of one and the same technology being employed in their production. One of the fibulae is restored, but this did not influence its D. Another pair of fibulae from Garbino (Kyustendil region), Nos. 3008a and 3008b, is of a similar morphological type as the above-mentioned ones. The fibulae of this pair are characterized by equal D (17.5 nm), differing however from the value of the Penkovtsi fibulae. These pairs are obviously products of different technologies and presumably different ateliers. A single silver fibula (No. 630) of unknown origin was analyzed in seven spots of the as-cast part and in one of the forged zone. All seven spot analyses show an exactly equal D (22.7 nm), while a single forged area shows some small difference (D = 21.7 nm). The D value of this fibula differs drastically from the above-mentioned two pairs, but it is quite similar to the D of the following objects. A pair of silver bracelets from Granitovo (Belogradchik region), Nos. 4036a and 4036b, show absolutely equal D (20.0 nm) (composition: Ag – 97.5%; Cu – 2.5% for No. 4036a, and Ag – 99%, Au – 1% for No. 4036b) to the abovementioned item, which may be a sign of a common origin. The analyses of the cast part and forged tip of the fragmented silver torque from Stolat (Sevlievo region), No. 3013, show similar sizes of crystals (D = 19.1 nm; 18.5 nm). ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? 3rd group of objects – 5 fragments of gold earrings, Krun (Bulgaria), No. 2682: All measured D values are identical (D = 17.8 nm), which confirms their common origin. Objects for Comparison of Crystallite Sizes First Group – Gold and Silver Nuggets: 1. Gold Nuggets a) Nugget – Samples of vein gold from the following regions were studied: Etropole, Bulgaria (D = 22.7 nm), Central Rhodope Mountains, Bulgaria (D = 19.8 nm), as well from the large deposits of Bestube (D = 20.0 nm), Aksu (D = 26.0 nm), and Zhana-Tyube (D = 31.8 nm) in Kazakhstan. Alluvial gold from the Rhodope Mountains rivers (D = 23.6 nm) and the Iskar River (D = 28.8 nm) was also studied. b) Nugget, exposed to thermal and mechanical influences. Samples from Bestube (northern Kazakhstan) were exposed to heating under and over the melting point (D = 19.8 nm and 16.2 nm, respectively), and to mechanical actions (forging) (D = 19.8 nm). 2. Silver Nuggets a) Nugget – Samples of the following deposits were studied: Chiprovtsi, Bulgaria (D = 28.1 nm), Sassa, Macedonia (D = 22.3 nm), Kongsberg, Norway (D = 22.4 nm), and Dzhezkazgan, Kazakhstan (D =28.0 nm). X-ray Diffraction method for determination of crystallite sizes of gold and silver items… b) Nugget, exposed to thermal and mechanical influences. Samples from Chiprovtsi and Dzhezkazgan were exposed to heating under (D = 28.1nm; D = 28.0 nm), and over (D = 17.7 nm; D = 17.6 nm) the melting point, and to mechanical actions (forging) (D = 28.0 nm; D = 27.7 nm). Second Group – Gold and Silver Artefacts, Differing Sharply in Terms of Chronology and Regions: 1. Gold items – 13th century gold appliqués from Preslav (D = 11.2 nm) and 18th century Turkish gold coins (D = 15.1 nm). 2. Silver items – The following objects were studied: 13th century silver belt appliqués from the regions of Shumen (D = 14.1 nm), Pliska (D = 15.4 nm) and Novi Pazar (D = 14.7 nm), Bulgaria; 18th century tobacco pipe cleaning set (D = 8.7 nm), hair decoration (D = 11.1 nm) and massive fork (D = 10.1 nm), Mongolia. The last item was also studied after a standard restoration process (D = 10.1 nm). Size splitting in groups is shown in Figure 3. 4. RESULTS AND DISCUSSION Multiple measurements of crystallize sizes of a given item (in one and the same point and in different points) always show the same results. Crystallite sizes of Au and Ag nuggets are larger than those of artefacts (gold and silver nuggets are formed at temperatures varying between 160 and 200 °C) (Spiridonov and Pletnev, 2002). Thermal treatment of Au and Ag has the following impacts: heating up to/over melting temperature – crystal- 49 lite sizes decrease sharply (30% D); heating under the melting point – no effect on crystallite sizes. Mechanical treatment (forging) either does not show any size decrease or shows a slight size decrease (0÷5% D). Surface chemical treatment (restoration) has no impact on crystallite sizes. Items from different cultures (regions) differ noticeably in terms of crystallite sizes. The objects with different chronology show differences in terms of crystallite sizes. 5. CONCLUSIONS AND GUIDELINES FOR FURTHER RESEARCH 1. Crystallite sizes, as a constant numerical feature of an item, can be used successfully as a ‘passport’ characteristic of the object. This value can be used confidently to prove an item’s authenticity, and it can also be used with certainty as an absolute protection against forging. While characteristics such as exterior features and chemical composition can be reproduced, imitating the nanostructure of the material is absolutely impossible, since it depends on the thermal history of the item. 2. Crystallite sizes can be used to study issues relevant to the treatment of forgeable metals in the ancient ages when man started using them (the processes of forging, heating under the melting point and melting only). 3. Crystallite sizes can provide information useful for distinguishing the different ateliers (metal workshops), since the thermal treatment of an item is a determinant feature for crystallite sizes. Figure 3 (See color plate): Distribution of crystallite sizes as per groups of studied items. Figure 3 (Voir planche couleur) : Distribution des tailles de crystallites selon les groupes des objets étudiés. ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? 50 4. Using XRD analysis in complex researches of a given archaeological site and particularly in researches allowing wider correlations, made according to data bases, creates opportunities for a more accurate interpretation of the studied archaeological material. 5. We are confident that working together with archaeologists on genuine items and in the framework of specific projects will contribute to both the clarification of issues relevant to cultural heritage and to the development of the method, rendering it more precise. The present work represents only a first stage of the application of the method for studying gold and silver archaeological items and an illustration of its possibilities. Detailed studies of the regularities that the objects in the specified groups show are a subject of another work. We believe that further precise researches of diffractogram profiles, comparing crystallite sizes and measured microstrains, as well as the correlation of the results obtained through different software, will provide new results, to be used successfully within archaeological sciences. Acknowledgments We owe a special debt of gratitude to: Prof. Georgi Kirov (St. Kliment Ohridski Sofia University) for participating in the discussions; Mr. Leonid Pautov (A. E. Fersman Mineralogy Museum – Russian Academy of Science, Moscow), Mr. Hristo Stanchev (Eurotest AD, Sofia) and Mr. Plamen Bonev (National Institute of Archaeology with Museum, Sofia) for precise analyses of the elemental composition of the studied items; Mr. Petar Delchev (Earth and Man National Museum, Sofia) for helping us with the computer processing of the information. The present work was carried out with the financial support of a project (Contract No. D01-1167, dated December 15, 2007) of the National Science Research Fund of the Ministry of Education and Sciences, and Yana & Joeri Buhrer Tavanier. ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? Denka YANAKIEVA et al. References BALZAR, D., 1993. X-ray Diffraction Line Broadening: Modeling and Applications to High-Tc Superconductors. Journal of Research of the NIST 98: 321-353. BALZAR, D., 1992. Profile Fitting of X-ray Diffraction Lines and Fourier Analysis of Broadening. Journal of Applied Crystallography 25: 559-570. BALZAR, D. and POPOVIC, S., 1996. Reliability of the Simplified Integral-Breadth Methods in Diffraction Line-Broadening Analysis. Journal of Applied Crystallography 29: 16-23. BALZAR, D., AUDEBRAND, N., DAYMOND, M.R., FITCH, A., HEWAT, A., LANGFORD, J.I., LE BAIL, A., LOUËR, D., MASSON, O., MCCOWAN, C.N., POPA, N.C., STEPHENS, P.W. and TOBY, B.H., 2004. Size-Strain Line-Broadening Analysis of the Ceria Round-Robin Sample. Journal of Applied Crystallography 37: 911-924. HOWARD, S.A. and PRESTON, K.D., 1989. Profile fitting of powder diffraction patterns, in D.L. Bish, J.E. Post (eds.), Mineralogical Society of America, Modern Powder Diffraction, Reviews in Mineralogy 20: 1-17. KRUMM, S., 1994. WINFIT 1.0 – A computer program for X- ray diffraction line profile analysis. XIII Conference on Clay Mineralogy and Petrology, Acta Universitatis Carolinae, Geologica 38: 253-256. SCARDI, P., LEONI, M. and DELHEZ, R., 1994. Line broadening analysis using integral breadth methods: a critical review. Journal of Applied Crystallography, 27: 345-357. SPIRIDONOV, E. and PLETNEV, P., 2002. Deposit of the cuprian gold Zolotaya Gora. Moscow, Scientific World (in Russian). STRUNZ, H. and NICKEL, E.H., 2001. Strunz Mineralogical Tables. 9th edition. Stuttgart, Schweizerbart. MITTEMEIJER, E.J. and SCARDI, P. , 2004. Applicabilities of the Warren-Averbach analysis and an alternative analysis for separation of size and strain broadening. In Diffraction Analysis of the Microstructure of Materials. Springer Series in Materials Science, Vol. 68, XXV, 550. UVAROV, V. and POPOV, I., 2007. Metrological characterization of X-ray diffraction methods for determination of crystallite size in nano-scale materials. Materials characterization 58: 883-891. A multi-analytical approach for the identification of technological processes in ancient jewellery Approche multi-analytique pour l’identification des procédés technologiques de fabrication de bijoux anciens Daniela Ferro*, Vania Virgili**, Adelia Carraro***, Edilberto Formigli**** and Lorenzo Costantini***** Abstract: he compositional and structural characterization of the constituent alloys of precious metal artefacts is the basis for understanding the technological processes used in the art of the ancient goldsmiths. his research presents a multi-analytical approach to the study of the technologies of ancient jewellery manufacturing through the use of non-traditional methods. he development of original methods, such as the application of hardware devices and software programs, allows us to overcome the application limits of traditional analysis, as in the study of Castellani’s earrings, where statistical analyses were carried out using Pearson’s Coeicient Correlation and Cluster Analysis to identify the correlation of elements in the soldering areas. For the Tarquinia ibula, a methodology based on Imaging Analysis has been used to measure the geometrical parameters involved in the construction of the jewel. For another study, a new experimental apparatus was set up in order to identify markers on the micro/nano scale. It consists of a scanning electron microscope (SEM-EDS) integrated with XRF, an equipment developed in order to obtain a greater depth of analysis and a better integration of the compositional data. he aim of the work has been to integrate data from diferent analytical methodologies in order to identify technology transfer and/or production features, as well as the circulation of the goldsmith’s artefacts. Résumé : La caractérisation de la composition et de la structure des alliages en métal précieux est fondamentale pour la compréhension des procédés technologiques liés à l’orfèvrerie antique. Ce travail décrit un nouveau protocole d’analyse dédié à l’étude des technologies de fabrication de bijoux anciens, grâce à l’utilisation de méthodes non traditionnelles. Le développement de méthodes originales, utilisant entre autres des dispositifs expérimentaux et des logiciels, permet de dépasser les limites d’application de l’analyse traditionnelle. Par exemple, dans l’étude de boucles d’oreilles de la collection Castellani, l’analyse statistique réalisée au moyen du Coeicient de corrélation de Pearson et de l’analyse par nuage de points ont permis d’identiier la corrélation entre les éléments entrant dans la composition de la soudure. Dans le cas de la ibule de Tarquinia, une méthodologie basée sur l’analyse d’images a été utilisée pour mesurer les paramètres géométriques associés à la construction du bijou. Simultanément, un nouvel équipement a été créé pour individualiser des marqueurs à l’échelle micro/nanométrique. Il s’agit du couplage d’un microscope électronique à balayage équipé d’un système de dispersion en énergie X (MEB-EDS) avec la luorescence à rayons X (FX) pour obtenir une profondeur d’analyse plus grande et une meilleure intégration et reproductibilité des données concernant la composition élémentaire. Le but de ce travail a été d’intégrer les données de diférentes méthodes d’analyse ain d’identiier le transfert de technologie et/ou de travail lié à la production et circulation d’objets d’orfèvrerie. Keywords: Ancient technology, Etruscan jewellery, Image analysis, SEM-EDS-XRF system, Micro/nano diagnostic, Statistical analysis Mots-clés : Technologie antique, bijoux étrusques, analyse d’image, MEB-EDS-FX, Micro/nano diagnostique, analyse statistique. * Consiglio Nazionale delle Ricerche – Institute for Nanostructured Materials (CNR-ISMN), c/o Department of Chemistry, Sapienza University of Rome – P.le A. Moro 5, 00185 Rome, Italy. (daniela.ferro@cnr.it) ** Department of Chemistry, Sapienza University of Rome – P.le A. Moro 5, 00185 Rome. *** National Etruscan Museum of Villa Giulia, Piazzale di Villa Giulia 9, 00196 Rome, Italy. **** AN.TE.A – Strada Comunale della Miniera 4, 53016 Murlo (SI), Italy. ***** Oriental Art National Museum – via Merulana 248, 00185 Rome, Italy. rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 51-57 52 Daniela FERRO et al. 1. INTRODUCTION 2. METHODS A deeper understanding of our cultural heritage is made possible through the contributions ofered by scientiic research. It is important to promote new research strategies capable of providing signiicant and potentially novel contributions towards the comprehension of the technologies used in ancient times, as well as of their transfer and difusion in time across geographical areas. Scanning electron microscopy (SEM) coupled with energy dispersive X-ray spectroscopy (EDS) can be considered one of the most versatile instruments for analyses carried out in the ield of cultural heritage, being able to provide data on almost all materials. he continuous development of instruments and their combination is necessary in order to detect data on the micro/nano scale, and for the interpretation of minute details from speciic areas of an object. In the case of a jewel, it is necessary to go beyond the traditional information derived from SEM-EDS investigations by elaborating either the morphological information or the compositional data with methods derived from other disciplines, such as Image Analysis (IA) and statistical analysis (Cluster Analysis and Pearson’s Correlation Coeicients). It is indispensable to improve traditional devices in order to create an innovative instrumentation able to acquire data generally inaccessible to forgers. In the present work, the coupling of the method of X-Ray luorescence (XRF) with SEM-EDS, recently developed (Bovani et al., 2007), is applied for the irst time to the study of jewellery. By focusing on the argument of falsiication, the aim of the present work is to ofer contributions towards two main directions: – Innovation: improving traditional methodologies through the integration of experimental apparatuses, software and/or statistical methodologies for XRF-EDS compositional data treatment; – Knowledge: the deinition of diagnostic markers as characteristics of composition and/or morphology, and of chemical-physical processes allowing the identiication of technologies of ancient jewellery manufacture. he present work deals with the application of methodologies based on SEM-EDS through the study of some important Etruscan jewels: the 8th century BC ibula from Tarquinia (Fig. 1), a pair of bauletto earrings dating to the 6th century BC from Castellani’s Collection, and a hoop earring from Adria, dated to the 4th century BC (Fig. 2). Investigation on the micro/nano scale is essential for the identiication of diagnostic markers related to particular processes applied to materials, ranging from prehistoric lithic tools to the trimmings of large bronze statuaries. In the case of metals, this entails identifying the traces left during the working process, regardless of the metal employed and its speciic structural modiications, due to thermal processes or mechanical stress and strain. Particular attention must be dedicated to distinguishing degradation phenomena, such as the formation of corrosion products or use traces, which can be confused with those left by the craftsmen of the past. A speciic diagnostic approach is necessary for each artefact. 3. RESULTS Concerning the study of the ibula (Fig. 1) found in an archaeological excavation in Tarquinia (Rome) and dated to the 8th century BC (presenting some details not pertinent to traditional Etruscan jewellery), the scope of the scientiic investigation was the characterization of decorative elements, granulation and iligree, through their geometrical and compositional parameters and their comparison with homologous decorations from the Mediterranean area. In a previous work (Ferro et al., 2005), a comparison of objective elements present among diferent technologies employed in the creation of some jewels originating from diferent geographical areas (Iran, Cerveteri, Adria, Syria) in the temporal range 21st century BC-3rd century AD, was carried out. Figure 1: Etruscan ibula, 8th century BC, Tarquinia (Viterbo, Italy). Figure 1 : Fibule étrusque, VIIIe siècle avant J.-C., Tarquinia (Viterbo, Italie). ArcheoSciences, revue d’archéométrie, 33, 2008, p. 51-57 53 A multi-analytical approach for the identiication of technological processes in ancient jewellery Figure 2: Attachment of the Etruscan hoop earring, 4th century BC, Adria (Rovigo, Italy). Figure 2 : Boucle d’oreille étrusque, IVe siècle avant J.-C., Adria (Rovigo, Italie). SEM-EDS analyses provided information on the morphology of the grains forming the granulation decoration and on the elemental composition of the areas characterizing a particular technology, such as mechanical joining, soldering and sheet working. In addition, the geometrical characterization of the small and numerous grains was carried out using Image Analysis (IA) of the SEM micro-photographs. IA analysis is not so straightforward when applied to SEM images. For processing the software elaboration (Leyca QWin), it is necessary to establish an appropriate methodological approach that includes speciic steps (Fig. 3). he acquisition of SEM images must be made with an adequate mode of contrast and magniication. he necessary linear measurement calibration data has to be determined on images taken with stereo microscopy, optic microscopy and SEM in order to evaluate the degree of error involved. In order to optimize the results, the images were submitted to a b an elaboration that comprises enhancement, threshold and selection of grain proiles characteristics so as to allow diameter measurements with a minimum of uncertainty. he characterization of the grains’ shape, roundness, or rather the eccentricity in the bi-dimensional image, was determined by introducing the values of two orthogonal diameters in the ellipse equation. he second evaluation regards grain packing in the granulation decoration. he covering of a surface is of superior quality if most of the grains are in touch, without noticeable gaps between them. By considering grains as spherical geometrical elements, the concept of sphere packing can be applied. It is obvious that if the diameters of grains have diferent dimensions, the decorative motif contains pattern displacements. he surface is not covered homogenously, but shows discontinuity, thus diminishing the object’s overall aesthetic efect. For the granulation of each jewel, the average diameter and its standard deviation (σ) have been calculated by the approximation of measurement data to Gaussian distribution using the “Origin” software. he standard deviation measures the variation in terms of diameter size from its mean value, and indirectly relects on the goldsmith’s ability to produce and select grains of the same dimension and roundness. Another evaluated granulation feature is its ‘tridimensional’ efect, related to the position of the gold grains in relief with respect to the jewel’s surface. his efect is linked to thermal processes of grain soldering. A previous study (Ferro et al., 2003) demonstrated that, especially in autogenous soldering, the granules tend to collapse into the surface support if the heating is prolonged and the surface of the grains and gold substrate start to melt (excessive surface lowing). In copper salt joining, the Cu that originates from the reduction of its salts difuses in gold, and little material is left on the join, a ine meniscus formed between the base and the grain (Ferro et al., 2008), creating the efect of spheres per- c d Figure 3: Several steps of Image Analysis: a) SEM image acquisition, b) binarization, c) selection of features, d) statistical analysis. Figure 3 : Diférentes étapes d’analyse d’images : a) acquisition d’une image MEB, b) binarisation, c) sélection des caractéristiques, d) analyse statistique. ArcheoSciences, revue d’archéométrie, 33, 2008, p. 51-57 54 ched on a base. Conversely, brazing by adding an alloy with a lower melting point partially loods the grains, with the consequent loss of tridimensionality and design. In this work, ive independent parameters have been selected to distinguish diferent granulation scenarios: 1) join typology (welding), 2) diameter average, 3) diameter standard deviation (σ), 4) grain eccentricity (circumference has eccentricity = 0), and 5) tridimensional efect. All these parameters provide a collection of data that characterizes the physical features of grains and their clustering, and can thus be utilised as diagnostic markers for distinguishing jewels from diferent geographical regions and historical periods. he results of the investigation have shown that in the Iranian and Syrian jewellery, the grain diameters are of the order of 700 microns, with a standard deviation of about 90 microns, an eccentricity of 0.04, and have identiied the use of brazing as a soldering methodology, with a consequent loss of tridimensionality. In the Etruscan area, the average diameter is reduced to about 200 microns, with a standard deviation of 65 microns and eccentricity of 0.02. Etruscan granulation is characterized by the use of copper salts in most cases. After 300 BC, Etruscan jewellery shows a general loss of precision in terms of roundness, even though grain dimensions remain very small and the use of the copper salt soldering continues. he results obtained for jewels in the previous study are reported in Table 1. Concerning the Tarquinia ibula (Fig. 1), made in electrum, the same procedure has been applied, with the resulting parameters attributable to a working process of oriental derivation: the average diameter of 650 microns, with a standard deviation of 70 microns, the high incidence of non-spherical grains, the lack of tridimensionality, and the use of brazing as a soldering method. his data, coupled with other stylistic observations, such as the shape of the Phoenician palmette, the use of iligree wires cut at a right angle, and the imperfect form of the pin clasp, supports the archaeologist’s hypothesis of a possible oriental origin of the working process employed for this object, found in an Etruscan context. In other studies, supericial characterization is not suiciently exhaustive for the complexity of the investigation, and it is necessary to resort to radiation analyses in order to obtain information from the entire volume of the sample. In this ield, an integrated instrumentation that integrates electron microanalysis (EDS) and X-ray luorescence (XRF) within the SEM environment has been developed (Bovani et al., 2007). he SEM-EDS-XRF (Fig. 4) setup was created by inserting in the SEM LEO450VP chamber a tungsten anode X-ray tube with a geometry of 45° with respect to the sample and 90° with respect to the solid state (SiLi) detector, the latter having an active surface suitable for detecting ArcheoSciences, revue d’archéométrie, 33, 2008, p. 51-57 Daniela FERRO et al. X-ray signals originating from both systems. he advantages derived from the use of an integrated instrumentation are notable, such as the increase of sensitivity in the detection of trace elements. XRF reveals element quantities of the order of ppm, and it is possible to detect signals both from the irst supericial layers (EDS) and from the inner volume (XRF). When used in an SEM environment, XRF warrants better eiciency with respect to its separate use for several reasons, from a suitable ixed geometry to the operability in high vacuum by avoiding the absorption of low energy X-rays. INCA300, the EDS signals processing software, supplies an optimum support to XRF through an X-ray energy spectra calibration obtained by standards. he characteristics of both devices allowed discriminating the presence of diferent phases in the precious hoop earring from Adria, dated to the 4th century BC (Fig. 2), where several metal laminas overlap and diferent materials have been employed for its construction. he earring, approximately 5 cm in diameter, is composed of a cylindrical tube formed by the torsion of several wires, each of which is produced with rolled lamina. he terminals of the ‘tube’ are decorated by two small gold lamina cylinders with granulation and are joined to two semi-spheres in gold lamina, found partially crushed. Stylistic and geometrical considerations on the technique used to decorate the object with granulation have been derived on the basis of the previously described methodology, revealing its appurtenance to the period of Etruscan decadence. he grains, ixed by autogenous welding, have small diameters, but their dispersion is very bad, with the sole function of covering the soldering area, without any pattern. he unusual dimensions of the earring are in contrast with its light weight. he illing material in the inner part of the cylinder was likely used to preserve the form and avoid damage to the ine wire wrapping. he traditional X-ray radiography with a radiation potential of 30 keV allows us only to individuate the presence or absence of any illing materials, but not its chemical nature. he presence of a continuous compact matrix in the cylinder and inside the two semi-spheres appears evident in photographs, despite the impossibility to penetrate through the wrapped gold wires without compromising the artefact. he investigation with the integrated EDS-XRF system was tested in order to verify the hypothesis of the presence of some illing material. he EDS data revealed only the composition of the gold metal surface constituting the object, while the XRF spectra detected the presence of elements characteristic of ‘earth’. It is important to note that the XRF method operating in the SEM chamber kept under high vacuum can also detect low energy X-ray emissions. On the basis of the collected data, it is plausible that the goldsmith used a clay 55 A multi-analytical approach for the identiication of technological processes in ancient jewellery Figure 4: Scheme of experimental instrument SEMEDS-XRF. Figure 4: Schéma du dispositif expérimental MEBEDS-FX. Dating Origin Welding hree-dimensionality Average of diameter ± σ(*) μm σ of diameter μm Eccentricity Pendant PREIS. 9 ca. 2000 BC West Iran autogenous moderate loss of deinition 587 ± 14 65 00:04 Pendant PREIS. 36 2000-1000 BC Iranian plateau autogenous moderate loss of deinition 782 ± 18 97 00:03 Tarquinia’s Fibula 8th century BC Tarquinia brazing moderate loss of deinition 653 ± 15 70 — Earring PREIS.47 3rd century AD Iranian plateau brazing moderate loss of deinition 519 ± 8 66 00:04 Fibula Inv. N. 5442 7th century BC Narce copper salt good 242 ± 8 50 00:02 Object rd Pendent IGAD 9295 3 century BC Adria copper salt good 160 ± 24 147 00:07 Bulla IGAD 9306 2nd century BC Adria autogenous good 727 ± 37 149 00:03 (*) he errors have been calculated as standard deviation and through statistic error propagation. Table 1: Table of comparison parameters. Tableau 1: Tableau de paramètres de comparaison. rod on which he wrapped the series of gold wires. he clay would have two functions: irst, to create a support enabling the object to maintain its shape, and, second, to protect the very thin gold lamina forming the wires from the heating processes necessary to solder the terminal elements. he study of the Etruscan bauletto earrings (6th century BC) from Castellani’s Collection (Museo Nazionale Etrusco di Villa Giulia, Rome) is more complicated. he earrings under consideration are an example of pastiches: an original Etruscan jewel with repairs carried out by the Castellani, a Roman family of goldsmiths, merchants and collectors operating during the 19th century. he respective century was characterised by the proliferation of so called ‘integrative’ restorations, i.e. the restitution of the full shape of a historical object so as to introduce complete archaeological artworks on the antiquarian market, because customers valued the integrity of an object d’art over its authenticity. he Castellani family studied, reproduced and restored ancient jewels, and contributed to the spreading of the Italian archaeological jewellery style (Moretti Sgubini, 2000). With the aim of individuating the restored elements integrated into genuine Etruscan jewels, an initial study was undertaken on a contemporary pair of earrings produced by Castellani in the 19th century, in order to identify ArcheoSciences, revue d’archéométrie, 33, 2008, p. 51-57 56 the characteristics and the tool traces of their oicina, and subsequently to recognize ‘their hand’ on restored objects. A local heating process that partly homogenized the alloy on a micrometric scale and caused a modest compositional variation renders a precise delimitation of the Castellani interventions problematic. With the purpose of interpreting, distinguishing and identifying the correlation among the chemical elements composing the gold base, the decorative elements and their soldering area, the statistical analysis has been carried out using Pearson’s Correlation Coeicient and Cluster Analysis (Virgili et al., 2008). his statistical tool has been processed with SPSS 8.0 software on the data collected by several EDS acquisitions. Pearson’s Correlation Coeicient is a measure of the linear dependence between two elements of precious alloy identiied as variables, and allows distinguishing whether an element belongs to a speciic procedure rather than another. he K-Means Cluster Analysis (Romesburg, 2004) aims to classify groups of homogeneous elemental composition in order to identify the alloys composing ancient and modern soldering and the diferent areas of ancient and modern brazing. For the earrings created by Castellani in the 19th century, the elaboration of the statistical data has been performed on 74 spectra acquired on diferent areas, previously selected on the basis of their technical content. Pearson’s correlation matrix of elements of the precious alloy shows a high negative value of linear dependence between Au and Ag. his evidence suggests that the soldering alloy had a higher silver content. Copper does not seem to participate in the chemical-thermal reactions, while the presence of cadmium points to its probable use in the form of cadmium mineral (CdS). he yellow colour of both the metal and the mineral also suggests its use to camoulage the gold soldering. he cluster analysis allowed evaluating the elemental composition of the lamina and the soldering area. Cadmium is present only in the soldering areas, with a variable percentage of circa 1%. Figure 5: (See colour plate) Bauletto earrings, 6th century BC, Cerveteri (Rome, Italy). he modern items are highlighted in red, and the ancient ones in green. Figure 5 : (Voir planche couleur) Boucles d’oreille “a bauletto”, VIe siècle avant J.-C., Cerveteri (Rome, Italie). Les éléments modernes sont mis en évidence en rouge et les anciens en vert. ArcheoSciences, revue d’archéométrie, 33, 2008, p. 51-57 Daniela FERRO et al. In the analysis of the other pair of earrings from Castellani’s collection “ori antichi”, each piece was studied individually, so as to determine the composite nature of each artefact. One of the two earrings (referred to as A) presents clear signs of restoration. A diferent decorated lamina has been added on the recto and ixed by illing it with organic materials, such as binder and ibre. In this item, Cd is absent, and the linear dependence Au-Cu demonstrates the use of copper soldering, a technology typical of Etruscan manufacture, while the one between Au-Ag emphasizes the presence of silver joining, characteristic of the Castellani production. his suggests that the Castellani soldering interventions were superimposed on those of the Etruscan artisan, probably because the joining parts are the most vulnerable in time. In the other ancient earring (B), the most signiicant evidence is Au decrement with respect to the Cu and Cd concentrations, as revealed by the variables correlation analysis. here is no direct proportionality relation detected between Ag and Cu. his implies that in the analyzed areas the two elements are not correlated. he evaluation of the results suggests the exclusion of the possibility of any Ag-Cu alloy utilization, and the consequential presence of copper salt soldering areas altered by 19th century interventions employing cadmium. he study on the Castellani collection earrings has induced us to hypothesize that not only had a novel procedure been adopted in their oicina, but that experimentation had been undertaken and applied to the production of modern jewels, as well as to the restoration of ancient Etruscan artefacts. he results obtained by the application of statistical analyses to the EDS compositional values are summarized by visualizing (Fig. 5) the areas where the Castellani interventions (highlighted in red) were performed versus the original Etruscan structure (highlighted in green). A multi-analytical approach for the identiication of technological processes in ancient jewellery 4. DISCUSSION AND CONCLUSIONS he integration of scanning electron microscopy (SEM) with other techniques, such as the simultaneous use of X-ray luorescence operating in an SEM chamber, computer elaboration of image analysis, or statistical methods for the EDS microanalysis data treatment, transforms the microscope into a potent analytical tool capable of providing data not detectable with traditional diagnostic methodologies, thus promoting a new vision on the identiication of fakes. In this work, some cases have been presented in which the electronic spectroscopy data has been interpreted in a diferent way, with the support of speciic analytic approaches. In the study of the composition of diferent granulation decorations, the possibility to establish selection criteria for the interpretation of the ancient techniques employed led to the formulation of a methodology based on the deinition of ive characteristic elements: 1) grain diameter, 2) grain roundness, 3) frequency of the average value of the diameter, 4) soldering chemical composition, and 5) tridimensionality. his methodology, applied to the study of gold items originating from diferent geographical areas, allowed us to establish important considerations on the ibula from Tarquinia. his information, integrated and completed by philological and historical-artistic studies, contributes to the discussion of technological transfers of craftsmen’s abilities and manpower, the difusion of procedural recipes, and the trade of jewellery in the Mediterranean Basin. he use of an integrated instrument created by inserting an X-ray tube in the SEM chamber ofered the possibility to analyze both the surface and the inner part of the earring from Adria, and furnished important data for the attribution of the nature of the constituent materials, which in turn may be considered a valid support for an eventual restoration. A statistical approach has been employed for the elaboration of the numerous data related to the composition of the restored areas in two Etruscan earrings from Castellani’s Collection, resulting in a new diagnostic interpretation. 57 A concise, tailored description of a precious object, integrated with historical, iconographic and artistic evaluations, is the most efective ‘weapon’ in the ‘war’ to identify counterfeits. Acknowledgments he authors are grateful to David Loepp for the editing. References BOVANI, S., FERRO, D., CAMPANELLA, L., KHISTAT, B. and AKASHEH, T., 2007. Contributo chimico analitico nell’individuazione di oicine metallurgiche a Petra (III a.C.) attraverso l’interpretazione di reperti ferrosi, in C. D’Amico (ed.), Atti del congresso nazionale AIAR – Pisa, 1-3 febbraio 2006. Bologna, Pàtron publisher, 575-587. FERRO, D., FORMIGLI, E., PACINI, A. and TOSSINI, D., 2003. Experimental archaeology approach to the study of “copper salt” weldings, in Archaeometallurgy in Europe: international conference, 24-25-26 September 2003, Milan, Italy: proceedings. Milano, Associazione Italiana di Metallurgia, 24-26. FERRO, D., FORMIGLI, E., PACINI, A. and TOSSINI, D., 2008. La saldatura nell’oreiceria antica. Roma, Kappa publisher. MORETTI SGUBINI, A.M. (ed.), 2000. La Collezione Augusto Castellani. Roma, L’Erma di Bretschneider Press. ROMESBURG, H.C., 2004. Cluster analysis for researchers. Belmont, CA, Lifetime Learning Publications. VIRGILI, V., CARRARO, A. and FERRO, D., 2008. Cluster Analysis: a Precious Tool for Study Antique and Etruscan Jewellery from Castellani Collection, in G. Visco (ed.), Proceedings CMA4CH 2nd International Meeting on Multivariate Analysis and Chemometry for Cultural Heritage and Enviroment. Torino, Marco Valerio publisher, 69-70. ArcheoSciences, revue d’archéométrie, 33, 2008, p. 51-57 Helium, uranium and thorium analyses of ancient and modern gold objects: estimates of their time of manufacturing Analyse de l’hélium, de l’uranium et du thorium dans des objets anciens et modernes en or : estimation de leur date de fabrication Otto Eugster*, Ernst Pernicka**, Michael Brauns**, Alex Shukolyukov***, Valerie Olive**** and Stefan Roellin***** Abstract: Authenticating archaeological gold objects is a major problem, as gold is probably the most diicult material in terms of detecting modern forgeries. In 1996 we published the results of the study of faked and genuine gold crystals from the Santa Elena gold mine in Venezuela, and demonstrated that the U/h – He dating method is a powerful tool for detecting forgeries among gold objects. U and h decay by emitting He atoms that remain stored in gold. By measuring U, h, and He, the time when the He trapping process started can be determined. In the 1996 publication we also mentioned that the same method can be applied to ancient gold objects, in order to determine the time when they experienced their last melting process and, thus, to verify their antiquity. In the present work we report on a systematic and quantitative study of He, U, and h in a large number of ancient and modern gold objects. Résumé : L’authentiication d’objets en or est un problème majeur, car l’or est probablement le matériau le plus diicile quand il s’agit de détecter des contrefaçons modernes. En 1996 nous avons publié les résultats d’une étude de cristaux d’or faux et authentiques de la mine d’or de Santa Elena, au Venezuela, et nous avons montré que la méthode de datation U/h – He est un outil puissant pour détecter les contrefaçons dans le cas des objets en or. La décroissance de l’U et du h se fait par émission d’atomes d’He qui restent stockés dans l’or. La mesure de l’U, du h et de l’He permet de déterminer le moment de départ du processus de piège de l’He. Dans notre publication de 1996, nous avons aussi mentionné que cette même technique peut être appliquée aux objets d’or anciens, de façon à déterminer le moment correspondant à leur dernière fonte et ainsi de vériier leur ancienneté. Dans cet article nous présentons une étude systématique et quantitative de l’He, de l’U et du h dans un large nombre d’objets d’or anciens et modernes. Keywords: Antique gold, forgeries, helium, uranium, thorium, mass spectrometry. Mots-clés : or ancien, contrefaçons, hélium, uranium, thorium, spectrométrie de masse. * Physics Institute, University of Bern – Sidlerstrasse 5, 3012 Bern, Switzerland. (eugster@space.unibe.ch) ** Curt-Engelhorn-Zentrum Archäometrie – C5, Zeughaus, 68159 Mannheim, Germany. (ernst.pernicka@cez-archaeometrie.de), (michael.brauns@ cez-archaeometrie.de) *** Scripps Institution of Oceanography, UCSD – 9500 Gilman Drive, La Jolla, CA 92093, USA. (ashukolyukov@ucsd.edu) **** SUERC – Ranking Avenue, East Kilbride G75 0QF, Scotland. (v.olive@suerc.gla.ac.uk) ***** Bundesamt für Bevölkerungsschutz – Labor Spiez, 3700 Spiez, Switzerland. (stefan.roellin@babs.admin.ch) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 59-65 60 1. INTRODUCTION When gold crystals are formed in the Earth’s crust, other elements, such as U, are incorporated into the crystal lattice. hus, gold from mines and from river beds always contains traces of U and h. he three long-lived isotopes 238U, 235U, and 232h decay to Pb by emitting α-particles. An α-particle is the nucleus of the 4He atom, so when two electrons combine with an α-particle, a 4He atom is formed. he 4He atoms remain stored in the metal. A fourth long-lived radionuclide, 147Sm, also disintegrates by α-decay, but contributes in antique gold objects only a few percent to the total radiogenic (produced by radioactive decay) 4He. Because gold is highly retentive of this gas up to about 500 °C, the He atoms remain trapped in the metal. Beginning with 1992, we have studied these characteristics in numerous natural gold samples from all over the world (for references, see Eugster et al., 2009). In Table 1, the data on He, U, and h obtained for placer gold samples are provided. In a subsequent work (Eugster, 1996), we presented results on the study of genuine and faked gold crystals (Fig. 1) from the Santa Elena gold mine in Venezuela. he data for the genuine samples provided in the aforementioned paper and new data obtained for three samples of genuine crystals analyzed recently (Table 1) yield an estimated age of 240 million years (Ma). his time may be consistent with the Palaeozoic-Mesozoic (542 – 65.5 Ma) intrusive rocks of the gold mineralisation (Beda Hofmann, private communication). In Eugster (1996), we also presented He, U, and h data for faked gold crystals that appeared on the mineral market. hese data, complemented with two new He analyses and h concentrations obtained recently, demonstrate that the fake crystals are undatably young, the He concen- Figure 1: Faked octahedral gold crystals of up to 30 grams, purported to originate from the Santa Elena gold mine (Venezuela). Figure 1 : Faux cristaux en or octogonal de plus de 30 g, dont la provenance supposée est la mine d’or de Santa Elena (Venezuela). ArcheoSciences, revue d’archéométrie, 33, 2009 p. 59-65 Otto EUGSTER et al. trations being below the detection limit of the mass spectrometer. he He quantities in samples of art objects are extremely small for two reasons: (1) the time during which He has accumulated is usually only a few thousand years, in contrast to geological gold that formed millions of years ago; (2) only very small samples of a valuable object are available for the analyses. Inspired by our initial publications, specialists in the scientiic investigation of ancient metals saw the potential for additional authenticity studies. Kossolapov et al. (1999) and Kossolapov and Chugunova (2002), working at the State Hermitage Museum in St. Petersburg, Russia, collaborated with the company SPECTRON ANALYT in St. Petersburg to build a highly sensitive mass spectrometer and measured the He concentrations in a selection of ancient gold objects. he present work is a continuation of our earlier investigation of a number of antique gold objects (Eugster et al., 2009). Some objects described in this earlier publication are also included in the present work, because we obtained additional He data for them, and, in particular, because U and h analyses were performed for these objects, for which, in our earlier investigation, we had to employ average U and h values. 2. INVESTIGATED OBJECTS Test objects of modern manufacture In the course of our experimental work, we repeatedly analyzed samples from gold objects of modern manufacture to verify the absence or low abundance of radiogenic He expected for the short time of U and h decay since their manufacture. Table 2 lists the three objects used for this purpose. Results for these objects have already been reported in our earlier publication (Eugster et al., 2009). Because in the meantime additional He results were obtained and the U and h measurements had not been completed earlier, we will discuss the results for these objects. he He concentrations in eight samples of typically 10 mg of a gold coin Napoléon III, minted in 1857, were measured. New results for a wedding ring of 1886 are presented, as U and h concentrations are now known, allowing us to verify the recent date of manufacture. Finally, we intended to conirm the absence of He in 11 samples of a commercial gold wire, and we also determined its U and h concentrations. Helium, uranium and thorium analyses of ancient and modern gold objects… 61 Table 1: Noble gases, U and h concentrations and estimated ages of geologic gold samples. Tableau 1 : Gaz rares, concentrations d’U et h, et âges estimés d’échantillons géologiques en or. Table 2: Test objects of modern manufacture. Tableau 2 : Objets de test de fabrication moderne. Gold objects to be tested for their antiquity he objects to be tested in this study are listed in Tables 3 and 4. In most cases, the purported origin and age have been indicated by the owners of the objects. here are three objects that deserve to be described in more detail: (i) A gold torc (Fig. 2), purported to originate from the Hallstatt/La Tène transition period, about 5th century BC. Its diameter is 16.5 cm; X-ray luorescence analyses yielded 8% Ag and 1% Cu; the object is said to originate from southern Germany. he torc was obtained from Dr. Pieter Meyers, Los Angeles County Museum of Art, on behalf of a client. (ii) A igurine made by an extremely ine granulated golden openwork technique. he surface is strengthened by golden wires, which form the basic structure of this igurine (Fig. 3). It is damaged and repaired at various locations. he present weight is 24 g at a height of 7.3 cm. Appearance and manufacturing techniques were discussed by Shalem (2002). According to this author, decorating jewellery pieces with gold granules of various sizes was probably known in the early mediaeval Islamic world. However, the earliest visual material of Islamic gold jewellery with granulated openwork presently known is usually datable to and associated with the Fatimid period (Fatimid dynasty of Egypt, 969-1171 AD). An expertise based on technical (CT, SEM/EDX), chemical (Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry, LA-ICP-MS), and our He, U, and h analyses has been compiled by Senn et al. (2009). (iii) A signet ring with the picture of a male bust in side-view and an inscription in relected face ‘HILDEB ERTISREGIS’ (Fig. 4). In a comprehensive study of this ring, Weber (2007) concluded that it can be attributed to one of the two kings Childebert I or II, of the sixth century Merovingian dynasty of Western Europe. he weight of this gold ring (40.56 g) corresponds almost exactly to the weight of nine Byzantine solidi (gold ArcheoSciences, revue d’archéométrie, 33, 2009, p. 59-65 62 Figure 2: (See colour plate) Gold torc purported to originate from the Hallstatt/La Tène transition period (5th century BC). Diameter: 16.5 cm. Figure 2 : (Voir planche couleur) Torque en or supposément attribué à la période de transition Hallstatt/La Tène (Ve siècle av. J.-C.). Diamètre : 16,5 cm. Otto EUGSTER et al. Figure 4: (See colour plate) Signet ring attributed to kings Childebert I or II, of the sixth century Merovingian dynasty of Western Europe. Figure 4 : (Voir planche couleur) Bague à sceau attribuée au roi Childéric I ou II, de la dynastie Mérovingienne d’Europe Occidentale, VIe siècle. obtained at the Curt-Engelhorn-Zentrum Archäometrie in Mannheim using LA-ICP-MS. In the present work, we report additional data, in particular the U and h concentrations for the Childebert ring, which were not yet available for our previous publication. 3. EXPERIMENTAL METHODS Figure 3: (See colour plate) Golden igurine decorated with granulation. Possible origin: Iran or central Asia (11th or 12th century AD). For size, see text. Figure 3 : (Voir planche couleur) Figurine en or décorée de granulation. Origine possible : Iran ou Asie centrale (XIe ou XIIIe siècle). Pour les dimensions voir texte. coins), indicating that coins of this type were used by the goldsmith to manufacture the royal ring. Regarding the chemical composition of the gold in the Childebert ring, we refer to Weber (2007), who discussed the results that were ArcheoSciences, revue d’archéométrie, 33, 2009 p. 59-65 For the details of sample preparation and He analyses we refer to our earlier publication (Eugster et al., 2009). he U and h analyses used in this work were performed by six diferent laboratories. In most cases, multiple analyses for a particular gold object, and numerous tests and blank measurements had to be carried out. herefore, the work load would have been too large for a single laboratory, as the U and h analyses of gold objects were not the main priority of the respective scientists’ activity. Furthermore, two of the collaborators went into retirement in the course of the present work. he following laboratories were involved in this study: (i) Curt-Engelhorn-Zentrum Archäometrie, Mannheim (Ernst Pernicka, Michael Brauns, Boaz Paz), using quadrupole inductively coupled plasma mass spectrometry after electrolytic separation of U and h from Au and other metals on graphite electrodes; (ii) Scripps Institution of Oceanography, UCSD, La Jolla, CA (Alex Shukolyukov, Paterno Castillo), using an inductively coupled plasma mass spectrometer (ICP-MS) after chemical separation of U and h from Au and other metals; (iii) Labor Spiez (Stefan Röllin), digestion of Au samples in aqua regia, direct analy- Helium, uranium and thorium analyses of ancient and modern gold objects… sis of U and h with an ICP-MS; (iv) Scottish Universities Environmental Research Center (Valerie Olive), using an ICP-MS after chemical separation of U and h; (v) Institute of Geology, University of Bern (Jan Kramers), using a multicollector inductively coupled plasma mass spectrometer with desolvating nebulizer, after chemical separation of U and h and addition of a 236U and a 229h spike; (vi) Department of Chemistry and Biochemistry, University of Bern (Urs Krähenbühl), using an ICP-MS after chemical separation of U and h. For an inter-laboratory comparison, a large number of different samples of object DL 807 (Table 3) were prepared. In ive diferent laboratories, samples of this ancient gold object were measured for their U concentration, the most important element for He production. he following results were obtained (in case of multiple analyses, the average values are given): 0.9 ppb (laboratory (i), as indicated above), 3.5 ppb (ii), 1.9 ppb (iii), 5.5 ppb (v), 12.8 ppb (vi). he diferences in the U concentration obtained by these laboratories are 63 presumably due to sample inhomogeneity. his shows that it is important to analyze the He, U, and h concentrations in the same section of the sample. 4. RESULTS AND DISCUSSION Objects of modern manufacture he results are given in Table 2. he He concentrations for the three objects are extremely low; for the wedding ring of 1886 and the commercial gold wire, they are at or below the detection limit of the mass spectrometer. For the Napoléon coin, we used typically 10 mg of gold material (after etching) to obtain a reliable signal. he age estimate is consistent with the time when the coin was minted. Table 3: Ancient gold objects Tableau 3 : Objets en or anciens ArcheoSciences, revue d’archéométrie, 33, 2009, p. 59-65 64 Ancient gold objects he gold objects that could be conirmed to be ancient are compiled in Table 3. Based on the 4He, U, and h concentrations, their ages were calculated using the formula T = [ 4He ] / (3.24 x 106 [ U ] + 7.70 x 105 [ h ] + 4.02 x 103 [ Sm ]) where 4He is expressed in atoms per g and U, h, and Sm in ppm. he U, h – He age results are in years. For the derivation of this formula and for the decay constants of the radionuclides and their relative abundances we refer to our previous publication (Eugster et al., 2009). Usually the main contribution of 4He comes from U, whereas Sm contributes only a few percent to the total He. he estimated U, h – He ages are consistent with the presumed date of manufacture of these gold objects. he experimental errors are up to 50% because in most cases the He, U, and h concentrations are very low. he counting statistics for the measurements for each of these elements lead to an uncertainty of about 30%. In the following, we discuss two objects in more detail: (i) as mentioned in section 2, the result for the signet ring of a Merovingian king was already presented in Eugster et al. (2009). In the present work, we obtain a more reliable age of 1460 ± 400 years, because U and h concentrations are now available. his age is in good agreement with the time when the kings Childebert I and II lived. (ii) Two diferent samples of the Islamic igurine were analyzed for He and chemical elements, and two diferent U, h – He ages were obtained. As outlined by Senn et al. (2009), the igurine is characteristic for the Islamic tradition and dated to the 11th or 12th century AD. Based on the trace element pattern, it can be inferred that the igurine derives from the area of Iraq Otto EUGSTER et al. and Syria. he igurine is manufactured from a native goldsilver alloy, using copper as soldering material. he measured alloy compositions of the feet vary considerably. herefore, it can be assumed that at least two diferent materials were used to manufacture the igurine or to repair it. Our dating of the igurine provides two dates, one to the period of 1800 ± 800 years ago, and a second one 170 ± 100 years ago (Table 3). he older material belongs to a sample taken from an area between the legs of the trousers, a section of the igurine that is well preserved. he younger material was taken from the spine, where the igurine is strongly damaged and where the elemental composition difers from that of the undamaged parts. It appears that repairs were carried out on the damaged spine in more recent times. In general, the natural scientiic study conirms the results of the stylistic study (Senn et al., 2009). Modern and undatable gold objects In Table 4, gold objects are listed that turned out to be modern or were undatable due to a He excess of unknown origin. For three objects, the He concentration was below the detection limit or extremely low, resulting in a modern date. One of these three objects is not speciied in Table 4, as the owner did not give the permission to mention the details in this paper. Some objects yielded He concentrations, for which, together with the U and h concentrations, an unreasonably high U, h – He age was obtained. hese objects must contain helium that was not completely outgassed when the object was manufactured. Among these objects are samples Java 1-4, for which, in an earlier report to the owners, we estimated an ancient origin. A more critical interpretation, after having found more similar cases Table 4: Modern and undatable gold objects Tableau 4 : Objets en or modernes et non datables ArcheoSciences, revue d’archéométrie, 33, 2009 p. 59-65 Helium, uranium and thorium analyses of ancient and modern gold objects… of He excesses, led us to the conclusion that the Java 1-4 samples are undatable. his leads us to acknowledge that the U, h – He dating method presented in this work for detecting forgeries in the market of antique gold objects is still far from being a reliable dating method. he reason for this are the extremely low He, U and h concentrations, which are at the limit of being precisely determined by the presently available techniques. We observed that some antique gold objects contain inclusions, such as micron sized quarz and feldspar grains that did not completely release all He when the gold object was manufactured. herefore, we plan to search for inclusions by electron microscopic investigations. On the other hand, it appears that in cases where the original material for the production of the gold object was already processed before, the dating method works ine. Examples are the Childebert signet ring, probably produced from Byzantine solidi, and the Napoléon gold coin. Knowing that gold is very often recycled, the determination of its chemical composition is not a inal test for authenticating an ancient object, and dating, performed using the method presented in this work, may be necessary. Acknowledgements he authors are greatly indebted to the institutions that supplied the gold samples studied in this investigation: Alex Kossolapov, State Hermitage Museum, St. Petersburg; Pieter Meyers, Los Angeles County Museum of Art; Jack Ogden, Gemmological Association and Gem Testing Laboratory, London; Melanie Roy, TK Asian Antiquities Williamsburg, VA, USA; Marianne Senn, Zentrum für Kulturanalytik, EMPA, Dübendorf, Switzerland; Dietrich Willers, University of Bern. We thank the collaborators, who provided U and h data in addition to those provided by the co-authors of this paper: Boaz Paz, Curt-Engelhorn-Zentrum Archäometrie, Mannheim, Germany; Urs Krähenbühl, University of Bern; Jan Kramers, University of Bern; Paterno Castillo, Scripps Institution of Oceanography, UCSD, La Jolla, CA. We also thank Arthur Ghielmetti, Armin Schaller and Markus Zuber for their technical assistance. 65 References EUGSTER, O., 1996. Applications for noble gas analyses of gold. Gold Bulletin 29(3): 101-104. EUGSTER, O., KRAMERS, J. and KRÄHENBÜHL, U., 2009. Detecting forgeries among ancient gold objects using the U,h – 4He dating method. Archaeometry 51(4): 672-681. KOSSOLAPOV, A.J. and CHUGUNOVA, X.S., 2002. Authenticating ancient gold using the U-He radiogenic clock, in R. Van Grieken, K. Janssens, L. Van’t Dack, G. Meersman (eds.), Proceedings ART 2002, 7th International Conference on Nondestructive Testing and Microanalysis for the Diagnostics and Conservation of the Cultural and Environmental Heritage, 2-6 June 2002, Congress Centre Elzenveld, Antwerp, Belgium. Antwerp, University of Antwerp, CD-ROM. KOSSOLAPOV, A.J., IVANOV, A.P. and KUZNETSOV, P.B., 1999. Helium radiogenic clock for dating of archaeological gold, in W. Mc Crone, D.R. Chartier, R.J. Weiss (eds.), Proceedings SPIE, v. 3315, Scientiic Detection of Fakery in Art. San Jose, CA, SPIE, 16-20. SENN, M., FLISCH, A., EUGSTER, O., GÜNTHER, D. and VONMONT, H., 2009. Test report No. 450’588, EMPA Ueberlandstrasse 129, Dübendorf, Switzerland. SHALEM, A., 2002. A note on a unique Islamic golden igurine. Iran 40: 173-180. WEBER, A.G., (ed.), 2007. Der Childebertring und andere frühmittelalterliche Siegelringe. Köln, Weber. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 59-65 Modern mineralogy of gold: overview and new data Minéralogie moderne de l’or : bilan et nouvelles données Ernst Spiridonov* and Denka Yanakieva** Abstract: We suppose that it should be useful for archaeologists to have an overview on gold mineralogy, because 1) in ancient times, part of the golden objects were made directly from natural golden nuggets; 2) most of the Au in ores exists as its own minerals. he major part of the Au in the planets and meteorites of our Solar system is found in high temperature solid solutions: metallic Fe-Ni and monosulides Fe-Ni and Fe-Cu. Au leaves them under luid or some other reworking. As a result, Au minerals are formed. hey are mainly developed in hydrothermal deposits of the upper part of Earth’s continental crust. Au is the noblest chemical element. hus, the most part of Au in deposits appears as native gold: Au-Ag, Au-Ag-Hg, Au-Cu, Au-Pd, and Au-Pt system minerals. he most important feature of native gold is its ineness, i.e. the Au content in the mineral species, expressed in ‰. Chalcogenides of Au – sulides and selenides, and, similar to them, tellurides, plumbotellurides, antimonides, bismuthides – are not so widespread. Data on the 44 Au mineral species known today are provided in this overview. he irst quantitative estimations of natural native gold nano-dimensional crystallites of several mesothermal deposits are enclosed. Résumé : Nous supposons qu’il est important pour les archéologues d’avoir une vue générale sur la minéralogie de l’or car : 1) dans le passé une partie des objets en or ont été produits directement à partir de pépites d’or ; 2) la plus grande partie de l’or existe sous la forme de ses propres minerais. La plus grande partie de l’or dans les planètes et les météorites de notre système solaire se trouve en solutions solides à haute température : Fe-Ni métallique et Fe-Ni monosulites et Fe-Cu. L’or les laisse sous la forme de luides ou sous une forme équivalente. Le résultat est la formation de minerais d’or. Ces minerais sont essentiellement développés dans des gisements hydrothermaux de la partie supérieure de la crôute terrestre. L’or est l’élément chimique le plus noble. Pour cette raison, la majorité de l’or dans les gisements se présente sous forme native : systèmes minéraux Au-Ag, Au-Ag-Hg, Au-Cu, Au-Pd, Au-Pt. La caractéristique la plus importante de l’or est son titre, c’est-à-dire la quantité d’or dans les espèces minérales en ‰. Chalcogenides d’or – sulites et sélénites et similaires comme les tellurites, plumbotellurites, antimonides, bismuthides – sont moins répendus. Les données des espèces minérales 44Au connues actuellement sont fournies dans cet article. Les premières estimations quantitatives de cristallites nano-dimensionnelles d’or natif naturel de divers gisements mesothermiques ont aussi été inclues. Keywords: mineralogy, gold, nanomineralogy. Mots-clés : minéralogie, or, nano-minéralogie 1. IntroductIon he major part of the Au in the planets and meteorites of our Solar system is found in high temperature solid solutions – metallic Fe-Ni and monosulides Fe-Ni (Mss) and Fe-Cu (Iss). Au leaves them under luid reworking. As a result, Au minerals are formed. hey are mainly developed in hydrothermal deposits of the upper part of Earth’s continental Earth. Au is the noblest chemical element. hus, the most part of Au in deposits appears as native gold – Au-Ag, Au-Ag-Hg, Au-Cu, Au-Pd, and Au-Pt system minerals. he most important feature of native gold feature is its ineness, i.e. the Au content in the mineral specimen, expressed in ‰. Chalcogenides of Au – sulides and selenides, and, similar to them, tellurides, plumbotellurides, * Department of Mineralogy, Lomonosov State University of Moscow – Vorobiovy Gory, 1192334, Russia. (mineral@geol.msu.ru) ** Museum “Earth and Man” – Bulevard Cherny Vrch 4, Soia, 1421, Bulgaria. (denkay@abv.bg) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 67-73 68 Ernst SPIRIDONOV, Denka YANAKIEVA antimonides, bismuthides – are not so widespread. Much Au is found as the inest impurity in hydrothermal nonstoichiometric As-bearing pyrite Fe(S,As)2-x and arsenopyrite FeAs1-yS1+y and bitumen-like Au-organic compounds. 2. Au – Ag system mInerAls he main part of native gold appears as cubic solid solutions – Au-Ag system minerals. According to the results of thousands of precision analyses (Brauns, 1906; Vernadsky, 1914; Lindgreen, 1933; Smolin, 1970; Editorial Committee, 1970; Petrovskaya, 1973; Berman et al., 1978; Nedra, 1984-1990; Oberthür and Saager, 1986; Shikazono and Shimizu, 1987; Amuzinsky et al., 1992; Nekrasov et al., 1994; Spiridonov, 1995; So et al., 1995; Frimmel and Gartz, 1997; Hytőnen, 1999; Knight and Leitch, 2001, etc.) it was established that the series Au-Ag is uninterrupted in composition. hese series are ixed in the limits of the series (Vernadsky, 1914; Petrovskaya, 1973): gold (0-30 wt% Ag; ineness 1000-700); electrum (30-70 wt% Ag; ineness 700-300); kuestelite (7090 wt% Ag; ineness 300-100); gold-containing silver (90100 wt% Ag; ineness 100-0). he conirmed mineral species are noted in bold. On observations of the authors, the phenomena of disintegration of solid solution in Au-Ag minerals are not shown. he size of gold crystals varies from colloidal (< 0.1 microns) up to several cm, usually around a fraction of mm. Colloidal gold colours a chalcedony-like quartz in green colour (Baley, Transbaykalia). he size of the granulated intergrowths of gold (nuggets) is up to one metre. As Al-Biruni (1963) noted, the largest nugget of gold discovered weighed ~ 2500 kg (Zaruban, Afganistan). High-standard gold (Fig. 1) is typical for plutonogenic hydrothermal gold-quartz deposits. hese formations are the main source for placer Au deposits. Electrum, kuestelite and Au-bearing silver, their dendrites, skeletal and wire-shaped crystals (Fig. 2) are typical for volcanogenic hydrothermal Au-Ag deposits (Goldschmidt, 1918; Lindgreen, 1933; Petrovskaya, 1973). Widespread in these deposits, electrum corresponds to the famous ‘electron’ of the ancient Greeks, and the favourite ‘white gold’ of the Incas. Gold crystals of primary endogenic deposits are zonal: later zones are enriched by Ag and Hg (usually insigniicantly). In metamorphosed deposits, grains of gold are azonal, and intergrowths of diferent composition are observed (Frimmel and Gartz, 1997). In placers, grains of gold are usually enclosed by margins of porous high-standard gold; silver from these margins is leached. In placers, native gold often is found together with minerals of platinum group elements (ferroplatinum Pt3Fe, alloys Os-Ir-Ru, etc.). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 67-73 Figure 1: (See colour plate) Gold (ineness 885) in quartz vein. 81 mm. Bestube, Northern Kazakhstan. Collection and photography – E.S. Perhaps this deposit was one of the gold sources for the well known ancient Scythian jewellery. Figure 1 : (Voir planche couleur) Or (titre 885) dans le ilon de quartz. Bestube, nord du Kazakistan. Collection et photographie – E.S. Peut-être ce gisement a été une source d’or, exploité pour la production d’orfèvrerie Scythe. 3. Au (Ag) – Hg system mInerAls After Ag, Hg is the second most typical element impurity in native gold. In ores formed under low f S2 conditions, there are cubic α-amalgams: mercurian gold (Au, Hg) (Nazmova and Spiridonov, 1979; Oberthür and Saager, 1986; Shikazono and Shimizu, 1988; Amuzinsky et al., 1992); mercurian electrum (Au, Ag, Hg), mercurian kuestelite (Ag, Au, Hg), gold-containing mercurian silver (Spiridonov and Pletnev, 2002). Mercurian gold contains up to 18 wt% Hg, and mercurian silver up to 26% Hg. Au amalgams enriched by Hg are much more rare. hey usually associate with native Hg: hexagonal weishanite (Au,Ag)3Hg2 (Li et al., 1984; Wu, 1993); cubic gold amalgama γ-Au2Hg3 (Berman and Harcourt, 1938). Almost all Au amalgams found in placer deposits are technogenic formations (Wu, 1993). 4. Au – cu system mInerAls Cu is the third most typical element impurity in native gold, after Ag and Hg. Au-Cu phases poor in silver and Au-Ag phases poor in copper are formed at temperatures below 350 °C (Lyakischev, 1996). Au-Cu minerals (cuprian 69 Modern mineralogy of gold: overview and new data of thin lattice solid solution disintegration structures: Au + tetraauricupride and tetraauricupride + mineral Au3Cu. he disintegration structures are known, where the matrix is Au and lamellas are mineral Au3Cu; probably they are products of the non-stoichiometric mineral Au3+xCu’s disintegration. he standard roentgenometric data that are required for the determination of Au-Cu system minerals are provided in Chvileva et al. (1988). 5. Au – Fe system mInerAls It was established by a combination of magnetometric and mineralogical methods that the high-magnetic ferrous gold with 4-5 wt% Fe described earlier shows ine crystal intergrowths of gold with 0-0.2 wt% Fe, and magnetite octahedrons 0.01-0.2 microns in size (Yacubovskaya et al., 1989). 6. Au – Pd, Pt, os, Ir system mInerAls Figure 2: (See colour plate) Golden wire in calcite. 4 mm. Tyrny-Auz, Northern Caucasus. Collection of V. Andreenko. Photography – M. Bogomolov. Figure 2 : (Voir planche couleur) Fil d’or dans de la calcite. 4 mm. Tyrny-Auz, nord du Caucase. Collection de V. Andreenko. Photographie M. Bogomolov. gold) are found in hydrothermal Au deposits, among listvenitized peridotites, komatiites and rodingites (Lozechkin, 1939; Spiridonov and Pletnev, 2002). hese last minerals replace native copper. Cubic auricupride AuCu3 is known (Ramdohr, 1967). Metastable cuproauride AuCu corresponds to the low ordered α-(Au, Cu) solid solution (Lozechkin, 1939; Chvileva et al., 1988). he mineral corresponding to the Au3Cu cubic modiication (Knipe and Fleet, 1997; Spiridonov and Pletnev, 2002) does not have an authorized name. Below, it is referred to as mineral Au3Cu. he products of cuproauride solid phase transformation are tetragonal tetraauricupride AuCu (Chen et al., 1982). Tetraauricupride can contain up to 17 wt% Pd (Spiridonov et al., 2003). Tetraauricupride – hongshiite CuPt solid solutions are known (Kwitko et al., 2002). After annealing, non-stoichiometric cuproauride Au1+xCu shows two types Au-Pd and Au-Pt system minerals are typical rare minerals of telethermal Au deposits formed at low f S2 and at heightened f O2. hey are cubic palladian gold (porpezite) (Au, Pd). (Vernadsky 1914; Kwitko et al., 2002); platinian gold (Au, Pt) (Levin et al., 1986); intermetallides with compositions similar to Pt3Au and Pt(Au,Pd). Even a low impurity of Pd (of the order of a few percent) results in the steel-grey colour of this type of native gold. 7. Au – BI, sB, PB, sn system mInerAls Intermetallides of gold – maldonite Au2Bi (Hytőnen, 1999) and aurostibite AuSb2 (Graham and Caiman, 1952; Spiridonov, 1995) are rather rare minerals of hydrothermal ores. Maldonite contains ~ 65 wt% Au, and aurostibite ~ 45 wt% Au. Aurostibite pseudomorphs after maldonite are rarely encountered; such aurostibite contains up to 9 wt% Bi. hese minerals are formed under low f S2. Maldonite usually replaces native bismuth. Aurostibite usually replaces native gold. he following three intermetallides: anyuiite AuPb2 (Razin and Sidorenko, 1989), hunchunite Au2Pb (Shangquan et al., 1992), yuanjiangite AuSn (Chen Lichang, 1994) are found only in placers. Perhaps these three intermetallides are technogenic products. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 67-73 70 Ernst SPIRIDONOV, Denka YANAKIEVA 8. Au – Ag (cu) – te system mInerAls 10. comPlex cHAlcogenIdes Au and Ag tellurides are typical minerals of hydrothermal Au deposits. In a series of cases, they are the main Au and Ag bearing minerals in ores. Hg and Cu are typical elemental impurities in these tellurides. Triclinic montbrayite Au2(Te,Sb,Pb,Bi)3 (Peacock and hompson, 1946; Chvileva et al., 1988) is a rare telluride, and it contains ~ 50 wt% Au and up to 5 wt% Sb. Monoclinic calaverite AuTe2 (Stillwell, 1931; Markham, 1960) contains ~ 45 wt% Au, and usually < 0.5 wt% Ag. In some deposits (Kalgoorly, Australia), there are calaverite aggregates reaching 1m3 in size. Rhombic krennerite Au3(Au,Ag,Cu)Te8 (Markham, 1960; Editorial Committee, 1970; Chvileva et al., 1988) contains ~ 40 wt% Au, and 0.5-6.5 wt% Ag; its composition varies from AuTe2 to Au3AgTe8. Monoclinic silvanite Au(Ag,Au,Cu)Te4 (Stillwell, 1931; Markham, 1960; Editorial Committee 1970; Chvileva et al., 1988) contains ~ 25 wt% Au. Kostovite Au(Cu,Ag,Au)Te4 (Terziev, 1966), similar to silvanite, is a typical reactionary mineral of volcanogenic Au ores. Mutmannite AuAgTe2 (Spiridonov and Chvileva, 1985), developed in the cementation zone of the gold-telluride deposits, is an example of mimicry in the mineral kingdom, because it is indistinguishable from petzite optically and on spectra of relection light, and from calaverite according to the X-ray pattern. Cubic petzite AuAg3Te2 (Markham, 1960; Chvileva et al., 1988) with garnet-like structure is the most widespread Au-Ag telluride; it contains ~ 25 wt% Au. Cubic solid solution (Ag,Au)2Te is rarely encountered as homogeneous phase; usually it is transformed into petzite and hessite intergrowths. Rhombic nagyágite AuPb5Te4-x(Sb,As)xS6 (Spiridonov, 1991a; Simon et al., 1994) – AuPb5Te2(Sb,Bi)S6 (Johan et al., 1994), monoclinic museumite AuPb5SbTe2S12 (Bindi and Cipriani, 2004), rhombic buckhornite AuPb2BiTe2S3 (Francis et al., 1992; Johan et al., 1994), monoclinic criddleite Au3Ag2TlSb10S10 (Harris et al., 1988), monoclinic jonassonite AuBi5S4 (Paar et al., 2006) are typical formations of Au and Au-Ag volcanogenic hydrothermal deposits. Nagyágite – ‘grey foliated gold ore’ – is the most widespread among them. 11. PlumBotellurIdes And stIBIo-PlumBotellurIdes oF tHe Au-cu-Fe system hese hypergenic minerals are developed in the cementation zone of gold-telluride deposits. hey are monoclinic (?) bilibinskite Au5Cu3(Te,Pb,)5 and bilibinskite-(Sb) Au6Cu2(Te,Pb,Sb)5 (Spiridonov, 1991b), rhombic (?) bogdanovite-(Cu) – Au5Cu3(Te,Pb)2 and bogdanovite(Fe) – Au5CuFe2(Te,Pb)2 (Spiridonov, 2008), monoclinic (?) bezsmertnovite Au4Cu(Te,Pb) (Spiridonov, 1991b). Bilibinskite, bogdanovite and bezsmertnovite contain ~ 9. Au – Ag – se, s system mInerAls Cubic fischesserite AuAg3Se2 (Johan et al., 1971), cubic uytenbogaardtite AuAg3S2 (Barton et al., 1978), monoclinic petrovskaite AuAgS - AuAg(S,Se) (Nesterenko et al., 1984; Nekrasov et al., 1988), hexagonal penzhinite AuAg4(S,Se)4 (Bochek et al., 1984), cubic solid solution (Ag,Au)2Se (Spiridonov et al., 2009), cubic solid solution (Ag,Au)2S (Samusikov et al., 2002) are typical minerals of Au volcanogenic hydrothermal deposits, from low sulide to rich sulide ones. Petrovskaite is also developed in crusts of weathering of sulide rich deposits, in association with jarosite and native sulphur. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 67-73 Figure 3: (See colour plate) Bogdanovite (pseudomorph after kostovite) in gold-quartz vein. 2 mm. Bogdanovite with strong coloured birelection. Polished section in relected light. Aginskoe, Kamchatka. Collection and photography – E.S. Figure 3 : (Voir planche couleur) Bogdanovite (pseudomorphe de kostovite) dans le ilon de quartz aurifère. Bogdanovite avec birélectione fortement colorée. Section polie sous lumière réléchie. Aginskoe, Kamchatka. Collection et photographie – E.S. 71 Modern mineralogy of gold: overview and new data 50, 65 and 75 wt% Au, respectively. Macroscopically, they have the appearance of bornite. In relected light, they are very speciic, being characterized by bright colour doublerelection (Fig. 3) and anisotropy. Bilibinskite, bogdanovite and bezsmertnovite replace kostovite, krennerite, silvanite, and nagyágite. In the oxidation zone, plumbotellurides of gold are substituted by inely scalloped gold and tellurides of Cu, Pb, and Fe. 12. conclusIon Brief data on the 44 gold mineral species known today are provided above. Each of them has an individual chemical composition, crystal structure, X-ray pattern, unit cell parameters, physical properties (colour, hardness, spectra of relection light, etc.). For many of these minerals, the most important feature is their ineness, i.e. Au content expressed in ‰. A main new feature – new characteristic of gold minerals – could probably be related to the area sizes of X-radiation coherent dissipation, i.e. the crystallite size. For the majority of crystal matters, the crystallites’ size is nano-dimensional, ranging from a few nm up to several hundreds of nm. he description of the crystallites size can be found in a paper by D. Yanakieva and co-authors in this issue of ArcheoSciences. he irst interesting data on the nano-dimensions of crystallites is obtained. High-standard gold from large mesothermal deposits in Northern Kazakhstan, with an age of 445 ± 4 Ma, is studied. hese deposits are of a diferent depth facies – hypabyssal Au-Sb Bestube, mesoabyssal Au Stepnyak, abyssal Au-telluride (Spiridonov, 1995). One of the studied specimens from Bestube is shown in Figure 1. he crystallite sizes of tens of studied Bestube gold specimens are 20.0 ± 0.2 nm; for Stepnyak gold, 21.6-21.8 nm; for Aksu gold, 26.0 ± 0.3 nm (investigator D. Yanakieva). hus, for this group of deposits, the direct correlation between the size of crystallites and the depth of formed gold ores is established. he size of the crystallites of hypergenic gold from Zana-Tube is quite diferent: 31.3-31.8 nm. Acknowledgements his research was supported by grant no. 07-05-00057 of the Russian Foundation for Basic Research. References Al-BIrunI, A., 1963. Materials for investigations of the precious stones. Leningrad: Academy of Sciences USSR (in Russian). AmuzInsky, V., AnIsImoVA, g. and zdAnoV, yu., 1992. Native gold of the Yakutia: Verchne-Indigirskii region. Novosibirsk: Nauka (in Russian). BArton, m., kIeFt, c., Burke, e. and oen, I., 1978. Uytenbogaardtite, a new silver-gold sulfide. The Canadian Mineralogist 16: 651-657. BermAn, H. and HArcourt, g., 1938. Natural amalgams. American Mineralogist 23: 761-764. BermAn, yu., BotoVA, m., BocHek, l. and PlescHAkoV, A., 1978. Natural system of the gold – silver. Geochemistry 9: 1351-1359 (in Russian). BIndI, l. and cIPrIAnI, c., 2004. Museumite, AuPb5SbTe2S12 , a new mineral from the gold-telluride deposit of Sacarîmb, Metaliferi Mountains, western Romania. European Journal of Mineralogy 16: 835-838. BocHek, l., sAndomIrskAyA, s., cHuVIkInA, n. and kHostoV, V., 1984. New selenian sulide of the silver, gold and copper – penzinite Au(Ag,Cu)4(S,Se)4. Zapiski Vsesoynznogo Mineralogicheskogo. Obshchestva 113: 356-360 (in Russian). BrAuns, r., 1906. Kingdom of minerals. St. Petersburg: A.F. Devrien. (in Russian). cHen, k., tInggAo, z. and Peng, z., 1982. Tetraauricupride, CuAu, discovered in China. Scientia. Geologica Sinica 11(1): 111-116 (in Chinese with English abstract). cHen l., , tAng, c., zHAng, J. and lIu, z., 1994. A new auriferous and stanniferous mineral, Acta Petrologica Et Mineralogica 13(3): 232-238 (in Chinese with English abstract). cHVIleVA, t., BezsmertnAyA, m. and sPIrIdonoV, e. (eds), 1988. Reference book for determination of the ore minerals in relected light. Moscow, Nedra (in Russian). edItorIAl commIttee For “IntroductIon to JAPAnese mInerAls”, orgAnIzIng commIttee ImA-IAgod meetIngs, ’70, 1970. Introduction to Japanese minerals. Tokyo, Geological Survey of Japan. FrAncIs, c., crIddle, A. and stAnley, c., 1992. Buckhornite, AuPb2BiTe2S3, a new mineral species from Boulder County, Colorado, and new data for aikinite, tetradymite and calaverite. he Canadian Mineralogist 30: 1039-1048. FrImmel, H. and gArtz, V., 1997. Witwatersrand gold particle chemistry matches model of metamorphosed, hydrothermally altered placer deposits. Mineralium Deposita 32: 523-531. nedrA 1984-1990. Gold deposits of the Soviet Union. Vol. 1-5. Moscow, Nedra Press (in Russian). goldscHmIdt, V.m., 1918. Atlas der Krystallformen. Gold. Band IV. Heidelberg, Carl Winters Universitatsbuchhandlung. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 67-73 72 grAHAm, A. and cAImAn, s., 1952. Aurostibite, AuSb2, a new mineral in the pyrite group. American Mineralogist 37: 461469. HArrIs, d., roBerts, A. and lAFlAmme, J., 1988. Criddleite TlAg2Au3Sb10S10, a new gold-bearing mineral from Hemlo, Ontario, Canada. Mineralogical Magazine 52: 691-697. Hytőnen , k., 1999. Suomen Mineraalit. Erillisjulkaisen, Geologian Tutkimuskeskus. JoHAn, z., dodony, I., morAVek, P. and PAšAVA, J., 1994. Le buckhornite, Pb2AuBiTe2S3, du gisement d’or de Jilove, Republique Tcheque. Comptes Rendus de l’Académie des Sciences Paris 318: 1225-1231. JoHAn, z., PIcot, P., PIerrot, P. and kVAček, m., 1971. La ischesserite, Ag3AuSe2, premier selenide d’or, isotype de la petzite. Bulletin de la Société française de minéralogie et de cristallographie 94: 381-384. knIPe, s. and Fleet, m., 1997. Gold-copper alloy minerals from the Kerr-Edison Mine, Ontario. he Canadian Mineralogist 35: 573-586. knIgHt, J. and leItcH, c.H.B., 2001. Phase relations in the system Au–Cu–Ag at low temperatures, based on natural assemblages, he Canadian Mineralogist 39: 889–905 knIttel u., 1989. Composition and association of arsenian goldieldite from the Marian Gold deposit, Northern Luzon, Philippines // Mineral. Petrol. 40: 145-154. kwItko, r., cABrAl, A., leHmAnn, B., lAFlAmme, J., cABrI, l., crIddle, A. and gAlBIAttI, H., 2002. Hongshiite, PtCu, from itabirite-hosted Au-Pd-Pt mineralization, Itabira district, Minas Gerais, Brazil. he Canadian Mineralogist 40: 711-723. leVIn, V., kotel′nIkoV, P. and kurnABAeV, e., 1986. Find of the natural intermetallide of gold and platinum intermetallide. Izvestiya Akademii Nauk Kazakhskoy SSR Seriya Geologicheskaya. 4: 76-80 (in Russian). lIndgren, w., 1933. Mineral deposits, 4th ed. New York, London, McGraw-Hill Book Company. lozecHkIn, m., 1939. New data about chemical composition “cuprian” gold. Doklady Academy of Sciences of the USSR 24: 454-457 (in Russian). lyAkIscHeV, n.P. (ed.), 1996. 1997, 2001, Diagrams of the state for double metallic systems. Vol. 1: 1996. Vol. 2: 1997. Vol. 3: 2001. Moscow: Maschino stroenie (in Russian). mArkHAm, n., 1960. Synthetic and natural phases in the system Au-Ag-Te. Economic Geology 55: 1148-1178, 1460-1477. nAzmoVA, g. and sPIrIdonoV, e., 1979. Mercurian gold. Doklady Academy of Sciences of the USSR 246: 702-705 (in Russian) nekrAsoV, I., sAmusIkoV, V., leskoVA, n., 1988. First ind of the sulide AuAgS – analog at petrovskaite. Doklady Academy of Sciences of the USSR 303: 944-947 (in Russian). nekrAsoV, I.y., lennIkoV, A.m., oktyABrsky, r.A., zAlIsHAk, B.l. and sAPIn, B.I. 1994. Petrology and platinum minerali- ArcheoSciences, revue d’archéométrie, 33, 2009, p. 67-73 Ernst SPIRIDONOV, Denka YANAKIEVA zation of the alkaline-ultramaic ring complexes, (Laverov, N.P. ed.). Moscow, Russia, Nauka, 381 p. (in Russian). nesterenko, g., kuznezoVA, A. and PAl’cHIk, n., 1984. Petrovskaite AuAg(S,Se) – new selenian sulide of gold and silver. Zapiski Vsesoynznogo Mineralogicheskogo. Obshchestva 113 (5): 602-607 (in Russian). oBertHür, t. and sAAger, r., 1986. Silver and mercury in gold particles from the Proterozoic Witwatersrand placer deposits of South Africa: metallogenic and geochemical implications. Economic Geology 81: 20-31. PAAr, w., Putz, H., toPA, d., roBerts, A., stAnley, c. and culetto, F., 2006. Jonassonite, Au(Bi,Pb)5S4, a new mineral species from Nagybörzsöny, Hungary. The Canadian Mineralogist 44: 1127-1136. PeAcock, m. and tHomPson, r., 1946. Montbrayite, a new gold telluride. American Mineralogist 31: 515-526. PetroVskAyA, n., 1973. Native gold. Moscow, Nauka (in Russian). rAmdoHr, P., 1967. A widespread mineral association, connected with serpentinization. Neues Jahrbuch Mineralogie Monatshefte 107S: 241-265. rAzIn, l. and sIdorenko, g., 1989. Anyuiite AuPb2 – new intermetallide of gold and lead. Mineralogical Journal 11(4): 88-92 (in Russian). sAmusIkoV, V. and nekrAsoV, I., leskoVA, n., 2002.Sulfoselenide (Ag,Au)2(S,Se) from Yakutskoe deposit. Zap. VMO. 131(6): 61-65 (in Russian). sHAngquAn, w., yAng, y. and song, q., 1992. A new gold mineral – hunchunite (Au2Pb). Acta Mineralogica Sinica 12: 319322. sHIkAzono, n. and sHImIzu, m., 1987. he Ag/Au ratio of native gold and electrum and the geochemical environment of gold deposits in Japan. Mineralium Deposita 22: 309-314. sHIkAzono, n., sHImIzu, m., 1988. Mercurian gold from the Tsugu gold-antimony vein deposit in Japan. he Canadian Mineralogist 26: 423-428. sImon, g., Alderton, d. and Bleser, t., 1994. Arsenian nagyagite from Sacarîmb, Romania: A possible new mineral species. Mineralogical Magazine 58: 473-478. smolIn, A., 1970. Nuggets of gold from the Urals. Moscow: Nedra (in Russian). so, c.s., duncHenko, y., yun, s., 1995. Te- and Se-bearing epithermal Au-Ag mineralization, Prasolovskoe, Kunashir Island, Kuril island arc. Economic Geology 90: 105-117. sPIrIdonoV, e., 1991a. Nagiagite AuPb5Te4-xSbxS6. he Fersman Mineralogical Museum, Russian Academy of Sciences 37: 128138 (in Russian). sPIrIdonoV, e., 1991b. About compositions and structures of the minerals bilibinskite – bogdanovite group. Trans. Fersman Mineral. Museum 37: 138-145 (in Russian). Modern mineralogy of gold: overview and new data sPIrIdonoV, e., 1995. Inversion plutonogenic gold-quartz formation in caledonides of the Central Kazakhstan. Geology ore deposits 37: 179-207 (in Russian). sPIrIdonoV, e., 2008. Ferrian bogdanovite Au5CuFe2(Te,Pb)2 from the cementation zone of the Aginskoe gold-telluride deposit, Kamchatka. he Fersman Mineralogical Museum, Russian Academy of Sciences 43: 143-145 (in Russian). sPIrIdonoV, e. and cHVIleVA, t., 1985. Mutmannite AuAgTe2 – new data. Doklady Academy of Sciences of the USSR 280: 994-997 (in Russian). sPIrIdonoV, e. and PletneV, P., 2002. Deposit of the cuprian gold Zolotaya Gora. Moscow: Scientiic World (in Russian). sPIrIdonoV, e., kulAgoV, e. and kulIkoVA, I., 2003. Platinianpalladian tetraauricuprid and associated minerals from Noril’sk-I ore deposit. Geology ore deposits 45: 267-277 (in Russian). sPIrIdonoV, e., FIlImonoV, s. and BzyzgAloV, I., 2009. Solid solution ischesserite –naumannite (Ag,Au)2Se from the vol- 73 canogenic gold deposit Ozernovskoe, Kamchatka. Doklady Academy of Sciences of the USSR 425: 391-394 (in Russian). stIllwell, F., 1931. he occurence of telluride minerals of Kalgoorlie. Proceedings of the Australian Institute for Mining and Metallurgy 84: 115-190. terzIeV, g., 1966. Kostovite, a gold-copper tellurid from Bulgaria. American Mineralogist 51: 29-36. VernAdsky, V.I., 1914. Experience of the descriptive mineralogy. Petrograd, Imperial Academy of Sciences (in Russian). wu, s., 1993. Hypothesis of amalgamation overgrowth of placer gold. 30th International Geological Congress, Beijing 1993. Quaternary Geology, A. Zhisheng, Z. Weijian eds. Vol. 3, VSP. Utrecht, he Netherlands. 3235. yAkuBoVskAyA, n., sPIrIdonoV, e., PonomAreVA, I. and gAPeeV, A., 1989. About “magnetic” native gold. Doklady Academy of Sciences of the USSR 309: 434-437 (in Russian). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 67-73 Bronze Age gold in Southern Georgia L’or en Géorgie du sud à l’Âge du Bronze Andreas Hauptmann* and Sabine Klein** Abstract: his study represents a joint pilot project between archaeological and natural sciences, aimed at exploring the technology and provenance of gold artefacts, exempliied for the case of Georgia. It is focussed on the prehistoric gold mine of Sakdrisi. he study included the investigation of both Bronze Age gold artefacts and native gold by chemical and lead isotope analyses using ICP-mass spectrometry. he analyses of the artefacts indicate that the gold which was used to manufacture the artefacts was alloyed with variable amounts of silver. Beside silver, the concentrations of platinum, osmium, and other trace elements in the gold are low. In addition to the analysis of the chemical composition, lead isotope analyses also provided insights into the provenance of the gold artefacts from the region Georgia-Anatolia-Armenia. No sophisticated treatment of the gold could be detected. Résumé : Ce travail est un projet pilote entre les sciences archéologiques et les sciences naturelles, dont le but est d’explorer les technologies et la provenance d’objets en or, présentés ici dans le cas de la Géorgie et, plus particulièrement, de la mine d’or préhistorique de Sakdrisi. Cette étude comporte l’analyse chimique et isotopique, au moyen de la technique ICP-MS, aussi bien d’objets en or datant de l’Âge du Bronze que d’or natif. L’analyse des objets a montré que l’or utilisé dans leur fabrication était allié à des concentrations variées d’argent. À côté de l’argent, les concentrations de platine, d’osmium et d’autres éléments trace de l’or sont faibles. En plus de l’analyse de la composition chimique, l’analyse des rapports isotopiques du plomb a fourni des indications sur la provenance des objets en or de la région Géorgie-Anatolie-Arménie. Aucun traitement élaboré de l’or n’a pu être identiié. Keywords: Gold, Prehistoric Mining, Silver, Copper, Mercury, Trace elements, Lead isotopy. Mots-clés : or, exploitation minière préhistorique, argent, cuivre, mercure, éléments trace, isotopes du plomb. 1. IntroductIon he region south of the Great Caucasus (present day Georgia) is known from the Greek myth of the “Golden Fleece”. Iason, a Mycenaean hero of royal origin, sailed with the Argonauts from Greece to the Colchis to demand the “Golden Fleece” from king Aietes. Iason successfully looted the leece with the help of the king’s daughter, Medea. he “Golden Fleece” stands as a symbol for the recovery of gold from placers using the skin of an ox or a sheep. he myth of the “Golden Fleece” is proof of the economic wealth of this region and its sources of gold, silver, iron and copper. he richness in gold is veriied extraordinarily by the excavations of 5th to 3rd century BC royal graves of the acropolis at Vani, the capital of the kingdom of Colchis. he graves, which are contemporaneous with the Greek colonisation of the eastern coast of the Black Sea, are under excavation by Kacharava and Kvirkvelia (2008). here is, however, also gold from earlier periods (KuraAraxes Culture) in Georgia. Since the Trialeti Culture (middle of the 3rd millennium BC), gold was more frequently used for jewellery and tomb oferings (Lordkipanidze, 1991). On the occasion of the ongoing excavations by the National Museum of Georgia and the Deutsches Bergbau-Museum Bochum of * Deutsches Bergbau-Museum Bochum – D 44787 Bochum. andreas.hauptmann@bergbaumuseum.de ** Institut für Mineralogie – Altenhöferallee 1, Goethe-University Frankfurt/Main, D 60438 Frankfurt. sabine.klein@kristall.uni-frankfurt.de rec. Oct. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 75-82 76 the prehistoric gold mine of Sakdrisi, ca. 50 km southwest of Tbilisi, close to the centre of the Middle Bronze Age Trialeti Culture, we analysed Bronze Age gold artefacts (beads and sheet metal) and native gold from the Great Caucasus and Transcaucasia. It is the aim of the present study to search for the source(s) of the Bronze Age gold in Georgia: are the artefacts made locally, or were they imported? In addition, we studied the composition of the gold in order to gain an insight into the metallurgical techniques possibly employed for producing the artefacts. he analytical data and the archaeological context will be published in greater detail in Hauptmann et al. (in press) and Stöllner et al. (in press). 2. Gold deposIts In GeorGIa he gold districts in Georgia are shown in Figure 1. hey were partly exploited at the beginning of the 19th century (Godabrelidze, 1933). Today, gold reserves in Georgia are estimated to ca. 100 t (Twaltschrelidze, 2001). (Sub-) recent placers occur in Swanetia and Racha in the Great Caucasus. his gold originates from hydrothermal veins in Cretaceous-Jurassic shales (Godabrelidze, 1933), and is possibly re-mobilised primarily from Proterozoic to Palaeozoic crystalline rocks with a geological age between 60 and > 570 Ma (Adamia, 2004). An estimated 8.000 kg of gold were exploited in this region. From the area north of Tbilisi, fossil placers are known in Pleistocene sediments. hey probably originate from hydrothermal quartz veins in Cretaceous shales, similar to Swanetia. Gold placers in the lower course of the Tschorochi river near Batumi may have their origin in the Artvin area, where copper-gold and silver deposits are under exploitation at Gümüshan and Murgul (Bayburtoğlu and Yildirim, 2008). he gold district of Sakdrisi, west of Bolnisi, is the most important one in Georgia (Fig. 1). Both gold placers and primary deposits occur here. Dozens of ‘ancient’ mines were surveyed. he gold deposits were formed mainly during the Alpidic metallogenesis (Jurassic- Cretaceous) and are part of the “Tethyan Eurasian Metallogenetic Belt” (TEMB) (Moon et al., 2001). hey are embedded in volcanic rocks. In addition, Palaeozoic-Proterozoic granitic rocks occur. Gold also occurs in volcanic rocks in the northern part of Armenia. Important copper bearing volcanogenic massive sulide deposits (VMS) with gold, namely those of David Garedschi and of Madneuli, are genetically connected with Sakdrisi. hey are also geologically comparable to ore deposits in the Artvin (Moon et al., 2001). Madneuli is bound to a rhyolitic dome located above an intrusive complex of granodioritic rocks. he K-Ar dating of Madneuli points to 85-93 Ma, ArcheoSciences, revue d’archéométrie, 33, 2009, p. 75-82 Andreas HAUPTMANN, Sabine KLEIN and the one of Sakdrisi to 78-84 Ma (Gugushvili et al., 2002). Due to its geochemical stability, gold is relatively enriched close to the surface. We therefore suggest that prehistoric mining at the outcrop of Madneuli also involved gold. 3. the prehIstorIc Gold mIne of sakdrIsI At Sakdrisi, a criss-crossing swarm of vertically dipping hydrothermal gold-quartz veins with hematite and minor amounts of copper minerals were exploited (Fig. 2). he ancient mining reached a depth of ca. 25 m. he ample prehistoric mining is attested by the masses of hammer stones collected from the waste dumps, and from the underground mining as well. Backillings left in the ancient working pockets contained Kura-Araxes pottery, and also charcoal from ire setting and from lightening. A set of seven 14C data carried out so far revealed mining activities spanning the period from the second half of the 4th to the middle of the 3rd millennium BC (Stöllner et al., 2008). Gugushvili et al. (2002) estimated an average of 1.03 ppm of gold in the Sakdrisi district. Local enrichments, however, reach > 10 and up to 500 ppm of gold (Hauptmann et al., in press; pers. comm. M. Natsvishvili, M. Tschochonelidze). Our own investigations proved that such enrichments occur in vugs illed with loose material including clay, hematite, and quartz. he gold that was washed from the backillings of the prehistoric mine proved to be a very ine grained material. he grains are hardly visible with the naked eye. It is questionable whether this is the material that was exploited in ancient times. Metallurgical processes that would be necessary to extract the gold from this type of material, such as amalgamation or lead melting, were not known in the Bronze Age. Perhaps local enrichments might have created classical bonanzas (see Stöllner et al., in press). Presently, it is too vague to come up with an estimation of the gold exploited at the ancient mine of Sakdrisi, even if survey work suggests a igure of 5000 m3 of worked rock at or in the mine. 4. analytIcal InvestIGatIon 20 alluvial gold specimens and samples of gold ore from the Sakdrisi mine were washed from several tons of material collected using sluice boxes and pans. In addition, 70 Middle to Late Bronze Age gold artefacts, mainly from Kurgans in Southern Georgia, were sampled. In our study, we analysed artefacts and native gold in a ‘destructive’ manner. Mounted samples were prepared for investigations by SEM for a irst overview. We applied elec- Bronze Age gold in Southern Georgia 77 Figure 1: (See colour plate) Gold districts in Georgia (upper right) and a simpliied geological overview of the Sakdrisi-Bolnisi gold district. A number of primary gold deposits are shown, combined with recent placer deposits. he prehistoric mine of Sakdrisi is closely connected with the volcanogenic massive sulphide deposits of Madneuli. Dozens of ‘ancient’ remains of mining activities were surveyed. he entire region contains mainly Cretaceous to Tertiary volcanic rocks, and Palaeozoic-Proterozoic magmatic rocks (Dambludka, Mamulo). Modiied after the geological map of the Sakdrisi-Bolnisi-Reviers (1:200.000), Georgia, unpublished version. With permission of M. Tschochonelidze, Tbilisi. Abbrevations: A = Artvin; B = Batumi with Tschorochi; S = Swanetia, Racha; SB = Sakdrisi, Bolnisi; T = Tbilisi. Figure 1 : (Voir planche couleur) Districts de l’or de la Géorgie (en haut à droite) et vue géologique simpliiée d’ensemble du district de Sakdrisi-Bolnisi. Certains gisements d’or sont localisés, combinés avec des placers récents. La mine préhistorique de Sakdrisi est très liée au gisement volcanogénique à sulfures massifs de Madneuli. Des dizaines de vestiges anciens d’activités minières ont été récoltés. De plus la région possède particulièrement des roches tertiaires et crétaciques volcaniques et des roches paléozoïque-protérozoïque magmatiques (Dambludka, Mamulo). Modiications selon la carte géologique de Sakdrisi-Bolnisi-Reviers (1:200.000), Géorgie, version non publiée. Avec la permission de M. Tschochonelidze, Tbilisi. Abréviations: A = Artvin; B = Batumi avec Tschorochi; S = Swanetia, Racha; SB = Sakdrisi, Bolnisi; T = Tbilisi. tron microscopy with a wavelength system (JEOL 8900 Superprobe) for the quantiication of the main elements and a selection of trace elements, in order to characterise the bulk chemical composition of the material. his preliminary step is necessary in order to identify the appropriate dissolution factor for solution based ICP-MS and also as a base for the LA calculation of the trace elements. Trace element and lead isotope analyses were carried out in Frankfurt using a Multi-Collector ICP-Mass Spectrometer (Finnigan MAT eNeptun). Generally, due to the high cultural value of the gold objects, destructive sampling such as the one permitted here is an exception. We therefore utilised, for comparison, Laser Ablation ICP-Mass Spectrometry analysis in a nondestructive mode as well. his method is most suitable for analysing objects from which only very little material or even none is available for analysis. In Frankfurt, a UP-213 laser ablation system (New Wave) was used, coupled with a Finnigan Element2 Mass Spectrometer. A measurement method was developed to combine extern standard solutions. Its reproducibility was veriied by multiple measurements (Bendall, 2003). In order to verify the precision of the method, the copper standard SRM C1252 was used. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 75-82 78 Andreas HAUPTMANN, Sabine KLEIN Figure 2: Aerial photograph of the prehistoric mine of Sakdrisi after the removal of the vegetation. Note the exploitation of criss-cross and vertically dipping quartz-hematite-gold veins. Figure 2 : Photographie aérienne de la mine préhistorique de Sakdrisi, après enlèvement de la végétation. L’on note l’exploitation de ilons de quartz-hématite-or en croisillons et descente verticale. With this procedure, a large number of elements can be analysed with a high degree of precision. he typical precision obtained for 207Pb/206Pb = +/- 0.00006 and for 208 Pb/206Pb = +/- 0.0002. he average extern precision is +/- 0.00009 for 207Pb/206Pb and +/- 0.0003 for 208Pb/206Pb. With these results, LA-ICP-MS represents the irst method of choice for ‘non-destructive’ analysis of gold objects. Problems occur if the overall low concentrations of lead in gold are extreme, and also if contaminations of mercury have taken place. his makes a precise analysis of 204Pb diicult, and it prevents its use in some cases. However, important archaeometric information is still available by the analysis of the remaining lead isotopes: 206Pb, 207Pb and 208Pb. We will present the speciicities of the diferent applications of this method in our upcoming paper. 5. results and dIscussIon By comparing the native gold collected from various localities in Georgia and the analysed gold artefacts, it becomes clear that the latter are generally richer in trace elements than the native gold. he native gold samples are, except for their silver concentrations, remarkably pure. Neither PGE inclusions, nor Ag-Au tellurides or Ag sulides were detected. We suggest that the higher trace element content of the artefacts is due to an incomplete separation of the gold from the associated heavy mineral fractions, as observed in ArcheoSciences, revue d’archéométrie, 33, 2009, p. 75-82 Mamulo (Fig. 3). Such impurities could be partly incorporated from the heavy minerals into the gold during the (s) melting process. Silver in gold Both gold artefacts and native gold show a large range of silver contents, from 1 to 38 wt% Ag. Silver is homogeneously distributed within the gold, and no enrichment of gold was observed as it may occur in nature, or caused by anthropogenic treatments. Ag concentrations of this range are not anomalous for native gold, and the match with the silver contents of the artefacts points out that silver was not intentionally added to the gold. If intentionally added, the alloy would be expected to be remarkably contaminated by lead from the cupellation process of Ag-bearing lead ores. Because the lead concentrations range within the lower ppm level (Fig. 4), we strongly exclude the possibility of an intentional adding of silver to the gold. Obviously, no parting was performed during the Middle and Late Bronze Age in Georgia. Mercury – a trappy element Among the gold grains that were washed, e.g., in a small river near Sakdrisi, gold-silver amalgams were identiied. here are two possible explanations for this occurrence: 1. the amalgams are native and occur in the ore deposits, and 2. the amalgams are anthropogenic products and represent 79 Bronze Age gold in Southern Georgia Figure 3: (See colour plate) Alluvial gold, Mamulo, SakdrisiBolnisi-district. he concentrate is contaminated by copper minerals (blue), magnetite (black) and quartz (white). We cannot exclude the possibility that comparable impurities would have been incorporated (in parts) in gold in prehistory. Especially copper minerals could lead to ‘anomalous’ concentrations in the inal object. Width of picture: ca. 2 mm. Figure 3 : (Voir planche couleur) Or alluvionnaire, Mamulo, district de Sakdrisi-Bolnisi. Le concentré est contaminé par des minerais de cuivre (bleu), magnétite (noire) et quartz (blanc). À ne pas exclure la possibilité d’incorporation dans l’or (en partie) d’impuretés comparables pendant la préhistoire. Les minerais de cuivre en particulier pourraient amener à des concentrations « anormales » dans l’objet ini. debris from historical gold washers’ activities. Aspects in the favour of the irst explanation: gold-mercury compounds occur in the deposits of the Sakdrisi-Bolnisi district (Moon et al., 2001). his is not uncommon, as natural gold may contain up to 6% mercury, as shown by the paleo-placer deposits of Witwatersrand (Oberthür and Saager, 1986). Placer gold in Scotland contained up to 8% mercury (Leake et al., 1998). An argument in favour of the second explanation is the fact that historical sources from Georgia report on multiple gold extraction by amalgamation during the 19th century. his seems to have been a widespread technique, and Dilabio et al. (1988) reports on worldwide anthropogenic pollution by mercury in gold placer deposits. he distinct porous texture, as shown in Figure 5, is typical for amalgamated gold, its origin being from the evaporation of mercury following the heating of the gold amalgam. As we could not exclude the possibility of an anthropogenic input of mercury rich mineral in the gold grains from Sakdrisi, ‘suspicious’ grains were sorted out prior to the lead isotope analysis in order to avoid contaminations of the natural gold. Hg-containing grains make the analysis of 204Pb impossible, because the isotopes 204 Hg and 202Hg interfere with 204Pb. Copper in gold Alluvial gold from Georgia contains far below 1% copper, while the gold artefacts contain between 1 and 7.7 wt% Cu. he question is whether copper was incorporated in the metal, due to the incomplete separation of the gold from the placer material, or rather if it was incorporated by a re-melting of the gold objects originally decorated with e.g. granulation, i.e., granules soldered with copper. We exclude this possibility, because granulation was not common in Georgia during this period (Dschaparidze, 2001). We ourselves suggest that copper was enriched in the gold, because the gold Figure 4: Pb/Ag diagram of gold artefacts from Georgia. Very low Pb concentrations and any missing correlation between Pb and Ag indicate that the noble metal was not intentionally added to the gold, but that it originates in the ore deposits. Figure 4 : Diagramme Pb/Ag des objets d’or de Géorgie. Les concentrations très faibles de Pb et l’absence de corrélation entre le Pb et l’Ag indiquent que le métal noble n’a pas été intentionnellement additionné à l’or, mais qu’il appartient au gisement. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 75-82 80 Andreas HAUPTMANN, Sabine KLEIN native gold from Georgia is almost devoid of tin. Based on the available data, we cannot claim with certainty that tin is a tracer for placer gold. Results of lead isotope analysis Figure 5: Gold-amalgam grain from a small river in the SakdrisiBolnisi-district. he Au-Ag-Hg alloy consists of 80-82% Au, 3.55.5% Ag and 14-15% Hg. he porous texture suggests a gold extraction by amalgamation rather than a natural amalgam (which is suggested to occur in this region as well). SEM image, secondary electron mode. Source: Hauptmann and Gambaschidze (2006). Figure 5 : Grain d’amalgame d’or provenant d’un ruisseau du district de Sakdrisi-Bolnisi. L’alliage d’Au-Ag-Hg contient 80-82% Au, 3,55,5% Ag et 14-15% Hg. Sa texture poreuse suggère une extraction d’or par amalgamation plutôt qu’un amalgame naturel (qui peut aussi être présent dans cette région). Image MEB, en mode d’électrons secondaire. Source : Hauptmann and Gambaschidze (2006). grains were irst insuiciently separated from copper mineral grains in the placer material (see above and Fig. 3), and subsequently introduced into the gold metal during the smelting or melting process. Results of trace element analysis Concentrations of other trace elements are very low. Platinum (up to 6 ppm) and osmium (up to 7 ppm) seem to be correlated. Platinum group elements are suggested to be typical for alluvial gold, especially in a luvial environment, and might be incorporated in gold. Gold in hydrothermal veins contain Pt only in the sub-ppb level. Tin reaches 3 – 50 ppm in the gold artefacts. his is in an order of magnitude lower by a factor of one to two than those analysed by Hartmann (1982) in European gold objects. In contrast, ArcheoSciences, revue d’archéométrie, 33, 2009, p. 75-82 Lead concentrations in native gold are in most cases close to the detection limit for isotope analyses; in the artefacts, lead is only up to 2 ppm. Hence, it was not possible to analyse the lead isotope ratios for all samples. he results obtained from gold artefacts, native gold and gold bearing heavy mineral fractions are shown in Figure 6. For comparison, we also included lead isotopes of some copper ores, because of their close paragenesis with gold. Most of the artefacts plot around a mixing line which crosses overlapping isotope ields of ore deposits in the Aegean, Anatolia and Armenia. his means that they are compatible with an origin of the ore deposits in the Mesozoic folded mountain belt of the Pontides and the Transcaucasus. More precisely, we found an accumulation of isotope ratios of artefacts which matches the composition of the Sakdrisi gold itself, and that of the ores from Madneuli. he results are also compatible with copper ores from Murgul, where at least prehistoric copper mining is attested (Lutz, 1990). High 208Pb/206Pb vs. 207Pb/206Pbratios, as represented by native gold samples from Mamulo, Pinezauri and Dambludka in the Sakdrisi-Bolnisi district (see upper right in Fig. 6), originate from geologically very old deposits, as they occur in Proterozoic crystalline rocks (see Fig. 1). Such isotope ratios were observed only in two gold artefacts, originating from a locality situated at a considerable distance from Sakdrisi, in Eastern Georgia. One of the two outliers in the lower left side of the diagram comes close to the isotopic pattern of 4th millennium BC litharge from Fatmali-Kalecik in the upper Euphrates area (unpublished data Bochum). here is evidence of prehistoric silver extraction at Fatmali-Kalecik (Hess et al., 1999), but the ore deposit is also gold bearing. 6. conclusIons Excavations proved that the Sakdrisi gold mine is the oldest in the world. his data shows that during the Early Bronze Age, gold exploitation was by no means conined to the exploitation of placers. Bronze Age gold artefacts in Georgia were made of natural gold-silver alloys. We could not detect exactly how gold was beneiciated, but we suggest that copper concentrations in the artefacts are caused by an insuicient separation of accessory minerals. As to the 81 Bronze Age gold in Southern Georgia Figure 6: (See colour plate) 208Pb/206Pb vs. 207Pb/206Pb-ratios of gold artefacts and native gold from Georgia. For comparison: lead isotope ratios of copper and lead ores and copper and lead-slags and artefacts from the Aegean, Anatolia and Armenia are shown (shaded area; according to Seeliger et al., 1985; Hauptmann et al., 2002; Pernicka et al., 2003; Meliksetian et al., 2003). Data from the VMS coppergold deposits at Madneuli from Dschaparidze (in prep.), and at Murgul from Lutz (1990). Abbreviations: VMS = Volcanogenic Massive Sulphide Deposit; MBA = Middle Bronze Age; LBA = Late Bronze Age. Figure 6 : (Voir planche couleur) Rapports 208Pb/206Pb vs. 207Pb/206Pb pour les objets d’or et l’or natif de Géorgie. Pour comparaison : rapports isotopiques de minerais de cuivre et de plomb, de scories de cuivre et de plomb et d’objets provenant de l’Egée, d’Anatolie et d’Arménie, comme illustré (zone ombrée ; selon Seeliger et al., 1985 ; Hauptmann et al., 2002; Pernicka et al., 2003; Meliksetian et al., 2003). Données des gisements VMS de cuivre-or de Madneuli selon Dschaparidze (en préparation) et de Murgul selon Lutz (1990). Abréviations: VMS = gisement à sulfures massifs volcanogéniques ; MBA = Âge du Bronze moyen ; LBA = Âge du Bronze récent. provenance of the gold artefacts analysed, we conclude that lead isotope analysis is a useful tool for this purpose. he genetic connection of the gold from Sakdrisi with volcanogenic ore deposits (VMS) suggests a comparison with copper ores from such sources as well. However, mercury concentrations have to be carefully analysed in order to verify their anthropogenic or natural origin. We were able to show that many of the gold artefacts from Georgia were most probably produced using gold from deposits located in the mountain range of the TEMB. he isotope data could support the use of local gold ores from the Sakdrisi and the Madneuli mines for the making of the artefacts, provided that mining activities from the Middle and Late Bronze Age will hopefully be discovered in the future. In order to provide more detailed answers to the questions posed by archaeology, we are in need of a more extensive database pertaining to the areas under consideration. Do gold artefacts found, e.g., in a single Kurgan, but with diferent isotopic signatures, indicate the fact that they come from diferent geographic sources? Or were they smelted using gold from one gold district only, such as the one of Sakdrisi-Bolnisi? What is the variety of isotopic compositions in a single district with diferent primary mineralisations and placers? Sampling of more gold artefacts and native gold appears necessary, and additional tracers are needed. Acknowledgements he authors gratefully acknowledge all the persons and institutions involved for providing us with gold objects and for help in extracting gold from the geological context. Our thanks go to the VolkswagenStiftung, Hanover, for inancial support. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 75-82 82 References adamIa, s. (ed.), 2004. Geological Map of Georgia. Tbilisi, Union GEO-ECO. BayBurtoğlu, B. and yIldIrIm, s., 2008. Gold and Silver in Anatolia, in Ü. Yalcin (ed.), Anatolian Metal IV. Der Anschnitt, Beiheft 21. Bochum, Deutsches Bergbau-Museum, 43-53. Bendall, c., 2003. he application of trace element and isotopic analyses to Celtic gold coins and their metal sources. Dissertation, University Frankfurt/Main, Germany. dIlaBIo , r.n.W., neWsome , J.W., m c Ivor, d.f. and loWensteIn, p.l., 1988. he Spherical Form of Gold: Manmade or Secondary? Economic Geology 83: 153-162. dschaparIdze, I., (in prep.). Bronze Production in Georgia (III-I millennia B.C.). Dissertation, Staatliche Ivane Dschawachaschwili-Universität, Tbilisi, Georgia. dschaparIdze, o., 2001. Zur frühen Metallurgie Georgiens vom 3. bis zum 1. Jahrtausend v. Chr, in I. Gambaschidze, A. Hauptmann, R. Slotta, Ü. Yalcin (eds.), Georgien – Schätze aus dem Land des Goldenen Vlies. Bochum, Deutsches Bergbau-Museum, 92-119. GodaBrelIdze, s.a., 1933. Mineral resources in the Socialistic Republic of Georgia. Tilis , Gostehizdat Gruzii “Tehnika da Šroma”, Tbilisi (in Russian). GuGushvIlI, v.I., akhvledIanI, r., natsvlIshvIlI, m. and hart, I., 2002. Two stages of gold mineralization within Sakdrisi deposit (Bolnisi Mining District, Georgia). Geologica Carpathica Special Issue 53: 1-5. hartmann, a., 1982. Prähistorische Goldfunde aus Europa II. Studien zu den Anfängen der Metallurgie 5. Berlin, Mann Verlag. hauptmann, a. and GamBaschIdze, I., 2006. Vom Kupfer zum Gold – Das Deutsche Bergbau-Museum in Georgien, in S. Brüggerhof, M. Farrenkopf, W. Geerlings (eds.), Montan- und Industriegeschichte. Festschrift R. Slotta zum 60. Geburtstag. Paderborn, Schöningh, 129-142. hauptmann, a., Bendall, c., Brey, G., dschaparIdze, I., GamBachIdze, I., kleIn, s., pranGe, m. and stöllner, t., In press. Gold in Georgien. Analytische Untersuchungen an Goldartefakten und Naturgold aus dem Kaukasus und dem Transkaukasus, in S. Hansen, A. Hauptmann, I. Motzenbäcker (eds.), Von Majkop nach Trialeti – Gewinnung und Verbreitung von Metallen und Obsidian in Kaukasien im 4.-2. Jahrtausend v. Chr. Archäologie in Eurasien. hess, k., hauptmann, a., WrIGht, h. and Whallon, r., 1999. Evidence of fourth millennium BC silver production at Fatmali-Kalecik, in h. Rehren, A. Hauptmann, J. Muhly (eds.), Metallurgica Antiqua: in honour of Hans-Gert Bachmann and Robert Maddin. Der Anschnitt, Beiheft 8. Bochum, Deutsches Bergbau-Museum, 57-67. kacharava, d. and kvIrkelIa, G., 2008. Wine, Worship and Sacriice of Ancient Vani. New York, Institute for the Study of the Ancient World. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 75-82 Andreas HAUPTMANN, Sabine KLEIN leake, r.c., chapman, r.J., Bland, d.J., stone, r., cameron, d.G. and styles, m.t., 1998. he origin of alluvial gold in the Leadhills area of Scotland: evidence from interpretation of internal chemical characteristics. Journal of Geochemical Exploration 63: 7-36. lordkIpanIdze, o., 1991. Archäologie in Georgien. Von der Altsteinzeit bis zum Mittelalter. Quellen und Forschungen zur prähistorischen und provinzialrömischen Archäologie. Bd. 5. Weinheim: VCH Verlagsgesellschaft. lutz, J., 1990. Geochemische und mineralogische Aspekte der frühen Kupferverhüttung in Murgul/Nordost-Türkei. Dissertation Nat.Math. Gesamtfakultät, Universität Heidelberg, Germany. melIksetIan, kh., pernIcka, e., avetIssyan, p. and sImonyan, h., 2003. Chemical and lead isotope characterisation of Middle Bronze Age bronzes and some Iron Age antimony objects, in Archaeometallurgy in Europe: international conference, 24-25-26 September 2003, Milan, Italy: proceedings. Milano, Associazione Italiana di Metallurgia, 311-318. mIGIneIshvIlI, I.r., 2002. A possible model of formation for the Madneuli copper-gold deposit. Proceedings of the Geological Institute of the Georgian Academy of Science, New series 117: 472-479. moon, c.J., GotsIrIdze, G., GuGushvIlI, v., kekelIa, m., mIGIneIshvIlI, r., otkhmezurI, z. and özGür, n., 2001. Comparison of Mineral Deposits between Georgian and Turkish sectors of the Tethyan Metallogenic Belt, in A. Piestrzynski (ed.), Mineral Deposits at the Beginning of the 21st Century. Lise, Balkema, 309-312. oBerthür, t. and saaGer, r., 1986. Silver and Mercury in Gold Particles from the Proterozoic Witwatersrand Placer Deposits of South Africa: Metallogenetic and Geochemical Implications. Economic Geology 81: 20-31. pernIcka, e., eIBner, c., öztunalI, ö. and WaGner, G.a., 2003. Early Bronze Age Metallurgy in the North-East Aegean, in G.A. Wagner, E. Pernicka, H.P. Uermann (eds.), Troia and the Troad. Scientiic Approaches. Berlin, Heidelberg, New York, Springer, 143-172. seelIGer, t., pernIcka, e., WaGner, G.a., BeGemann, f., schmItt-strecker, s., eIBner, c. and öztunalI, ö., 1985. Archäometallurgische Untersuchungen in Nord- und Ostanatolien. Jahrbuch Römisch-Germanisches Zentralmuseum 32: 597-659. stöllner, t., GamBaschIdze, I., hauptmann, a., mIndIaschWIlI, G., GoGotschurI, G. and steffens, G., in press. Goldbergbau in Südostgeorgien – Neue Forschungen zum frühbronzezeitlichen Bergbau in Georgien, in S. Hansen, A. Hauptmann, I. Motzenbäcker (eds.). Von Majkop nach Trialeti – Gewinnung und Verbreitung von Metallen und Obsidian in Kaukasien im 4.-2. Jahrtausend v. Chr. Archäologie in Eurasien. tWaltschrelIdze, a.G., 2001. Erzlagerstätten in Georgien, in I. Gambaschidze, A. Hauptmann, R. Slotta, Ü. Yalcin (eds.), Georgien – Schätze aus dem Lande des Goldenen Vlies. Bochum, Deutsches Bergbau-Museum, 78-89. Minéralogie et signature isotopique du plomb des minerais auro-argentifères exploités durant l’époque romaine à Alburnus Maior (Roşia Montană, Roumanie) Mineralogy and lead isotope signature of the gold-silver ores exploited during the Roman period at Alburnus Maior (Roşia Montană, Romania) Călin Gabriel Tămaş*, **, ***, Sandrine Baron*** et Béatrice Cauuet*** Résumé : Alburnus Maior (Roşia Montană, Roumanie), est un site minier important à l’époque romaine pour ses métaux précieux. Des études géologiques, gîtologiques et géochimiques ont été efectuées sur les corps de minerais exploités à l’époque romaine dans les chantiers miniers mis en évidence par l’archéologie minière. La démarche principale réside dans l’étude de minerais géo- et chrono- référencés par l’archéologie et la géologie pour identiier les sources des métaux précieux pendant l’antiquité romaine. Les études géologiques menées dans le massif de Cârnic, ont permis d’identiier quatre phases de minéralisation. Trois d’entre elles ont été exploitées par les Anciens. Chacune se traduit par un corpus d’associations minéralogiques en métaux précieux dont deux d’entre elles possèdent des minéraux rares à tellures et à germanium. Par ailleurs, les teneurs élémentaires des autres éléments chimiques étant très hétérogènes, l’isotopie du plomb (Pb) est l’outil idéal pour optimiser la caractérisation. Chaque phase de minéralisation possède sa propre signature en Pb. Malgré les diférences isotopiques de Pb observées entre chacune d’elles, les signatures en Pb de Roşia Montană (RM) constituent un champ isotopique très restreint. En comparaison avec d’autres districts miniers de Roumanie, le champ isotopique de RM est identiiable et ainera la iliation chimique à échelle régionale dans l’est de l’Europe. Abstract : Alburnus Maior (Roşia Montană, Romania) was an important precious metals mining site during Roman times. Geological, ore deposit and geochemical studies have been carried out on the ore bodies exploited during the Roman period that have been discovered during archaeological excavations. he primary purpose of the present paper is to demonstrate the importance of the study of geologically and chronologically referenced ores by means of mining archaeology and geology for identifying the sources of precious metals from Roman Antiquity. he geological studies carried out in the Cârnic Massif allowed the identiication of four mineralisation phases. hree of these four phases have been exploited by the ancient miners. Each mineralised phase is characterised by a speciic precious metals mineral association, and two of them contain rare Te and Ge bearing minerals. As their elemental composition is heterogeneous, the use of lead isotopes represents a good methodology for reining their characterisation. Each mineralised phase has its own lead signature. In spite of some observed diferences in terms of lead isotopes for each phase, the overall signature of Roşia Montană (RM) displays a well deined isotopic ield. When compared to other mining districts in Romania, the isotopic ield of RM is distinguishable, and this allows a better discrimination that can contribute to the provenance study of minerals from Eastern Europe. Mots-clés : Minerais Au-Ag, minéralogie, géochimie, isotopie, archéologie minière, mine romaine, Roşia Montană. Keywords : Au-Ag ores, mineralogy, geochemistry, isotopes, mining archaeology, Roman mine, Roşia Montană. * Université de Fribourg, Département de Géosciences, 6, Ch. du Musée, 1700 Fribourg, Suisse. ** Université Babeş-Bolyai – Faculté de Biologie et Géologie, 1, M. Kogălniceanu, 400084 Cluj-Napoca, Roumanie. (calin.tamas@ubbcluj.ro) *** Laboratoire Travaux de Recherches Archéologiques sur les Cultures, les Espaces et les Sociétés, Université Toulouse 2 Le Mirail, CNRS UMR 5608 – Maison de la Recherche, 5 allées Antonio-Machado, 31058 Toulouse cedex 09, France. rec. Aug. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 83-89 84 Călin Gabriel Tămaş, Sandrine Baron, Béatrice Cauuet 1. IntroductIon Roşia Montană (RM) est l’ancienne Alburnus Maior, une agglomération minière romaine mentionnée (entre autres) sur la plus ancienne des tablettes cirées (tabulae ceratae) retrouvées sur le site et datée du 6 février 131 (CIL TC XVII, Sîntimbrean, 1989), un quart de siècle après la conquête de la Dacie par les Romains. RM représente un gisement auro-argentifère de très grande taille exploité à ciel ouvert, mais également en souterrain dès l’époque romaine (Cauuet et al., 2003). À l’heure actuelle, RM reste le plus grand gisement de métaux précieux d’Europe (Manske et al., 2006) avec environ 410 tonnes d’or et 1840 tonnes d’argent pour des teneurs limites d’exploitation de 1,3 g/t Au et 6 g/t Ag (www.gabrielresources.com ; Jan 3, 2009). Un projet minier s’intéresse à ces ressources, il s’agit de la compagnie minière roumaine « Roşia Montana Gold Corporation », une iliale de la compagnie canadienne « Gabriel Ressources ». Dans le contexte d’une probable reprise minière du site de RM, des fouilles d’archéologie préventive se sont déroulées dans le cadre du Programme National de Recherche Alburnus Maior, sous la direction du Musée National d’Histoire de Roumanie. Jusqu’à ce jour, d’importantes fouilles archéologiques ont été achevées dans plusieurs secteurs miniers du site de RM (secteurs de Cetate et Cârnic) et d’autres fouilles sont actuellement en cours dans d’autres secteurs miniers de RM (secteurs Ţarina, Orlea, Păru Carpeni, etc.) (Cauuet, sous presse). L’étude présentée ici concerne le massif de Cârnic. Une approche interdisciplinaire a été efectuée dans le cadre des fouilles archéologiques menées à RM. En efet, des études géologiques très détaillées de travaux miniers datant de l’époque romaine ont pu être efectuées grâce aux résultats fournis par l’archéologie minière. Nous avons donc eu la possibilité d’examiner et d’étudier les parois et les fronts de tailles creusés par les mineurs romains et ainsi les corps de minerais exploités réellement par les anciens mineurs. L’étude géologique de terrain a été poursuivie par des analyses minéralogiques et par un cortège d’analyses géochimiques. Le but a été d’établir une caractérisation minéralogique et géochimique de minerais auro-argentifères archéologiques – c’est-à-dire efectivement exploités par les Anciens. 2. LocaLIsatIon géographIque et contexte géoLogIque Le gisement de RM fait partie du « Quadrilatère de l’or » (Ghiţulescu et Socolescu, 1941), une région d’environ 900 km2 située dans la partie sud des Monts Apuseni ArcheoSciences, revue d’archéométrie, 33, 2009, p. 83-89 (Roumanie). Dans cette région, de nombreux gisements de porphyres cuprifères et des gisements épithermaux d’or et d’argent, liés au volcanisme néogène sont répertoriés, RM étant l’un des plus importants. Ce gisement (Fig. 1) est encaissé dans un lysch Crétacé supérieur, pénétré par trois épisodes volcaniques, dont le premier a été responsable de la mise en place des corps de métaux précieux de RM. Plus de détails sur la géologie de la région et sur les particularités du gisement de RM sont reportés ailleurs (Tămaş, 2007). De plus, divers types de corps de minerai sont connus à l’échelle du gisement, il s’agit de ilons, brèches, stockwerks, imprégnations, ainsi que placers et paléoplacers. Du point de vue minéralogique, les métaux précieux se présentent sous forme d’électrum, d’une grande variété de minéraux d’argent, de tellurures et de sulfures communs (Petrulian, 1934 ; Tămaş et al., 2004; 2006 ; Ciobanu et al., 2004). 3. MatérIeLs et Méthodes L’étude géologique a permis d’établir une cartographie géologique détaillée de tous les travaux miniers romains qui ont été fouillés dans le massif de Cârnic. Cette étude a permis d’efectuer la pétrographie des roches encaissantes, d’identiier les corps de minerais exploités par les Romains, ainsi que les altérations hydrothermales associées (notamment l’adularisation et la siliciication), mais également les relations mutuelles entre divers corps de minerais. L’échantillonnage systématique (roches et minerais) et l’analyse des teneurs élémentaires des minerais efectuées au laboratoire Chemex (voir ci-dessous) ont complété l’étude géologique de terrain. L’étude minéralogique des minerais issus de l’exploitation romaine a compris un examen microscopique (lames minces et lames polies), nécessaire pour l’identiication des minéraux porteurs des métaux précieux et de minéraux associés. Des observations et analyses au microscope électronique à balayage (Jeol-JSM 6360 LV) ont permis de détailler la minéralogie des minéraux métalliques. Des analyses quantitatives efectuées sur les associations de minéraux métalliques ont été réalisées à l’aide d’une microsonde électronique de type CAMECA SX50 en utilisant une tension d’accélération de 25 kV, un courant d’analyse de 20 nA et le diamètre du faisceau de 3 x 3 µm2. Des minéraux naturels et synthétiques, ainsi que des métaux purs et des alliages ont servi de matériaux de références. Les analyses chimiques élémentaires ont été efectuées par Chemex, un laboratoire de service analytique de rang international, dont il n’est pas nécessaire de présenter ici toutes les caractéristiques, pour plus d’informations, se reporter au site web (www.alsglobal.com). Minéralogie et signature isotopique du plomb des minerais auro-argentifères exploités durant l’époque romaine… 85 Figure 1 : Géologie simplifiée du gisement de Roşia Montană. Dans l’encadré de la Roumanie, RM correspond au gisement de Roşia Montană et MM correspond au district de Maramureş, la partie ouest représente la région de Baia Mare et la partie est celle de Baia Borşa. Figure 1: Simpliied geology of Roşia Montană (RM). Inside the box representing Romania, RM corresponds to the Roşia Montană ore deposit and MM corresponds to the Maramureş metallogenetic district, which consists of the Baia Mare area in the west and the Baia Borşa area in the east. Les analyses isotopiques ont été conduites selon une méthode de préparation dont tous les détails ont été exposés antérieurement (Baron et al., 2006). L’appareil utilisé est un MC-ICP-MS (Multi-Collector inductively coupled plasma-mass spectrometer, Neptune, hermo scientiic). La méthode de mesure utilisée pour les rapports isotopiques du plomb (Pb) est le dopage au thallium dont la présentation peut se retrouver dans la bibliographie (Baron et al., 2006). Le spectromètre de masse, équipé de neuf cages de Faraday, permet une mesure simultanée de tous les isotopes du plomb, du thallium et du mercure (200Hg). Les mesures répétées du matériau de référence NIST NBS 981 Pb ont permis de calculer les valeurs de reproductibilité suivantes (2* écart type de la moyenne des valeurs du standard NIST NBS 981 Pb) : 93 ppm pour le rapport 208Pb/206Pb, 80 ppm pour le rapport 207Pb/206Pb, 60 ppm pour le rapport 208 Pb/204Pb, 50 ppm pour le rapport 207Pb/204Pb et 100 ppm pour le rapport 206Pb/204Pb. 4. résuLtats Les analyses géologiques et minéralogiques ont permis l’identiication de diférents types de corps de minerais exploités à l’époque romaine dans le massif de Cârnic (ilons et corps de brèches). Ces corps de minerais correspondent à trois grandes phases de déposition de métaux précieux, chacune avec une minéralogie qui lui est propre, par conséquent des teneurs en métaux précieux spéciiques. Une quatrième phase de déposition a également été identiiée ailleurs dans les travaux miniers modernes du massif de Cârnic. Tandis que les trois premières phases ont été exploitées par les Romains, la quatrième phase métallogénique n’a pas été exploitée par les Anciens. Les phases de déposition de minerais se caractérisent pour chacune par (Tableau 1) : Phase #1 – Des structures de brèches phréatiques à ciment hydrothermal riche en quartz. Ce sont des corps de minerais très riches en or et argent (plus d’or que d’argent), avec une ArcheoSciences, revue d’archéométrie, 33, 2009, p. 83-89 86 Călin Gabriel Tămaş, Sandrine Baron, Béatrice Cauuet Phase de minéralisation Composition minéralogique Minéraux métalliques Minéraux de gangue Éléments traces Observations #1 électrum, polybasite, cuivre gris quartz, adulaire et chinga (ciment argentifère, galène argentifère, pyrite, riche en quartz et matière chalcopyrite et sphalérite ; organique) ; - phase exploitée par les mineurs romains #2 électrum, polybasite, pyrite, chalcopyrite, sphalérite, cuivre gris, marcasite et covellite ; - phase exploitée par les mineurs romains #3 achanthite, stéphanite, polybasitepéarcéite, argent natif, cuivre gris argentifère, galène argentifère, pyrite, quartz, adulaire, chinga ; chalcopyrite, sphalérite, bornite, électrum, argyrodite ; traces de Teargyrodite et altaïte ; Ge, Te phase exploitée par les mineurs romains #4 cuivre gris, galène, sphalérite, chalcopyrite, pyrite, hessite (très abondant), altaïte, sylvanite, Teargyrodite, électrum et marcasite. Te, Ge phase non exploitée par les mineurs romains quartz, adulaire ; rhodonite, rhodochrosite, quartz, adulaire. Tableau 1 : Composition minéralogique des quatre phases de minéralisation mises en évidence dans le massif de Cârnic, Roşia Montana, Roumanie. Table 1: Mineralogical composition of the four ores deposition phases from the Cârnic Massif, Roşia Montană, Romania. composition minéralogique dominée par l’électrum et divers minéraux d’argent (sulfosels) avec des sulfures communs. Phase #2 – Des ilons à gangue quartzeuse avec des teneurs très élevées en or et argent, mais avec des valeurs relativement similaires entre ces deux métaux précieux. L’électrum et les minéraux d’argent (sulfosels) sont majoritaires, tandis que les sulfures communs sont très rares. Phase #3 – Des structures de brèches phréatiques rebréchiiées avec l’association de ilons de quartz très riches en argent et des teneurs beaucoup plus faibles en or. La minéralogie de cette troisième phase de déposition est complexe avec une grande variété en sulfosels d’argent, argent natif, sulfure d’argent accompagnés d’électrum et de minéraux en base germanium, à savoir argyrodite (Ag8GeS6) et argyrodite à tellure (Ag8GeTe2S4). Le tellure est systématiquement présent dans le sulfure d’argent (achantite) et la galène. Phase #4 – Des ilons à gangue rhodochrosite-rhodonite, avec plus de 1,15 kg d’argent et quelques grammes d’or à la tonne. La minéralogie de la quatrième phase est complètement diférente des trois premières phases, avec des tellurures d’argent, d’or-argent et de plomb, de l’argyrodite à tellure, des sulfures communs et de l’électrum. Les analyses chimiques élémentaires efectuées sur les minerais de chaque phase de minéralisation montrent des teneurs très hétérogènes, ce qui ne permet pas d’utiliser ces dernières en vue d’un éventuel traçage. Nous ne discuterons donc pas ces données. Néanmoins, les teneurs assez élevées en Te (∼250 ppm) et Ag (10 à > 1 000 ppm) dans les ArcheoSciences, revue d’archéométrie, 33, 2009, p. 83-89 minerais d’or de la phase 2 et 3 seront de bons traceurs élémentaires. En efet, comme il a été démontré par les études minéralogiques, les minéraux porteurs de ces éléments chimiques seraient une particularité des minerais de RM. Les compositions isotopiques du plomb des quatre phases (ϕ) de minéralisations sont reportées dans les graphiques de la Figure 2 (données en cours d’acquisition). Comme tous les rapports isotopiques montrent les mêmes tendances, seul le rapport isotopique 206Pb/204Pb sera pris en compte dans les discussions ain ne pas alourdir le texte. Chaque phase de minéralisation possède un champ de composition isotopique qui lui est propre (en dehors de l’incertitude totale externe des échantillons) et dont les diférences sont signiicatives, à cette échelle (Fig. 2). Par contre, en moyenne, les quatre phases présentent un champ de compositions isotopiques très homogène. Par exemple, les valeurs moyennes, pour le rapport 206Pb/204Pb, et pour chaque phase, sont les suivantes : 18.657 ± 0.007 (1σ : écart type par rapport à la moyenne des valeurs isotopiques d’une phase donnée) ; 18.651 ± 0.003 (1σ) ; 18.657 ± 0.011 (1σ) et 18.635 ± 0.001 (1σ) pour les phases ϕ 1, ϕ 2, ϕ 3 et ϕ 4 respectivement. Les Anciens ayant exploité uniquement trois des quatre phases, nous ne retiendrons que les compositions isotopiques de Pb de RM des phases ϕ 1, ϕ 2 et ϕ 3. Ainsi, la composition isotopique en Pb de Roşia Montană est une moyenne des trois phases, pour chaque rapport isotopique. La signature est, pour l’instant : 2.0778 ± 0.0015 (1σ) pour le rapport 208Pb/206Pb, 0.8395 ± 0.0004 (1σ) pour le rapport Minéralogie et signature isotopique du plomb des minerais auro-argentifères exploités durant l’époque romaine… 87 Pb/206Pb, 38.765 ± 0.023 (1σ) pour le rapport 208Pb/204Pb, 15.663 ± 0.004 (1σ) pour le rapport 207Pb/204Pb et 18.657 ± 0.008 (1σ) pour le rapport 206Pb/204Pb (acquisitions inales des données en cours). 207 5. dIscussIons Les analyses géologiques et minéralogiques de minerais exploités à l’époque romaine indiquent le caractère auroargentifère des deux premières phases de déposition et le caractère argentifère subordonné aurifère de la troisième phase. Dans ces trois phases de déposition, l’or apparaît sous forme d’électrum et il est encaissé dans le ciment hydrothermal des brèches phréatiques, dans le quartz de ilons et de brèches, ainsi que dans la pyrite. Mais, il est rarement inclus dans d’autres minéraux métalliques (sulfures et sulfosels). L’argent est concentré dans l’électrum et les sulfosels d’argent. En ce qui concerne la minéralogie de la troisième phase de déposition, nous devons souligner son caractère particulier en raison de l’occurrence de minéraux de germanium (argyrodite et argyrodite à tellure) et de la grande abondance de sulfosels d’argent avec une présence systématique du tellure. La quatrième phase de déposition, argentifère par excellence, mais non exploitée par les Romains dans le massif de Cârnic, s’individualise par la participation notable des tellurures (Ag, Ag-Au, Pb) et de l’argyrodite à tellure. Les éléments chimiques retenant notre attention sont Ag, Te et Ge. De plus, une étude de fusion simple réalisée sur du minerai de RM (condition oxydante et réductrice avec une température > 1200 °C) a montré que les teneurs en Ag et Te restent relativement identiques entre le minerai de départ et le minerai fondu (Hauptmann et al., 1995). Ag et Te ne semblent pas être afectés par le processus métallurgique. Par conséquent, Ag et Te seraient des éléments « conservatifs » dans ce cas précis, et ils seront de très bons indicateurs, en complément de l’isotopie du plomb, pour les futures études de traçabilité minerais – objets. Figure 2 : Diagrammes Pb/Pb des quatre phases de minéralisations des minerais en Au-Ag du gisement de Roşia Montană, massif de Cârnic (Roumanie). Figure 2: Pb/Pb diagrams of the four Au-Ag ore deposition phases from the Roşia Montană ore deposit, Cârnic Massif (Romania). Les compositions isotopiques du plomb des minerais de RM. Sur ces diagrammes classiques Pb/Pb (Fig. 2), les compositions isotopiques du plomb des minerais de Roşia Montană sont identiiables pour chacune des quatre phases déjà mises en évidence pas les études géologiques et minéralogiques. Chaque phase est l’illustration d’un évènement minéralisateur (mise en place de minerais) dont la source est un réservoir géologique plus ou moins profond et qui possède une composition isotopique qui lui est propre. Les compositions isotopiques sont assez homogènes et, outre le temps de mise en place assez court du district de RM, ArcheoSciences, revue d’archéométrie, 33, 2009, p. 83-89 88 Călin Gabriel Tămaş, Sandrine Baron, Béatrice Cauuet ces signatures en Pb indiquent surtout une source primaire commune (rapport U/h identique et homogène) des différents luides minéralisateurs, ce qui n’est pas forcément le cas dans les gisements métallifères. En efet, un seul et même gisement, ou une seule et même veine minéralisée, peut présenter des hétérogénéités isotopiques, ce qui peut perturber la iliation minerais-objets (Baron et al., 2006 ; 2009). Par conséquent, dans le cadre de problématiques archéologiques et historiques, il est indispensable de devoir caractériser un district minier d’un point de vue archéologique, géologique, mais aussi géochimique ain d’apporter des données pertinentes et complémentaires aux problématiques archéologiques et historiques, notamment sur la provenance des métaux. Le district de RM en contexte régional (à échelle de la Roumanie) Sur le graphique 207Pb/204Pb vs 206Pb/204Pb (Fig. 3), sont reportées les compositions isotopiques du plomb disponibles dans la littérature pour la Roumanie (Bird et al., 2008 ; Marcoux et al., 2002). Quatre groupes s’individualisent : le groupe des Monts Apuseni, les deux groupes de Maramures (celui de Baia Mare et de Baia Borşa) et Roşia Montană. Ce dernier groupe se distingue très bien au milieu du champ de compositions isotopiques de plomb des valeurs de minerais des Monts Apuseni au sens large. Sur le diagramme de la Figure 3, nous pouvons observer que, globalement, les compositions isotopiques de Pb de la région de Maramures, au nord du pays, sont plus radiogéniques, notamment pour le rapport 207Pb / 204Pb, que celles des Monts Apuseni, au sud. La source minéralisatrice qui a conduit à ces gisements de Maramures est donc diférente de celle du secteur des Monts Apuseni. Baia Borşa est le district minier qui possède la signature la moins radiogénique pour le rapport 206Pb / 204Pb et de façon globale, c’est le secteur qui se distingue le plus des autres. Par conséquent, au vu du graphique, chaque secteur minier est individualisable à échelle régionale. Par la suite, la traçabilité sera considérablement ainée dans le Nord/ Nord-Ouest de la Roumanie. Cela est prometteur dans la perspective des futures études de iliations des objets précieux antiques connus en Roumanie. En efet, nous pourrons mieux cerner géographiquement les aires d’extraction des minerais ayant servi à élaborer les objets en or découverts en Europe de l’Est au sens large. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 83-89 Figure 3 : Compositions isotopiques de plomb 207Pb / 204Pb versus 206 Pb / 204Pb des districts miniers de Roşia Montană (inclus dans les Monts Apuseni), Baia Borşa et Baia Mare (région de Maramureş). Figure 3: 207Pb/204Pb versus 206Pb/204Pb ratios of the ore deposits from Roşia Montană (Apuseni Mountains), Baia Borşa and Baia Mare (Maramureş district). 6. concLusIons À Roşia Montană les études géologiques et minéralogiques menées récemment sur les travaux miniers mis au jour par l’archéologie minière ont permis de caractériser de façon précise les corps de minerais exploités pendant l’époque romaine. Sur les quatre phases de minéralisations mises en évidence par la géologie, trois phases ont été exploitées par les Romains. Elles se caractérisent par des teneurs élevées en Ag et Te, dont il sera possible de retrouver les traces (pour Ag et Te) dans les objets précieux prochainement étudiés, étant donné que la réduction simple de minerai n’altère pas, a priori, les teneurs. L’isotopie du plomb, de haute résolution, efectuée sur ces matériaux archéologiques permet de proposer une signature de type « Roşia Montană ». La traçabilité des métaux précieux va être ainée à échelle régionale. La présente étude marque le point de départ de l’établissement d’une banque de données sur des matériaux archéologiques géo- et chrono- référencés. Cette étude est le fruit d’une démarche interdisciplinaire que nous souhaitons rendre systématique. Minéralogie et signature isotopique du plomb des minerais auro-argentifères exploités durant l’époque romaine… Remerciements Nous tenons à remercier la compagnie minière Roşia Montana Gold Corporation (RMGC) pour le soutien inancier apporté non seulement aux études de terrain (exploration, fouilles archéologiques et géologie des travaux romains), mais aussi aux analyses minéralogiques et isotopiques. Le second auteur a bénéicié d’une bourse postdoctorale pour l’étude géochimique et isotopique. Les analyses MEB, microsondes électroniques et isotopiques ont été efectuées au Laboratoire des Mécanismes de Transferts en Géologie (LMTG-CNRS), Université Paul Sabatier, Toulouse, France. Bibliographie BIrd, g., Brewer, p. a., MackLIn, M. g., BaLteanu, d., serBan, M., drIga, B. et ZaharIa, S., 2008. River system recovery following the Novat-Rosu tailings dam failure, Maramures County, Romania, Applied Geochemistry 23: 3498-3518. Baron, s., Mahé-Le-carLIer, c., carIgnan, J. and pLoquIn, A., 2009. Archaeological reconstruction of medieval lead production : Implications for ancient metal provenance studies and paleopollution tracing by Pb isotopes, Applied Geochemistry 24: 2093-2101. Baron, s., carIgnan, J., Laurent, s. et pLoquIn, A., 2006. Medieval lead making on Mont-Lozère Massif (CévennesFrance) : tracing ore sources by using Pb isotopes, Applied Geochemistry 21: 241-252. Cauuet, B., sous presse. L’espace minier romain. Le cas des mines d’or et d’argent d’Alburnus Maior en Dacie romaine (Roşia Montană, Roumanie), Actas del V Congreso de la Sociedad Española para la Protección del Patrimonio Geológico y Minero. Homenaje a Claude Domergue. cauuet, B., anceL, B., rIco, ch. et tăMaş, C., 2003. Ancient mining networks. he French archaeological missions 19992001. In Damian P. (dir.), Alburnus Maior Monographic series, I, Bucharest, p. 467-526. cIoBanu, c.L., cook, n.J, tăMaş, c., Leary, s., Manske, s., o’connor, g. et MInuţ, A., 2004. Telluride-gold-base metal associations at Roşia Montană : the role of hessite as 89 gold carrier. In N.J. Cook, C.L. Ciobanu (ed.). Gold-silvertelluride deposits of the Golden Quadrilateral, South Apuseni Mts., Romania : guidebook of the International Field Workshop of IGCP project 486, Alba Iulia, Romania, 31st August-7th September 2004, (IAGOD Guidebook Series ; 12), 187-202. GhIţuLescu, t.p. et socoLescu, M., 1941. Etude géologique et minière des Monts Métallifères (Quadrilatère aurifère et régions environnantes), Annuaire Institute Géologique Roumain XXI: 181-465. HauptMann, a, rehren, th. et pernIcka, E., 1995. he composition of gold from the ancient mining district of Verespatak/ Roşia Montana, Romania. In G. Morteani, J.P. Northover (ed.), Prehistoric Gold in Europe, p. 369-381. Manske, s.L., hedenquIst, J.w., o’connor, g., tăMaş, c., cauuet, B., Leary, s. et MInuţ, A., 2006. Roşia Montană, Romania : Europe’s largest gold deposit, Society of Economic Geologists Newsletter 64: 1, 9-15. Marcoux, e., grancea, L., LupuLescu, M. et MILésI, J. P., 2002. Lead isotope signatures of epithermal and prophyrytype ore deposits from the Romanian Carpathian Mountains, Mineralium Deposita, 37, p. 173-184. petruLIan, N., 1934. Étude chalcographique du gisement aurifère de Roşia Montană (Transylvanie, Roumanie), Annuaire Institute Géologique Roumain XVI: 499-539. sîntIMBrean, A., 1989. Muzeul mineritului din Roşia Montană, Editura Sport Turism, Bucureşti, 219 p. (en roumain). TăMaş, c.g., BaILLy, L., ghergarI, L., o’connor, g. et MInuţ, A., 2004. First reported tellurides occurrence in Roşia Montană, Apuseni Mountains, Romania, 32nd International Geological Congress Abstract Volume, Florence, Italy, August 20-28: 273. TăMaş, c.g, BaILLy, L., ghergarI, L., o’connor, g. et MInuţ, A., 2006. New telluride and argyrodite occurrences in Roşia Montană, Apuseni Mountains, Romania and their metallogenic signiicance, Canadian Mineralogist, 44: 689-705. tăMaş, c.g., 2007. Structuri de brecii endogene (breccia dyke breccia pipe) şi petrometalogenia zăcământului Roşia Montană (Munţii Metaliferi, România), Ed. Casa Cărţii de Ştiinţă, ClujNapoca, 230 p. (en roumain). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 83-89 he earliest gold objects in Italy: a review of the archaeological evidence Les plus anciens objets d’or découverts en Italie : bilan des données archéologiques Giovanna Bergonzi* Abstract: his paper aims to provide updated information on prehistoric gold from Italy, originating from pre-Iron Age contexts. he evidence dating from the beginning of the Metal Age is scanty indeed: to date, only 4 objects appear to be from an earlier period than the Middle Bronze Age. More gold inds were discovered from the MBA onwards, in part from funerary contexts, representing property or funerary gifts pertaining to individuals of both sexes who attained a high status within the community. Résumé : Le but de ce travail est de fournir des informations mises à jour sur les objets d’or préhistoriques découverts en Italie et datant des périodes précédent l’Âge du Fer. Les objets datant du début de l’Âge des Métaux sont très peu nombreux : jusqu’à présent seulement quatre objets semblent dater d’avant l’Âge du Bronze moyen. A partir de cette période, les objets d’or, parfois trouvés en contexte funéraire sous forme de parure ou d’ofrande, deviennent plus nombreux. Ces parures ou ofrandes accompagnent seulement quelques individus, hommes ou femmes, dont le niveau social était élevé. Keywords: Italy, prehistoric, gold. Mots-clés : Italie, préhistorique, or. 1. INTRODUCTION Gold is scarce in prehistoric contexts from Italy: objects are very few, and the quantity of metal very small. hat is very likely one of the reasons why Italy was not included in the SAM gold project (Hartmann, 1970; 1982), while it had been included in a previous project concerning copper / bronze inds (Junghans et al., 1960; 1968). A review of the earliest gold inds from central Italy was published at the time by an archaeologist (von Hase, 1975). Later on, Bronze Age gold objects were neither included in projects on Bronze Age metalwork (Artioli et al., 2005), nor in studies concerning ancient gold from Italy, which were focused on later and more attractive inds (e.g. Cristofani and Martelli, 1983). In 1995, M. Primas put on a map of Europe the gold and silver objects dated earlier than 2600 BC and observed that, while in some areas silver was more ancient, in others it was gold (Primas, 1995; 1996). In Italy, silver metallurgy is the more ancient one: until today, several ancient small silver inds were identiied on the coasts of the Tyrrhenian Sea and in Sardinia (Bergonzi forthcoming), while early gold objects are very limited in number and dating to a later period, very likely not earlier than the second half of the third or the beginning of the second millennium BC. - rec. Oct. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 91-97 Giovanna BERGONZI 92 his review article aims to provide a synthetic but comprehensive description of the prehistoric gold inds and of the archaeological contexts they originate from, as well as to investigate how they changed over time. 2. METHODS In order to understand the role played by gold objects for Bronze Age society, the various types of objects are briely discussed and archaeological contexts are taken into consideration as well. hese data and the geographical provenance are sketchily summarised in Figures 1-3. 3. RESULTS AND DISCUSSION During the period between the Copper Age (hereafter CA) and the Early Bronze Age (hereafter EBA), approximately 3500-1700 BC, gold inds are scanty indeed (List 1, Fig. 1). In Sardinia, a small electrum torc with open ends was discovered in a CA/EBA megalithic grave at Bingia ‘e Monti, Gonnostramatza, containing multiple burials and, among other grave goods, also several bell beakers (List 1, n. 3). In Northern Italy, two small gold wire spiral rings were found in an EBA fossa grave cemetery near Verona, at Gazzo Veronese, under the skulls of two skeletons; they were also part of a headdress (List 1, n. 2). he fossa tombs appear to contain either poor or no grave goods at all, but might have also been looted, a frequent practice in contemporary Austrian cemeteries where similar gold inds were common (NeugebauerMaresch and Neugebauer, 1988/89). In the lake dwellings at Lavagnone a curved gold wire, 3 cm long, with one end found broken, was identiied in a late EBA stratigraphic unit; it “might have been part of a double spiral pendant” (List 1, n. 1). In the south of Italy, a tiny blue glass bead with gold foil from a settlement at Mursia on Pantelleria Island probably belongs to the irst half of the second millennium BC Figure 1: Gold inds before the Middle Bronze Age. Full symbols: objects from tombs. Figure 1 : Trouvailles datant d’avant l’Âge du Bronze moyen. Symboles pleins : objets provenant de tombes. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 91-97 he earliest gold objects in Italy: a review of the archaeological evidence (List 1, n. 4). hese small objects – a torc, two hair spirals, a possible pendant, one bead – are adornments mostly worn on the head or around the neck by high status individuals, and might have been imports; they do not appear to be in the vicinity of mineral deposits (e.g. Piana Agostinetti et al., 1995; Lehrberger, 1995). During the time span between the Middle Bronze Age (hereafter MBA) and the early phase of the Late Bronze Age (hereafter LBA), approximately 1700-1200 BC, the indings of gold objects increase in number (List 2, Fig. 2). Such objects are found mainly in two areas, northern Italy, in and around the Po plain, where gold objects originate primarily from settlements, and eastern Sicily, where gold inds came to light in chamber tomb cemeteries and are in part regarded as likely imports from the Aegean, while those in northern Italy might come from the north, possibly from the Alps. he most frequent gold objects are rings, and spiral rings are especially common in the Italian peninsula; the largest spiral 93 rings (List 2, n. 4) are regarded as a pair of arm rings. Quite a few of these artefacts are inger rings, like those from Sicily, plain bands, or rings with a middle expansion imitating a signet ring. In Sicily, gold foil pendants and beads, which may have been imports, were discovered, as well as local daggers with a small gold rivet in the hilt. All these objects were very likely worn by individuals who belonged to the dominating elite, not only on the head or around the neck, but also on the hands (inger rings), or on symbols of power, like the daggers (rivets). Quite diferent was the function of the gold foil discs, which were not pendants but had to be aixed on a support. hey very likely were solar symbols used in ceremonies, possibly religious ones, and they were compared to the well known Trundholm chariot (e.g. Bettelli, 1997). During the late phase of the Late Bronze Age (the socalled ‘Bronzo inale’ or Final Bronze Age), approximately 1200-950 BC, while there were still several inds in north- Figure 2: Gold inds from the Middle Bronze Age and early phase of Late Bronze Age. Full symbols: objects from tombs. Figure 2 : Trouvailles datant de l’Âge du Bronze moyen et de la première phase de l’Age du Bronze inal. Symboles pleins : objets provenant de tombes. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 91-97 Giovanna BERGONZI 94 ern Italy and in eastern Sicily, by this time gold objects had become more widespread: such artefacts were also found in central and southern Italy, and in Sardinia (List 3. Fig. 3). Most of them were found in tombs rather than settlements. To a great extent, they still represent the same typologies: rings and spiral rings are the most frequent ones, followed by gold rivets in the hilts of swords or daggers, and gold foil objects, among which the repoussé decorated gold foil discs are the most remarkable. During this time span, the link with the dominating elite appears to be especially evident. Tiny quantities of gold could be worn both by males, such as the warrior from the Frattesina Le Narde tomb 227 (List 3, n. 4A), who, alone among over 500 burials, had a gold ring, gold rivets in the sword hilt and buttons with gold inlays, and by females, such as the lady from the Morano tomb 12 (List 3, n. 1), who, among several other objects, had a tiny open ended gold ring, which according to the excavators might have been an earring. Several gold objects were found in two extraordinary settlements on the Adriatic coast, which were very likely ‘emporia’: Frattesina (List 3, n. 4) in the north and Roca Vecchia (List 3, n. 8) in the south. In Frattesina, in addition to the already mentioned objects from the Le Narde tomb 227, a gold foil disc and several gold foil fragments were found in the settlement area. Yet more spectacular are the inds from Roca Vecchia: at least four gold foil discs, forming at least two pairs, three spiral rings, and one gold foil bead were found on the loor and in a small hole under the loor of a large building (approximately 40 x 15 m), which the excavators interpreted as a ‘hut- temple’. In any case, when considered in its entirety, this most spectacular of all Italian inds weighed little more than 46 g. Metal quantities are also very small: the CA/EBA Sardinian torc (List 1, n. 3) weights less than 12 g; an MBA spiral ring from Fiavè (List 2, n. 3) only 1 g. Nevertheless, the symbolic value of gold is evident: it is linked to high status individuals, but also, during the LBA, to ceremonies, possibly reli- Figure 3: Gold inds from the late phase of Late Bronze Age. Full symbols: objects from tombs. Figure 3 : Trouvailles datant de la in de l’Âge du Bronze inal. Symboles pleins : objets provenant de tombes. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 91-97 he earliest gold objects in Italy: a review of the archaeological evidence 95 gious ones, and to ‘emporia’ which had intensive exchanges with the Aegean area, such as Frattesina and Roca Vecchia. 13B) Orsi, 1912: tab. VI, 23; Peroni, 1956: ig. 9, 142 S.O. 14) Orsi, 1912: col. 373, tab. XVII, 1. 4. REFERENCES FOR FIGURES 1-3 Acknowledgements List 1, Figure 1: LCA – EBA: 1) de Marinis, 2000: 116, tav. XIX, b. 2) Salzani, 1998/1999: igs. 3; 4; 7, 6; 8, 1. 3) Atzeni E. 1998: 258, igs. 8, 3; p. 314, n. 143, ig. 31; Atzeni C. et al., 2003: 103-104. 4) Giardino and Merkouri, 2008: 118. his contribution is part of a long standing project with several participants coordinated by P. Piana Agostinetti (Piana Agostinetti et al., 1995); to her I owe many stimulating discussions and many thanks. We are grateful to R. Peroni, who brought to our attention a still unpublished spiral (List 3, n. 6). Many thanks to G. Montali for his unfailing friendly help with the illustrations. List 2, Figure 2: MBA to early LBA: 1) Bertone et al. 2004: 67, ig. 37. 2) Perini, 1987: 36, ig. 15b. 3) Perini, 1987: 36, ig. 15a. 4) Caldiero, Rocca di Caldiero (Verona): Aspes and Salzani, 1997: 538, ig. 1; 540; 552, nr. 28-29. 5) Guerreschi et al., 1985: 87, tab. XXIII. 6) Peschiera: A violin bow bronze ibula plated with gold foil: von Hase 1975: 101, n. 57, ig. 4. 7) Aspes, 1987. 8) Mutti et al., 1988: 161. 9) Bermond Montanari, 1990: 72; Bettelli, 1997: 733, ig. 732a, b. 10) Peroni, 1963; Bietti Sestieri, 1973; Bettelli, 1997: 734-735. 11).Giardino, 1998: 160, ig. 6; Marazzi, 1999, 420, ig. 7. 12) Voza, 1973; Voza, 1997: 174. 13A) Orsi, 1899: coll. 46-47, tab. VII, 17; Peroni, 1956: ig. 3; Procelli et al., 2004: igs. 2, 7. 13B) Orsi, 1899: coll. 55-56, tab. VIII, 9. 13. 16; Peroni, 1956: ig. 6; Procelli et al., 2004: ig. 2, 2-6. 14) Orsi, 1904: 86, ig. 42. Two more inger rings lack precise dating: Orsi, 1904: 73, ig. 9; 77, ig. 22. 15A) Panvini, 1997: 501, ig. 20. 15B) Panvini and Pappalardo, 2005. Other rings might go back to the early LBA: Ribera (Agrigento), contrada Anguilla, tholos tomb 6, two inger rings (Panvini Di Stefano, 1986: 114); Milena (Caltanissetta), Monte Campanella, a wire ring and a band ring (Militello, 1992: 17). List 3, Figure 3: LBA, late phase. 1) Venturino Gambari and Luzzi, 1999: 122, n. 28, igs. 103, 21. 2) Salzani, 1978: 119. 3) Bianchin Citton, 2003: 28, ig. 4. 4A) Salzani, 1990: 16, igs. 16, 3, 5; 4, 6, 7, 9. 4B) Salzani, 2003: 44, ig. 4, A and B. 5) Bettelli, 1997: 734, ig. 432, a and b. 6) Allumiere (Roma). A spiral wire from a burial urn (excavations by R. Peroni). 7) Müller-Karpe 1959: 103, tab. 12, 7; Gierow, 1966: 360-361. 8) Ugas, 1993: 30, ig. a, 25. 9) Lo Schiavo and Usai, 1995: 174, ig. 19, 2-4; Lo Schiavo et al., 2004. 10) Bozzini et al., 2003; Guglielmino, 2003; Pagliara, 2003; Butalag et al., 2004; Pagliara and Guglielmino, 2005). 11) Pacciarelli, 2001: 192 ss., igs. 111, B and 114, 1 12) Bernabò Brea and Cavalier, 1960: 117-119, tav. XLI, 3). 13A) Orsi, 1899: tab. VIII, 15; Peroni, 1956: ig. 9, 62. References ARTIOLI, G., BIETTI SESTIERI, A.M., CARDARELLI, A., DE MARINIS, R.C. and GIARDINO, C., 2005. Archeometallugia dell’Italia Protostorica. Progetto di ricerca dell’Istituto Italiano di Preistoria e Protostoria, at http://win.iipp.it/progetti1/archeometallurgia (13.06.2005). ASPES, A., 1987. Franzine Nuove di Villabartolomea, in Prima della Storia. Inediti di 10 anni di ricerche a Verona. Verona: Museo Civico di Storia Naturale di Verona, 99-102. ASPES, A. and SALZANI, L., 1997. Gli oggetti ornamentali nella regione Veneta nella preistoria, in L. Endrizzi, F. Marzatico (eds.), Ori delle Alpi. Trento: Provincia autonoma di Trento, Servizio Beni Culturali, 538-541; 552, nn. 28-29. ATZENI, C., CICCOTTI, A., MASSIDA, L. and SANNA, U., 2003. Prenuragic copper, arsenic, silver and gold, in Archaeometallurgy in Europe. International Conference, 24-26 September 2003, Milan, Italy: proceedings, vol. 2. Milano, Associazione Italiana di Metallurgia, 97-106. ATZENI, E., 1998. La tomba ipogeico-megalitica di Bingia ‘e Monti, in F. Nicolis, E. Mottes (eds.), Simbolo ed enigma. Trento, Provincia autonoma di Trento, Servizio Beni Culturali, 254-260. BERGONZI, G. forthcoming, L’argento più antico: metallurgia dell’argento tra la Sardegna e la penisola italiana, Atti della XLIV Riunione Scientiica dell’Istituto Italiano di Preistoria e Protostoria. BERMOND MONTANARI, G., 1990. La lamina aurea di Redù, in Nonantola. Antiquarium: tracce del tempo. Nonantola: Comune di Nonantola, 69. BERNABÒ BREA, L. and CAVALIER, M., 1960. Meligunis Lipara I. Palermo: Flaccovio. BERTONE, A., CARANZANO, S. and ROSSI, P., 2004. Dalla pietra ai metalli, in A. Bertone, F. Fozzati (eds.), La civiltà di Viverone. Biella: Eventi e progetti, 53-69. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 91-97 96 BETTELLI, M., 1997. Elementi di culto nelle terramare, in M. Bernabò Brea, M. Cremaschi, A. Cardarelli (eds.), Le Terramare. Milano, Electa, 720-741. BIANCHIN CITTON, E., 2003. Montagnana – Borgo San Zeno (PD): l’abitato e le sue necropoli, in L. Malnati, M. Gamba (eds.), I Veneti dai bei cavalli. Treviso: Canova, 28-29. BIETTI SESTIERI, A.M., 1973. he metal industry of continental Italy, 13th-11th century and its Aegean connections. Proceedings of the Prehistoric Society 39: 383-424. BOZZINI, B., GIOVANNELLI, G., PAGLIARA, C., GUGLIELMINO, R. and MAGGIULLI, G., 2003. Ceremonial Gold Artefacts of the Late Bronze Age from the Deposition Site of Roca: a Metallurgical Investigation, in Archaeometallurgy in Europe. International Conference, 24-26 September 2003, Milan, Italy: proceedings, vol. 1. Milano, Associazione Italiana di Metallurgia, 267-276. BUTALAG, K., DEMORTIER, G., QUARTA, G., MUSCOGIURI, D., PAGLIARA, C., MAGGIULLI, G. and CALCAGNILE, L., 2004. Checking the homogeneity of gold artefacts of the inal bronze age found in Roca Vecchia, Italy by Proton Induced X-Ray Emission, in Program and Book of abstracts: 8th European Conference on Accelerators in Applied Research and Technology (= ECAART). Paris: 116. CAUUET, B., 1999. L’or dans l’Antiquité, De la mine à l’objet, Aquitania, Suppl. 9. CRISTOFANI, M. and MARTELLI, M., 1983. L’oro degli Etruschi. Novara: Istituto Geograico De Agostini. DE MARINIS, R.C., 2000. Il museo civico Giovanni Rambotti: una introduzione alla preistoria del lago di Garda. Desenzano del Garda: Comune di Desenzano del Garda. GIARDINO, C., 1998. I metalli nel mondo antico. Roma and Bari, Laterza. GIARDINO, C. and MERKOURI, C., 2008. Greece and Southern Italy: the precious connection, in S.A. Paipetis, C. Giannopoulou (eds.), Cultural cross fertilisation of Southern Italy and Western Greece through history. Patras: N.E.O. Patron-Athinon, 108128. GIEROW, P.G., 1966. he Iron Age culture of Latium 1. Acta Instituti Romani Regni Sueciae. Series prima in 4°. Lund, C.W.K. Gleerup. GUERRESCHI , G., L IMIDO , C. and CATALANI , P., 1985. L’insediamento preistorico dell’Isolone del Mincio (Volta Mantovana): scavi Mirabella, Rittatore, Zorzi, 1955-56. Cavriana, Museo archeologico dell’Alto Mantovano. GUGLIELMINO, R., 2003. Il sito di Roca Vecchia: testimomianze di contatti con l’Egeo, in F. Lenzi (ed.), L’archeologia dell’Adriatico dalla Preistoria al Medioevo. Firenze, All’ insegna del giglio, 91-119. HARTMANN, A., 1970. Prähistorische Goldfunde aus Europa I. Studien zu den Anfängen der Metallurgie 3. Berlin, Mann. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 91-97 Giovanna BERGONZI HARTMANN, A., 1982. Prähistorische Goldfunde aus Europa II. Studien zu den Anfängen der Metallurgie 5. Berlin, Mann. VON HASE, F.W., 1975. Zur Problematik der frühesten Goldfunde in Mittelitalien. Hamburger Beiträge zur Archäologie 5: 99-182. JUNGHANS, S., SANGMEISTER, E. and SCHRÖDER, M., 1960. Metallanalysen kupferzeitlicher und frühbronzezeitlicher Bodenfunde aus Europa, Studien zu den Anfängen der Metallurgie 1. Berlin, Mann. JUNGHANS, S., SANGMEISTER, E. and SCHRÖDER, M., 1968. Kupfer und Bronze in der frühen Metallzeit Europas. Studien zu den Anfängen der Metallurgie 2. Berlin, Mann. LEHRBERGER, G., 1995. he gold deposits of Europe: An overview of the possible metal sources for prehistoric gold objects, in G. Morteani, J.P. Northover (eds.), Prehistoric Gold in Europe. Mines, Metallurgy and Manufacture. Dordrecht, Boston, London, Kluwer Academic Publishers, 115-144. LO SCHIAVO, F. and USAI, L., 1995. Testimonianze culturali di età nuragica: la grotta Pirosu in località su Benatzu di Santadi, in V. Santoni (ed.), Carbonia ed il Sulcis. Archeologia e territorio. Oristano, Editrice S’Alvure, 147-186. LO SCHIAVO F., ANTONA A., BAFICO S., CAMPUS F., COSSU T., FONZO O., FORCI A., GARIBALDI P., ISETTI E., LANZA S., LEONELLO V., PERRA M., PUDDU M. G., RELLI R., ROSSI G., SANGES M., USAI A. and USAI L. 2004. Articolazioni cronologiche e diferenziazioni locali . La metallurgia, in D. Cocchi Genick (ed.), L’età del Bronzo recente in Italia, Viareggio (Lucca), Mauro Baroni, 357-382. MARAZZI, M., 1999. I Micenei a Vivara o i Micenei di Vivara? Un bilancio delle ricerche a vent’anni dall’inizio degli scavi, in V. La Rosa, D. Palermo, L. Vagnetti (eds.), Epì Ponton Plazomenoi. Roma, Scuola archeologica italiana di Atene, 415-421. MILITELLO, P., 1992. Due anelli d’oro dalle pendici sud-ovest di Monte Campanella. Quaderni dell’Istituto di Archeologia della Facoltà di Lettere e Filosoia dell’Università di Messina 6: 17-21. MÜLLER-KARPE, H., 1959. Vom Anfang Roms. Heidelberg, Kerle. MUTTI, A., PROVENZANO, N., ROSSI, M.G. and ROTTOLI, M., 1988. La terramara di Castione dei Marchesi. Studi e Documenti di Archeologia 5. Bologna, Nuova Alfa Editoriale. NEUGEBAUER-MARESCH, C. and NEUGEBAUER, J.W., 1988/1989. Goldobjekte aus den Frühbronzezeitnekropolen Franzenhausen I und II und Gemeinlebarn F. Mitteilungen der Anthropologischen Gesellshaft in Wien 118/119: 101-134. ORSI, P., 1899. Pantalica e Cassibile. Monumenti Antichi dei Lincei 9: 33-149. ORSI, P., 1904. Siculi e Greci a Caltagirone. Notizie degli Scavi 5: 65-98. ORSI, P., 1912. La necropoli sicula di Pantalica e la necropoli sicula di Monte Dessueri. Monumenti Antichi dei Lincei 21: 301-408. he earliest gold objects in Italy: a review of the archaeological evidence PACCIARELLI, M., 2001. Dal villaggio alla città. Firenze, All’Insegna del Giglio. PAGLIARA, C., 2003. Il sito di Roca Vecchia nell’età del Bronzo, in F. Lenzi (ed.), L’archeologia dell’Adriatico dalla Preistoria al Medioevo. Firenze, All’Insegna del Giglio, 74-90. PAGLIARA, C. and GUGLIELMINO, R., 2005. Roca: dalle curiosità antiquarie allo scavo stratigraico, in S. Settis, C. Parra (eds.), Magna Graecia. Archeologia di un sapere. Milano, Electa, 298321. PANVINI DI STEFANO, R., 1986. La necropoli preistorica di contrada Anguilla di Ribera, in M. Marazzi, S. Tusa, L. Vagnetti (eds.), Traici micenei nel Mediterraneo. Taranto, Istituto per la storia e l’archeologia della Magna Grecia, 113-122. PANVINI, R., 1997. Osservazioni sulle dinamiche formative socioculturali a Dessueri, in S. Tusa (ed.), Prima Sicilia. Alle origini della società siciliana 1. Palermo, Ediprint, 493-502. PANVINI, R. and PAPPALARDO, G., 2005. Due pugnaletti con manico d’avorio dalla necropoli protostorica di Dessueri, in L. Vagnetti, M. Bettelli, I. Damiani (eds.), L’avorio in Italia nell’età del bronzo. Roma, CNR – ICEVO, 105-110. PERINI, R. 1987. Scavi archeologici nella zona palaitticola di FiavéCarera II. Trento, Servizio Beni Culturali della Provincia Autonoma di Trento. PERONI, R., 1956. Per una distinzione in fasi della necropoli del II periodo siculo a Pantalica. Bullettino di Paletnologia Italiana 65: 387-432. PERONI, R., 1963. Ripostigli delle età dei metalli 3. Ripostigli dell’appennino umbro-marchigiano. I, 6 – I, 8, Inventaria Archaeologica Italia fasc. 3. Firenze, Sansoni. PIANA AGOSTINETTI, P., BERGONZI, G., CATTIN, M., DEL SOLDATO, M., GAMBARI, F.M. and TIZZONI, M., 1995. Gold in the Alps: A View from the South, in G. Morteani e J.P. Northover (eds.), Prehistoric Gold in Europe. Mines, Metallurgy and Manufacture. 97 Dordrecht, Boston, London, Kluwer Academic Publishers, 199-218. PRIMAS, M., 1995. Gold and silver during the 3rd Mill. cal. BC, in G. Morteani and J.P. Northover (eds.), Prehistoric Gold in Europe. Mines, Metallurgy and Manufacture, Dordrecht, Boston, London, Kluwer Academic Publishers, 77-93. PRIMAS, M., 1996. Velika Gruda I. Bonn, Habelt. PROCELLI, E., ALBANESE PROCELLI, R.M. and PANVINI, R. and TUSA, S., 2004. Sicilia. Le necropoli, in D. Cocchi Genick (ed.), L’età del Bronzo recente in Italia. Viareggio (Lucca), Mauro Baroni, 335-346. SALZANI, L., 1978. La necropoli dell’età del Bronzo a Fontanella Mantovana. Preistoria Alpina 14: 115-162. SALZANI, L., 1990. Necropoli dell’età del Bronzo Finale alle Narde di Fratta Polesine. Prima nota. Padusa 25: 5-42. SALZANI, L., 1998/1999. Sepolture dell’antica età del Bronzo da Valserà di Gazzo Veronese (VR). Padusa 34/35: 63-75. SALZANI, L., 2003. Recenti rinvenimenti nel Polesine. Fratta Polesine. Il ripostiglio n. 4 e altri reperti da Frattesina. Quaderni di Archeologia del Veneto 19: 40-45. UGAS, G., 1993. La metallurgia del piombo, dell’argento e dell’oro nella Sardegna prenuragica e nuragica, in T.K. Kirova (ed.), L’uomo e le miniere in Sardegna. Cagliari, Edizioni della Torre. VENTURINO GAMBARI, M. and LUZZI, M., 1999. Catalogo, in M. Venturino Gambari (ed.), In riva al iume Eridano. Una necropoli dell’età del Bronzo inale a Morano sul Po. Torino, Edizioni dell’Orso, 103-141. VOZA, G., 1973. hapsos, in Archeologia della Sicilia Sud-Orientale. Napoli, Centre Jean Berard, 30-52. VOZA, G., 1997. Schede del Catalogo in S. Tusa (ed.), Prima Sicilia. Alle origini della società siciliana 1, Palermo, Ediprint, 166-175. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 91-97 Gold in early Southeast Asia L’or dans le Sud-Est asiatique ancien Anna T. N. Bennett * Abstract: Gold is fairly widely, though irregularly, distributed throughout Southeast Asia in igneous and metamorphic hard rock deposits and in sedimentary placer deposits. he region was known to the Indian merchants of the 1st millennium BPE as Suvarnabhumi: ‘Land of Gold’, which is thought to refer to the mainland, including lower Burma and the hai Malay Peninsula, and Suvarnadvipa: ‘Islands of Gold’, which may correspond to the Indonesian Archipelago, including Sumatra. he historical sources inform us that the Chinese were clearly impressed by the quantities of gold present, and there is evidence to suggest that the gold deposits were one of the stimulating factors in the development of early contacts with India and China. Gold irst appears in the archaeological record in 400 BPE, at about the same time as iron, semiprecious stone polishing and glass working, suggesting that the techniques of gold extraction and working were quite plausibly introduced to Southeast Asia via Indian and/or Chinese merchants seeking gold ores. Résumé : L’or est d’une façon irrégulière plutôt abondant dans le sud-est de l’Asie, grâce à la présence de gisements de roches métamorphiques et ignées et à des placers de nature sédimentaire. La région était désignée par les marchants Indiens du Ier millénaire comme Suvarnabhumi : ‘Terre de l’Or’, ce qui doit correspondre à la partie continentale, comprenant la Birmanie et la Péninsule de hai Malay, et comme Suvarnadvipa : ‘Îles de l’Or’, ce qui peut correspondre à l’Archipel Indonésien, comprenant Sumatra. Les sources historiques nous renseignent que les Chinois ont été clairement impressionnés par les quantités d’or de ces régions, et des évidences suggèrent que les gisements d’or étaient un des facteurs stimulant les premiers contacts entre l’Inde et la Chine. L’or apparaît pour la première fois cité dans les sources archéologiques en 400 BPE, sensiblement en même temps que le fer, le polissage de pierres semi-précieuses et le travail du verre, ce qui suggère que les techniques d’extraction et de travail de l’or ont été vraisemblablement introduites dans le sudest de l’Asie par des marchants Indiens et/ou Chinois à la recherche de minerais d’or. Keywords: Gold, Southeast Asia, Historical sources, Archaeological evidence, Maritime trade, 1st millennium B.P.E. Mots-clés : or, Asie du sud-est, sources historiques, évidence archéologique, commerce maritime, Ier millénaire BPE. 1. DISTRIBUTION OF GOLD ORES IN SOUTHEAST ASIA he words ‘gold’ and ‘golden’ are often associated with place names throughout the Southeast Asian region – Ptolemy’s Geography refers to the ‘Golden Khersonese’, and an early name for hailand is Laem hong, which translates as ‘golden peninsula’. Gold is fairly widely, though irregularly, distributed throughout Southeast Asia in igneous and metamorphic hard rock deposits and in sedimentary placer deposits. Luzon and Mindanao in the Philippines, the Barisan mountain range in west Sumatra, western Borneo, Timor, parts of the Malaysian and hai Peninsula, northern Burma, north and central Vietnam, Laos, northwest Cambodia in the Oddar Meanchey province, near Banteay Chhmar, its north-central area, in the Preah Vihear province in Rovieng district, and in the northeast in Rattanakiri province have all acquired a reputation as gold producers at one time or another, and the majority of these still produce small amounts of placer * www.analyzeart.com – 36, rue Paul-Émile Janson, 1050 Brussels, Belgium. (atnbennett@analyzeart.com) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 99-107 100 gold (Bronson, 1992: 83; Miksic 1990: 12-13; 2003: 19; Reinecke et al., 2002: 222; 2009a: 113-114; Workman, 1972) (Fig. 1). here is some evidence of ancient gold mining in Southeast Asia – ancient shafts have been reported in Central Vietnam at Kham Duc (pers. comm. local villagers), an area which is linked by the river to the protohistoric sites of Go Ma Voi, and at Go Mun, 65 km southwest (Nguyen Kim Dung et al., 1995; Nguyen Kim Dung, 2001). In addition, Classical gold coins are reported to have been found near the mines of the Rejang area, Bengkulu, in southwest Sumatra (Miksik, 1999: 17). Most recently, Iron Age bronzes were found in a mine shaft in Laos, at the Sepon gold and copper mine, known as the ‘Dragon Field’ (hong Na Nguak) on the 11th of April 2009 (UCR, 2009). he mine shafts were reported to be extremely well preserved, having been waterlogged and unexposed since the time they were abandoned. Most of the gold in the prehistoric and early historic periods would, however, undoubtedly have been extracted by panning alluvial sediments, a technique requiring little capital investment in equipment and no specialist technology, but unfortunately leaving no discernable archaeological signature. Many of the secondary deposits, which are today regarded as containing too little gold to be worked commercially, would have been quite successfully panned in the past. Today, the economic viability of gold panning will in part be determined by the price of gold on the international market 1. When gold prices are at their highest, it becomes more economically viable to pan the streams – the villagers sell their gold to middlemen who themselves sell on to merchants, and the number of transactions through which the gold may pass before inding its way to a town jeweller may be considerable (Fig. 2). Anna T. N. BENNETT Figure 1: Map showing the gold producing regions of Southeast Asia. Figure 1 : Carte avec localisation des régions productrices d’or du Sud-est de l’Asie. 2. TEXTUAL EVIDENCE Historical sources inform us that the Chinese were clearly impressed by the quantities of gold available in Southeast Asia, and one of the earliest recorded descriptions of the preAngkor Funan Kingdom is that of the Chinese emissaries who visited it during the Jin Dynasty (265-419 PE). hey referred to the Funan Kingdom as having already expanded its sphere of inluence to encompass parts of the Malay Peninsula, and as having a tax on gold, suggesting that the metal was readily available, if not locally produced. “here are many walled towns, palaces and houses… hey mostly 1. In 2008/2009 there has been some revival of gold mining in hailand, and one or two old locations are being reinvestigated (Bangkok Post). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 99-107 Figure 2: Myanmar jeweller producing gold beads which stylistically replicate ancient ones. Figure 2 : Orfèvre à Myanmar fabricant des perles en or stylistiquement identiques aux anciennes. Gold in early Southeast Asia take their food on silver utensils. he taxes are paid in gold, silver, pearls and perfumes” (Dube, 2003: 17). Another early Chinese text, known as “he History of Southern Ch’i”, which was written in the ifth century, described the inhabitants of Funan as having gold, silver and silk: “he inhabitants of Funan make rings and bracelets of gold and plates of silver” (Cœdès, 1968: 58). It seems quite probable that the gold ores available in Southeast Asia were not only partly responsible for the early trade with China, but that they were one of the stimulating factors in developing early contacts with India. he Ramayana mentions Sugriva sending his envoys to Hiranyanabha Mountain to the east, i.e. in the direction of Southeast Asia, which was known to be rich in gold, in search of Rama’s wife Sita. Sanskrit synonyms for gold, such as Hiranya, Kanaka, Hema and Suvarna, have been used to name rivers and mountains containing either alluvial or vein deposits of gold, and the region was known to the Indian merchants of the Mauryan Empire (321-185 BPE) as Suvarnabhumi: ‘Land of Gold’, which is thought to refer to the mainland, including lower Burma and the hai Malay Peninsula, and Suvarnadvipa: ‘Islands of Gold’, which may correspond to the Indonesian Archipelago, including Sumatra. A ifth century Sanskrit text narrates the story of merchants who travelled to Suvarnabhumi by sea. Having crossed the ‘unfathomable’ sea, they arrived at a coast, and embarked on a long voyage on foot, past the forested foothills of high mountains, which they then crossed, inally reaching a forest and a shallow river rich in gold sands (Dube, 2003). An eighth century Indian text known as the Samaraiccakaha describes a sea voyage to Suvarnadvipa and the making of bricks from the gold rich sands which they inscribed with the name dharana and then baked (Dube, 2003: 6). he sintering of such gold rich sands, by mixing the sand with water, which were shaped into bricks and then baked, would have ofered the early gold explorers a relatively easy method of transporting the raw material, which could then be carefully separated from its sand matrix in the more suitable environment back in their homeland, where the workforce would be plentiful. he scale of the sea trade in this early historic period is illustrated by the tenth century text Kathakosa (932 PE), which mentions a voyage with ive hundred ships (Dube, 2003, referring to Majumdar, 1936). A further indication of the importance of this trade is related to Indian coinage of this period, which is stamped with the picture of ships (Ray, 1994: Pl 12). he interpretation of early travel writings is not, however, always easy. he Javanese embassies to China in 860 and 873 PE refer to Java as rich in gold, although it was in fact devoid 101 of any deposits, and would have had to import its gold possibly from neighbouring Malaya, Sumatra or Borneo, where gold was still being mined in the 19th century and where ancient mining sites are reputed to have been located (Colless, 1975; Miksic, 1999: 19; Manning et al., 1980). In the 14th century, the Sumatran ruler Adityavarman, who had an honoriic title of Kanakamedinindra, meaning ‘Gold Land Lord’, is thought to have moved his capital to west Sumatra, notwithstanding the inhospitable coast, because of the rich gold deposits in the Minangkabau highlands (Miksic, 2004). Even though Java did not have its own gold deposits, the texts make frequent references to the existence of goldsmiths, and it is clear from the archaeological evidence that this culture had developed a sophisticated gold working technology, which relied on the importing of substantial quantities of metal (Wahyono Martowikrido, 1994; 1999). A gold and silver currency had been in place since the tenth century, and although copper coins imported from China during the Late Classical Majapahit period in the early 14th century replaced it, this replacement seems to have been more for the convenience of small denominations than because of a shortage in the supply of gold. In 1225 PE, the Chinese writer Zhao Rukuo referred to the Majapahit’s commander in chief and his 30,000 soldiers being paid in gold (Miksic, 1999; 2004). Elaborate descriptions of the complex jewellery representing gods, goddesses, kings, queens, warriors and priests were used in the Javanese literary texts and inscriptions to provide to the reader precise information pertaining to the status of the characters concerned (Hinzler, 1999). 3. GOLD COMPOSITION here is some evidence that the early explorers were aware of the diferences in compositions of gold. An 11th century Indian text refers to the gold coming from Survarnadvipa as being of diferent types: of yellow gold colour and white shell colour, which is presumed to refer to the amounts of silver or copper contained in the alloy. An 11th century Sanskrit inscription of the Cham king Rudravarman III at Po-Nagar temple mentions a donation by the king of seven panas of red coloured gold (rakta kaladhautam) and twenty-three kattikas and two panas of gold of superior white gold (sitatara kaladhautam). Majumdar (1986: 158-159) translated rakta kaladhautam as gold and sitatara kaladhautam as silver, but Dube (2003) has re-evaluated the texts and reached a more convincing explanation, of rakta kaladhautam being used for red gold and sitatara kaladhautam for white gold. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 99-107 Anna T. N. BENNETT 102 he goldsmith themselves would probably only rarely, if ever, have been able to acquire suicient capital to control their own stocks of metal and would generally have obtained their gold from the wealthy individual or institution who was commissioning the desired article. heir sources of raw material would have included recycled, damaged, or no longer sought after goods, which may have contained other baser metals, such as copper. It is unclear to what extent the goldsmiths undertook the relatively straightforward process of cupellation, which involves the addition of lead to a crucible of molten gold, which is then oxidised using a blast of cold air. he base metal oxides are absorbed into the lead oxide, leaving the gold with only silver remaining. Certainly touchstones, such as the ones found at Kota Cina in northeast Sumatra (Swan and Scott, 1986), and the one at Khuan Lukpat, Krabbi Province, in Peninsular hailand, which is dated to the 3rd century AD and has a Tamil inscription, Perumpadan Kal, identifying it as the property of Perumpadan (Srisuchat, 1996: 250), were part of the goldsmiths’ tool set. An old Javanese text compares the methods used to test the quality of human beings with those employed to test the purity of gold (Hinzler, 1999: 27), and an 11th century inscription provides details of the four ways in which gold may be tested: by using a touchstone, by hammering, cutting, and by melting. he earliest reference to the use of cementation to remove silver from the gold is contained in an Indian text known as he Kautiliya Arthasastra, which is believed to have been written by the chief minister of the Mauryean Emperor Chandragupta, who ruled northern India in the fourth century BC (Kangle, 1960). Gold dust or thin hammered sheets were placed in baths of acidic salts, which attacked the silver, forming a silver chloride vapour and leaving a puriied gold metal. Ornamental gold of the best kind, possessed of excellent colour, passed through an equal amount of lead, turned into leaves by heating made bright with Indus-earth, becomes the base of blue, yellow, white, green and parrot-feather colours (Kangle, 1960: 2.13.47). he cupellation with lead would have removed any base metal oxides, and the Indus earth, thought to be a saline soil rich in salt, nitre and ammonium salts which is collected from the Indian river valleys and used for the patination of zinc rich Bidri ware, would have removed the silver. For Southeast Asia, there is little analytical data concerning the chemical characterisation of the alloys used. he largest group of analyses is that of the Hunter hompson Collection of Javanese gold (Miksic, 1990), conducted by Riederer (1994). Although the material in question did not have a certain provenance, Riederer did detect a trend which ArcheoSciences, revue d’archéométrie, 33, 2009, p. 99-107 seemed to suggest that the earlier items in the collection contained a higher ratio of silver to gold. Two scientiic studies carried out on the gold inds from Kota Cina in northeast Sumatra, looking at the composition of inished gold sheet and attempting to locate the source of this gold, supported the view that the gold had been retrieved locally (Manning et al., 1980; Swan and Scott, 1986). Provenance studies have suggested that in the 12th century Western Borneo was producing gold which was being transported to the Malay Peninsula (Treloar, 1977). Gold panned from the Sarawak rivers in Borneo has been found to contain traces of antimony with platinum and arsenic (Treloar and Fabris, 1975), gold from Sulawesi may be characterised by the presence of nickel as the only impurity, while gold from Sumatra contains tin, lead and mercury (Stambolov, 1999). However, the chances of successfully matching an object to its source of raw material decreases with the degree of processing and recycling to which the object has been submitted during its manufacture. Also, because gold is never thrown away but has usually been recycled and traded, impurity patterns will generally be far from clear. 4. ARCHAEOLOGICAL EVIDENCE Given the amount of gold available in Southeast Asia, remarkably little archaeological evidence has been found in the late prehistoric sites, and presently none can be dated to a period before 400 BPE. Much of the gold retrieved from archaeological sites has not been adequately recorded but, because of its metal value, has been melted or in some cases sold into the art market. Gold beads from the Lopburi Artillery site in Central hailand were reportedly found in a pre-Dvaravati context (Bennett 1988a: 37; 1988b); hammered sheet gold inds are said to originate from a site near Battambang, west of Angkor. In 1999, Iron Age gold items, which are said to have included a bronze helmet inlaid with gold and gold jewellery, were plundered from the cemetery site of Phum Snay, north of the Tonle Sap in Cambodia 2, where about 50 inhumations were discovered and dated from about 100 BPE to 500 PE (O’Reilly and Pheng Sytha, 2001: 265266; Higham, 2002: 214; Stark, 2004: 96). he analysis of two gold earrings from Phum Snay showed that they had a composition of 70% gold and 30% silver and were possibly 2. he site, which is in the Preah Net Preah district of Banteay Mean Chey Province, Cambodia, was excavated in 2001 by O’Reilly and Pheng Sytha (2001), and, although they found no gold, they conirmed it as a rich Iron Age cemetery with military paraphernalia. Gold in early Southeast Asia manufactured from an alluvial electrum (Hieda et al., 2008: 141; Reinecke et al., 2009b). Notwithstanding the objects that may have been lost to looters, the gold inds, together with other highly prized materials recovered from Iron Age sites, suggest that during this period gold was being used as a mark of social rank. At the large Iron Age cemetery site of Noen U-Loke in the Mun Valley, Northeastern hailand, dated to the 3rd-4th centuries PE, a woman wore a necklace of agate and hollow square-shaped gold beads, together with bronze bangles and bronze and silver inger and toe rings; a man who wore silver ear coils covered in gold leaf was buried with three bronze belts, seventy ive bronze bangles on each arm, numerous bronze inger and toe rings, and glass bead necklaces and anklets (Chang, 2001; Higham and hosarat, 1998: 151158; Higham, 2004: 63). Earlier protohistoric sites have produced very little gold and, when found, it is usually in the form of a bead or two. Some of the earliest gold was found at the coastal sites of Giong Phet and Giong Ca Vo, in Southern Vietnam, which are dated to the 3rd-2nd century BPE, and which are claimed to be the earliest documented glass manufacturing sites in Vietnam, if not in the whole of Southeast Asia (Nguyen Kim Dung et al., 1995; Nguyen Kim Dung, 2001: 109). At the only slightly later sites of Go Ma Voi and Go Mun 3, on the banks of Ba Ren river, a branch of the hu Bon, near Hoi An in Quang Nam province, central Vietnam, Reinecke excavated two gold beads at Go Ma Voi, and recorded beads from the burial ground of Go Mun which were located in the local village collections, together with agate, nephrite and quartz beads, lingling-o earrings, and a bimetallic axe (Reinecke and Lê Duy Son, 2000; Reinecke et al., 2002: 23 and 176). he gold content of the looted bead from Go Mun was 94%, with 5% silver, suggesting that it was made of natural placer gold (Reinecke and Lê Duy Son, 2000). Not until the recent excavations at the protohistoric cemetery of Prohear, about 65 km east of Phnom Penh and 40 km west of the Vietnamese border, dated to a period between the 2nd century BPE and the 1st century PE (Reinecke et al., 2009b), did a signiicant corpus of early gold material which could be analysed exist. In 2007, the site was extensively looted by the villagers, who within a few months had dug up almost the entire cemetery and sold the artefacts, which included gold items and many Dongson bronze drums, on the art market and for scrap metal 4 (Reinecke et al., 2009b). 3. hese are the sites which are linked by river to Kham Duc, 65 km northeast, an area where gold mining is still undertaken today and where evidence of ancient shafts have been reported (villagers, pers. comm.). 4. he villagers of Prohear sold the iron oferings they found in the burials for 900 Riels (0.2 US $) per kilo to the scrap yard (Reinecke et al., 2009b). 103 Archaeologists were able to photographically document or see photographs of more than ifteen gold or silver objects originating from looted burials, and which had already been or were yet to be sold. In the last two years, a scientiic excavation along the main village street at Prohear has produced seventy nine gold or silver objects in 52 burials, and has revealed the full extent of the richness of this protohistoric site (Reinecke et al., 2009b). Out of the seventy nine gold and/or silver objects that were excavated, thirty fragments, mostly of small wire spiral rings, were analysed using LA-ICP-MS (Laser Ablation – Inductively Coupled Plasma – Mass Spectrometry) (Schlosser, 2009). he resulting data has indicated that slightly less than half of the objects analysed contained more silver than gold, and suggests that the inhabitants of Prohear were using both a native panned electrum and were also producing intentional alloys. he natural electrum samples contained between 35-44% silver and less than 0.2% copper, while the intentional alloys showed a similar low copper content, but a higher silver content, between 56 and 68%. Platinoid inclusions were present in greater concentration in the intentional alloys with a higher gold content and lower silver content, since the concentration of the platinoids originally contained within the alluvial gold is diluted by the intentional addition of silver. he fact that the early inhabitants of the region were engaged in trade as well as manufacture is clearly demonstrated by the inds from Oc Eo, in the Mekong Delta in Southern Vietnam (Malleret, 1959-1963; Vo Si Khai, 2003). Oc Eo is not only one of the earliest commercial centres in Southeast Asia engaged in the trade of a wide range of specialised goods, such as horses and elephants from India, but it is also the earliest excavated site to produce evidence of a gold workshop. Hammers, awls, bivalve casting moulds, droplets of gold, fragments of copper, gold sheet, and partially inished repoussé gold plaques were recovered by Mallaret in 1944. his was accompanied by signiicant amounts of gold jewellery, including rings, some surmounted by images of Nandi, the sacred bull, linked chains, inscribed gold sheets, gold plaques decorated with repoussé images of Hindu deities, and over nine hundred gold beads with various shapes (Higham, 2002: 233; Le Xuan Diem et al., 1995: 338-365; Malleret, 1962: 41; Manguin, 2004; Miksic, 2003: 18-19). he analysis of a 7 gram gold ingot showed that it contained 19% silver and 5% copper, suggesting that intentional alloying took place prior to the manufacture of the object, and perhaps also suggesting that the gold was traded as an intentional alloy (Malleret, 1962: 460). he only other analysis from Oc Eo was that of a fragment of gold wire, which ArcheoSciences, revue d’archéométrie, 33, 2009, p. 99-107 104 had a diferent composition, containing only 5% silver and 1% copper. Another site located in the vicinity, the cemetery of Go hap, excavated in 2001, produced some of the earliest known Hindu iconographic images in Southeast Asia, including three hundred and twenty two gold plaques with simple incised decorations, depicting turtles representing Vishnu and his mount, the eagle Garuda, water bufaloes, elephants, snakes, conch shells, the sun, plants, and houses on stilts (Fig. 5). Deposited as an ofering together with ive gold discs, three gold rings, a gold lower, eight precious stones and seven pieces of glass, the site has been dated to a period as early as the 2nd century PE, although Higham dates the site to 400-600 PE (Higham, 2001: 31). he large coastal and riverine settlements around Oc Eo in the Mekong Delta (Malleret, 1959-1963; Vo Si Khai, 2003), Giong Ca Vo (Nguyen Kim Dung, 2001) in Southern Vietnam, the early river port of Khao Sam Kaeo in eastern peninsular hailand (Bellina and Silapanth, 2006), together with the Tabon caves in the Palawan Islands of the Philippines represent the earliest sites yielding gold inds in maritime Southeast Asia. Although the gold inds from these sites have not all been excavated and are often chance inds, recorded in local villagers’ collections, they share many highly characteristic features which provide suicient evidence to indicate that a trade network linking the areas of the Southeast Asian mainland, the outer islands, the Philippines and India was already in place by the 2nd half of the irst millennium BPE. Central knobbed gold beads excavated at Giong Ca Vo (Dang and Vu, 1997: 41, Plate 17), and which have also been recorded at other Sa-Huynh related sites, strongly resemble surface inds from the peninsular site of Khao Sam Kaeo (Pryce et al., 2008: Fig. 14). A gold ive-lobed rosette plate, a surface ind from Khao Sam Kaeo, is similar to artefacts recovered by Fox from the Guri and Tadyaw Caves of the Tabon complex in the Palawan Islands, which Fox dates to the Early Metal Age (between 500 and 300 BPE) and the Developed Metal Age (100 BPE to 300 PE), respectively (Fox, 1970). Several of the gold ornaments recorded in private collections around Khao Sam Kaeo, including the characteristic polyhedral gold beads (Fig. 3), have also been found in local private collections and bear strong stylistic similarities to inds recorded from Oc Eo (Malleret, 1962: Plate XII), to beads from numerous Pyu sites in Myanmar (Richter, 2000), and to others observed in private collections (Fig. 4) (Pryce et al., 2008: 310). he present archaeological evidence indicates an emergent, complex, hierarchical society that was to some degree already in place by the beginning of the Christian era, preceding the development of the early Indic states of the midArcheoSciences, revue d’archéométrie, 33, 2009, p. 99-107 Anna T. N. BENNETT Figure 3: (See colour plate) he polyhedral gold bead from Khao Sam Kaeo, in eastern peninsular hailand, is stylistically identical to those from Oc Eo in the Mekong Delta, and to numerous similar beads from Pyu sites in Burma. Figure 3 : (Voir planche couleur) La perle d’or en forme de polyèdre de Khao Sam Kaeo, Est de la haïlande péninsulaire, est stylistiquement identique à la perle d’Oc Eo, delta du Mékong, et à beaucoup d’autres perles provenant des sites de Pyu, Burma. irst millennium PE and later. While much archaeological work is still necessary to clarify the socio-economic conditions in the period between 500 BPE and 500 PE, there is ample evidence that mainland Southeast Asian societies were engaged in extensive and sophisticated commercial trading, either directly or indirectly, with India, Central and Western Asia, Insular Southeast Asia and China throughout the last two and a half thousand years. he fact that gold appears in the archaeological record in the second half of the irst millennium BPE, at about the same time as iron, semiprecious stone polishing and glass working, suggests that the Gold in early Southeast Asia 105 techniques of gold extraction and working were quite plausibly introduced to Southeast Asia via Indian or Chinese merchants seeking the gold ores available in Southeast Asia. It also seems quite plausible that the conquest of the Upper hai-Malay Peninsula by the Funan Kingdom, which is clearly mentioned in Chinese texts, may have been stimulated by a desire to control this area’s mineral resources. his elaborate international trade network was not a homogenous system, but was rather based on diferential economies, which may nonetheless have eventually stimulated local gold panners to retrieve excess metal to be sold in the context of a wider distribution. he rulers of the large coastal and riverine settlements around Oc Eo in the early Funan Kingdom are reported to have sent more than twenty-ive missions to China between 226 and 649 PE, bearing gold and other precious gifts. Acknowledgements Figure 4: (See colour plate) Gold beads worn by a woman at the large Iron Age cemetery site of Noen U-Loke in the Mun Valley, Northeastern hailand, dated to the 3rd-4th century PE. Figure 4 : (Voir planche couleur) Perles en or portées par une femme enterrée dans le cimetière de l’Age du Fer du site de Noen U-Loke, vallée de Mun, nord-est de la haïlande, datées des IIIe-IVe siècles PE. I am very grateful to Dr Bérénice Bellina-Pryce and Praon Silapanth, the directors of the archaeological mission of Khao Sam Kaeo in Eastern Peninsular hailand for allowing me access to the material from the site, to Kim Dung Nguyen for information concerning Giong Phet and Giong Ca Vo, in Southern Vietnam, to U Win Muang for allowing free access to his collection, and to Charles Higham for allowing me to use the photograph from Noen-U-Locke. I am, as always, deeply indebted to Ian Glover for his help, support and encouragement. Grateful acknowledgements are made to the following for useful discussions and communications: Oliver Pryce, John Guy, John Miksic. Maria Korkel has taken enormous care and trouble in the prooing of this paper and I am most grateful to her. References Figure 5: (See colour plate) he earliest known Hindu iconographic images in Southeast Asia include three hundred and twenty two gold plaques with simple incised decorations from Go hap. Figure 5 : (Voir planche couleur) Les plus anciennes images iconographiques hindoues connues dans le sud-est de l’Asie incluent trois cents vingt-deux plaques en or avec de simples décorations incisées de Go hap. BELLINA, B. and SILAPANTH, P., 2006. Khao Sam Kaeo and the Upper hai Peninsula: Understanding the mechanism of early trans-Asiatic trade and cultural exchange, in E.A. Bacus, I.C. Glover, V.C. Pigott (eds.), Uncovering Southeast Asia’s Past – Selected Papers from the Tenth Biennial Conference of the European Association of Southeast Asian Archaeologists. Singapore, National University Press, 379-392. BENNETT, A., 1988. Prehistoric copper metallurgy in Central hailand, in P. Charoenwongsa, B. Bronson (eds.), Prehistoric Studies: he Stone and Metal Ages in hailand. Bangkok, Fine Arts Department, 125-135. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 99-107 106 BENNETT, A., 1988. Copper Metallurgy in Central hailand. Ph.D. dissertation, Institute of Archaeology, University College London. BRONSON, B., 1992. Patterns in the early Southeast Asian Metals Trade, in I.C. Glover, P. Suchitta, J. Villiers, (eds.), Early Metallurgy, Trade and Urban Centres in hailand and Southeast Asia. Bangkok, White Lotus, 63-114. CHANG, N.J., 2001. Personal ornaments in hai Prehistory, Nong Nor, Bam Lum Khao, and Noen-U-Locke. Unpublished thesis, University of Otago, Dunedin, New Zealand. CŒDÈS, G., 1968. The Indianized States of Southeast Asia. Honolulu, University Press of Hawaii. COLLESS, B.E., 1975. Were the gold mines of ancient Java in Borneo? he Sarawak Museum Journal 5(1): 14-22. DANG V.T. and VU, Q.H., 1977. he excavation at Giong Ca Vo site. Journal of Southeast Asian Archaeology 17: 30-37. DUBE, R.K., 2003. Suvarnahumi, Suvarnadvipa: Origin, Identity and its Richness in Gold in Ancient and Medieval Times. Bulletin of the Metals Museum 36: 3-23. FOX, R.B., 1970. he Tabon Caves: Archaeological Explorations and Excavations on Palawan Island, Philippines. Manila, National Museum. HIEDA, S., YOSHIMITSU, H. and SHIGERU, K., 2008. Chemical Composition of Bronze Artifacts Unearthed from Phum Snay Archaeological Site in Cambodia, in Yashuda Yoshinori (ed.), Final Report on Excavation in Phum Snay in 2007,Chuch Phoeurn eds., Phnom Penh, 133-143 (in Khmer). HIGHAM, C., 2001. he Civilization of Angkor. Berkeley and Los Angeles, University of California Press. HIGHAM, C., 2002. Early Cultures of Mainland Southeast Asia. Bangkok: River Books. HIGHAM, C., 2004. Mainland Southeast Asia from the Neolithic to the Iron Age, in I.C. Glover, P. Bellwood (eds.), Southeast Asia. From prehistory to history. New York, Routledge Curzon. HIGHAM, C. and THOSARAT, R., 1998. Prehistoric hailand from early settlement to Sukhothai. Bangkok: River Books. HINZLER, H., 1999. Gold and Jewellery in Hindu-Buddhist Java from the 10th to the 15th century, in W.H. Kal (ed.), Precious Metals in early South East Asia. Amsterdam, Royal Tropical Institute, 27-38. KANGLE, R.P., 1960. he Kautiliya Arthasastra. Part II. An English translation with critical and explanatory notes. Bombay, University of Bombay. LE THI LIEN, 2004. he excavations at Minh Su Mound (Go hap site in Dong hap province South Vietnam), 20002003, proceedings from European Association of Southeast Asian Archaeologists, 10th International Conference, London 14th-17th Sept 2004. LE XUAN DIEM, DAO LINH CON, VO SI KHAI and VAN HOA, 1995. Oc Eo. Oc Eo cultures, recent discoveries. Hanoi, Social Science ArcheoSciences, revue d’archéométrie, 33, 2009, p. 99-107 Anna T. N. BENNETT Publishing House (In Vietnamese but with line drawings and photographs of the gold). MAJUMDAR R.C., 1986. Suvarnadvīpa: ancient Indian colonies in the Far East, Vol. 1. 1936, Reprint. Cosmos Publications, New Delhi. MALLERET, L., 1959-1963. L’Archéologie du Delta du Mékong, Vols. 1-4. Paris, Ecole Française d’Extrême Orient. MANGUIN, P, 2004. he archaeology of early maritime polities of Southeast Asia, in I.C. Glover, P. Bellwood (eds.), Southeast Asia. From prehistory to history. New York, Routledge Curzon, 282-313. MANNING, A., MCKINNON, E.E. and TRELOAR, F.E., 1980. Analysis of gold artefacts from Kota Cina site near Medan, Sumatra. Journal of the Malaysian Branch of the Royal Asian Society 53(2): 102-116. MIKSIC, J., 1990. Old Javanese Gold. Singapore, Ideation. MIKSIC, J., 1999. Clustered gold inds: hoards or sub-assemblages? in W.H. Kal (ed.), Precious Metals in early South East Asia. Amsterdam. Royal Tropical Institute, 14-22. MIKSIC, J., 2003. Introduction: the Beginning of Trade in Ancient Southeast Asia: he Role of Oc Eo and Lower Mekong River, in J.C.M. Khoo, Art and Archaeology of Fu Nan. Bangkok, Orchid Press, 1-34. MIKSIC, J., 2004. he Classical Culture of Indonesia, in I.C. Glover, P. Bellwood (eds.), Southeast Asia. From prehistory to history. New York, Routledge Curzon, 234-256. NGUYEN, KIM DUNG, TRINH CAN, DAN VAN THANG, VU QUOC HIEN and NGUYEN THI HAU, 1995. Ornaments from jar burial sites in Can Gio District, Ho Chi Minh City. Khao Co Hoc 2: 27-46. NGUYEN and KIM DUNG, 2001. Jewellery from late prehistoric sites recently excavated in Southern Vietnam in Indo-Paciic Prehistory: he Melaka Papers Volume 5. Bulletin of the IndoPaciic Prehistory Association 21: 107-113. O’REILLY, D.J.W. and PHENG SYTHA, 2001. Recent excavations in northwest Cambodia. Antiquity 75(288): 265-266. PRYCE, T., BELLINA-PRYCE, B. and BENNETT, A., 2008. he development of metal technologies in the Upper hai-Malay Peninsula: initial interpretation of the archaeometallurgical evidence from Khao Sam Kaeo. Bulletin de l’Ecole Française d’Extrême-Orient 93(2006): 295-315. RAY, H.M., 1994. he Winds of change, Buddhism and the maritime links of early South Asia. Delhi, Oxford University Press. REINECKE, A. and LÊ DUY SON., 2000. Ein neu entdecktes Gräberfeld der Sa-Huynh-Kultur von Go Mun in Mittelvietnam. Beiträge zur Allgemeinen und Vergleichenden Archäologie 20: 5-43. REINECKE, A., NGUYEN CHIEU and LAM THI MY DUNG., 2002. Neue Entdeckungen zur Sa-Huynh-Kultur. Köln. Linden Soft. Gold in early Southeast Asia REINECKE, A., LAYCHOUR, VIN. and SONETRA, SENG., 2009A. Discovered: he rich past of Prey Veng. he Cambodian Scene 41: 40-41. REINECKE, A., LAYCHOUR, VIN. and SONETRA, SENG., 2009B. he irst Golden Age of Cambodia: Excavation at Prohear. Bonn, Die Deutsche Nationalbibliothek. RICHTER, A., 2000. Jewelry of Southeast Asia. New York, Harry N. Abrams Inc. RIEDERER, J., 1994. he technological analysis of early gold objects from Java, in W.H. Kal, (ed.), Old Javanese Gold (4th-15th century). Amsterdam, Royal Tropical Institute, Tropenmuseum, 46-57. SCHLOSSER, S., 2009. Nothing but gold and silver – it depends on the right mixture, in A. Reinecke, Vin. Laychour, Seng. Sonetra (eds.), 2009, he irst Golden Age of Cambodia: Excavation at Prohear. Bonn, Die Deutsche Nationalbibliothek, 110-116. SRISUCHAT, A. (ed.), 1996. Ancient Trades and Cultural Contacts in Southeast Asia. Bangkok, he Oice of the National Culture Commission. STAMBOLOV, T., 1999. Gold Mines, ancient Indonesian ornaments and the exact sciences, in W.H. Kal (ed.), Precious Metals in early South East Asia. Amsterdam, Royal Tropical Institute, Tropenmuseum, 23-26. STARK, M., 2004. Pre-Angkorian Cambodia, in I.C. Glover, P. Bellwood, Southeast Asia. From Prehistory to History. New York, Routledge Curzon, 89-119. 107 SWAN, R. and SCOTT, D., 1986. Gold Work from Kota Cina: a technical study. Southeast Asian Archaeology. B.A.R. International Series. Oxford, Archaeopress, 275-286. TRELOAR, F.E., 1977. he composition of gold artefacts from Santubong and Gedong, Sarawak. Sarawak Museum Journal 25(46): 9-18. TRELOAR, F.E. and FABRIS, G.J., 1975. Evidence for the contemporary existence of two Kedah sites. Journal of the Malaysian Branch of the Royal Asian Society 48: 74-76. UCR, 2009. University of California, Riverside, CA 92521. At http://lists.ucr.edu, April 2009. VO SI KHAI, 2003. he kingdom of Fu Nan and the Culture of Oc Eo, in J.C.M. Khoo, Art and Archaeology of Fu Nan. Bangkok, Orchid Press, 35-86. WAHYONO M., 1994. he gold of Wonoboyo, in W.H. Kal (ed.), Old Javanese Gold (4th-15th century). Amsterdam, Royal Tropical Institute, 30-45. WAHYONO M., 1999. Der Goldschatz von Plosokuning, in W.H. Kal (ed.), Precious Metals in early South East Asia. Amsterdam, Royal Tropical Institute, 73-76. WORKMAN, D.R., 1972. Mineral Resources of the Lower Mekong Basin and Adjacent Areas of Khmer Republic Laos, hailand and Republic of Viet-Nam. Mineral resources Development 39. Bangkok and New York, Economic Commission for Asia and the Far East, United Nations. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 99-107 S  O:      ÉTUDES D’OBJETS : COMPÉTENCES TECHNIQUES ET CHOIX DES ALLIAGES Technological study of gold jewellery pieces dating from the Middle Kingdom to the New Kingdom in Egypt Étude technologique de pièces d’orfèvrerie datées du Moyen Empire au Nouvel Empire en Égypte Lore G. Troalen*, Maria Filomena Guerra**, Jim Tate* and Bill Manley*** Abstract: his paper presents a technological study of items of Egyptian jewellery from the collections of the National Museums Scotland: a pendant from the 19th century BC; objects from the 16th century BC royal burial unearthed at Qurneh; two gold inger-rings dated to the 14th century BC; and a group of pendants from the 13th century BC. he objects were examined using optical microscopy, X-radiography and Scanning Electron Microscopy (SEM). In addition, the elementary compositions of alloys and solder joints were investigated using air-path X-ray luorescence (XRF), proton-induced X-ray analysis (micro-PIXE), and energy-dispersive X-ray analysis (SEM-EDS). his preliminary study provides information about the evolution of alloy composition and the use of alluvial gold, and illustrates the Ancient Egyptian goldsmiths’ skills in working with wires, granulation and joining techniques. Résumé : Cet article présente l’étude technologique de pièces égyptiennes d’orfèvrerie faisant partie de la collection des musées nationaux d’Écosse : un pendentif daté du XIXe siècle av. J.-C., des pièces d’orfèvrerie provenant de la tombe royale de Qurneh datées du XVIe siècle av. J.-C., deux bagues datées du XIVe siècle av. J.-C. ainsi qu’un groupe de pendentifs datés du XIIIe siècle av J.-C. Les objets ont été observés sous microscopie optique, par radiographie de rayons-X et par microscopie électronique à balayage en mode électron secondaire (MEB-SE). En parallèle, la composition des alliages et des soudures a été déterminée par luorescence des rayons-X (FX), par analyse par faisceaux d’ions (micro-PIXE) et par énergie dispersive de rayons-X (MEB-EDS). Cette étude préliminaire donne des informations sur l’évolution des alliages ainsi que sur l’utilisation d’or alluvionnaire, et illustre les techniques utilisées par les orfèvres de l’Egypte Ancienne, en particulier l’utilisation du iligrane, de la granulation et des techniques de soudure. Keywords: Goldwork, Egypt joining, base-alloys, NMS. Mots-clés : Orfèvrerie, Egypte, soudures, alliages, NMS. * National Museums Scotland, Department of Conservation & Analytical Research – 242 West Granton Road, Edinburgh EH5 1JA. (l.troalen@nms. ac.uk and j.tate@nms.ac.uk) ** Laboratoires du Centre de Recherche et de Restauration des Musées de France, UMR171 CNRS – 14, quai François Mitterrand, 75001 Paris, France. (maria.guerra@culture.gouv.fr) *** National Museums Scotland, Department of World Cultures, – Chambers Street, Edinburgh EH1 1JF. (b.manley@nms.ac.uk) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 111-119 Lore G. TROALEN et al. 112 1. INTRODUCTION Numerous works on jewellery originating from Ancient Egypt are conveniently cited in studies by Lucas (1929, reedited 1962), Ogden (1982; 2000) and Andrews (1990). However, relatively little work is available combining elemental analysis with technological study; the principal exceptions are Lucas (1962) and Schorsch (1995) on jewellery from the burial of Wah, and Lilyquist (2003) on jewellery from the burial of the foreign wives of hutmose III, while others are cited by Schorsch (2001). As a result, little is known about the alloys used in Ancient Egypt, or about manufacturing techniques such as iligree and granulation. In order to shed more light on this technology, we applied a multi-disciplinary approach to the analysis of several artefacts in the collections of National Museums Scotland (NMS). NMS holds about 6,000 objects from Ancient Egypt and Sudan, including gold and electrum jewellery that has hitherto not been the subject of signiicant scientiic study. he collections cover all periods from the Predynastic to the Roman Empire, and a good proportion of the artefacts derive from reputable archaeological excavations carried out during the 19th and 20th centuries. An example of this is the intact royal burial from Qurneh, near hebes, acquired by the former Royal Scottish Museum (a precursor of NMS) in 1909, and dated to the 16th century BC. he Qurneh burial included the ornately decorated coin and mummy of a young adult woman and the simpler coin and mummy of a child (Petrie, 1909), both of whom had been buried with a range of grave goods, including jewellery of gold, electrum, ivory and faience (Eremin et al., 2000; Manley et al., 2002). In addition to the gold necklace described elsewhere in this volume (Tate et al., 2010), the woman was provided with: four gold bracelets made from a D-section bar bent into a ring and soldered (A.1909.527.16 + A-C); an electrum girdle consisting of twenty-six scaraboid beads and twelve barrel beads threaded on a double string (A.1909.527.17); and two gold earrings consisting of four penannular hoops soldered together (A.1909.527.18 + A). he child’s mummy wore a simpler gold necklace (A.1909.527.11), and two loops, each consisting of three-and-a-half gold rings, which are usually described as earrings (A.1909.527.43 +A). To the purpose of carrying out an initial investigation into the varieties of alloys and production techniques used in Ancient Egypt, we studied the Qurneh jewellery along with a small collection of other gold items spanning the period between the 19th and the 13th centuries BC, containing: a ish-shaped pendant (A.1914.1081) from a tomb at ArcheoSciences, revue d’archéométrie, 33, 2009, p. 111-119 el-Harāgeh from the mid-19th century BC 1; two gold ingerrings with decorated bezels (A.1883.49.2 + A.1883.49.8) from the Amarna royal tomb of the late 18th Dynasty (14th century BC); 2 and sixteen ly-shaped pendants (A.1977.140 A-P), assumed, on stylistic grounds, to date from the 19th Dynasty (13th century BC) 3. 2. METHODS he objects were examined at NMS using a stereomicroscope (Olympus SZX12 x 7-90) equipped with a digital camera (Olympus DP70) and CamScan Scanning Electron Microscopy in Secondary Electron mode (SEI), and their construction investigated using a 300kV Pantak X-radiography system. he elemental composition of the alloys was determined using an Oxford Instruments ED 2000 air-path X-ray luorescence (XRF), with a Rhodium target X-ray tube collimated to a point of about 2 x 1.5 mm, coupled to a Si(Li) detector using the Oxford Xpert Ease software. In addition, proton-induced 3MeV X-ray analysis (micro-PIXE) was carried out at the external beamline of the AGLAE accelerator at the Centre de Recherche et de Restauration des Musées de France (Guerra and Calligaro, 2003; Dran et al., 2004; Guerra, 2008). Energy Dispersive X-ray analysis (SEM-EDS), using the Noran Vantage system and Vista software, was used to characterise inclusions of platinumgroup elements (PGE) and investigate other compositional information. All SEM-EDS measurements were performed at the analytical working distance of 35.0 mm at 300 s measurement with a voltage of 20 kV or 25 kV. Table 1 presents the compositional analysis of ternary alloys of gold-silver-copper standards, showing the interinstrument compatibility. he compositions of the objects can be found in Table 2, and are summarised in the ternary diagram in Fig. 3. For the Qurneh jewellery, all compositional results are from PIXE analysis, except for the girdle, which, like most of the jewellery, was analysed with XRF. 1. he pendant is 41 mm long by 19 mm high. It was excavated in the burial of a ten years old girl in Cemetery A/Tomb 72. Fish-shaped pendants, known as ńkhā, were usually worn by children or young women. 2. See Martin (1974, nos. 275 and 280). Dimensions: A.1883.49.2, diameter of hoop 28 mm (internal 16 mm), bezel 12 mm by 13 mm. A.1883.49.8, diameter of hoop 22 mm, bezel 13 mm by 9 mm. 3. he ly-pendants might have formed part of an oicial’s usekh-collar, cf. an example in he Egyptian Museum, Cairo (CG 52677). he pendants are nearly identical and are 11 mm long by 7 mm high. Technological study of gold jewellery pieces dating from the Middle Kingdom to the New Kingdom in Egypt Standards NMS Rolled gold GCS5 GCS4 GCS7 Au % Ag % Cu % heoretical Value 80.0 10.0 10,0 PIXE - 2 values 80.0 10.7 9,3 XRF - 5 values standard deviation 79.4 10.6 10,0 0.3 0.3 0,1 SEM-EDS - 4 values standard deviation 80.9 9.6 9,5 0.9 0.7 0,3 heoretical Value 80.0 17.1 2,9 PIXE - 2 values 79.4 17.9 2,7 XRF - 5 values standard deviation 78.5 18.6 3,0 0.5 0.4 0,1 SEM-EDS - 3 values standard deviation 79.2 17.9 2,9 1.1 1.4 0,4 heoretical Value 61.9 35.1 3.0 XRF - 8 values standard deviation 61.4 35.6 3.0 0.7 0.8 0.1 heoretical Value 45.0 39.8 15.2 XRF - 6 values standard deviation 44.5 40.0 15.5 0.1 0.2 0.3 Table 1: Compositional results (in wt%) of gold-silver-copper alloy standards by PIXE, XRF and SEM-EDS. Tableau 1 : Compositions (en %) des alliages étalons en or-argentcuivre par PIXE, FX et MEB-EDS. 3. RESULTS All the jewellery showed indications of wear, particularly the Qurneh girdle and one of the Amarna inger-rings (A.1883.49.2). For the girdle, we were able to observe loose a 113 decoration, and deformations of the edges of the beads and of holes in the wallet spacers where they sit against the beads (Fig. 1a). Occasionally, some small barrel beads are trapped within wallet beads (Fig. 1b). For the inger-ring, we observed that the granules on the bezel are lattened, except those immediately beside the bezel hole, which remain perfectly spherical (see Fig. 5). Gold Alloys We observed and analysed PGE inclusions, ranging from a few to a few hundred microns in diameter, in all the jewellery items except the ly-pendants. PGE inclusions are characteristic for gold from alluvial deposits (Ogden, 1976), and can be found in all the Qurneh objects. However, only a single inclusion was found in the woman’s earrings, and a single small inclusion was found in the tail of the ishpendant (Fig. 2a). By comparison, the Amarna inger-rings contain numerous inclusions, which may indicate the use of gold from other deposits. he inclusions show a range of composition, with an average of 36% iridium, 42% osmium and 22% ruthenium, corresponding well with compositions reported by Meeks and Tite (1980) in their investigation of Egyptian objects from the British Museum, and also with the range otherwise expected for Egyptian deposits (Ahmed, 2007). We also observed variable ratios of iridium, osmium and ruthenium in the PGE inclusions within the same object, making the formulation of any conclusions regarding the gold deposits diicult. However, the absence of platinum, rhodium or palladium indicates that the gold is not associated with the alluvial deposits of Yubdo (Wallaga) in Ethiopia (Molly, 1959). With regard to the alloys (Table 2; Fig. 3), we observed that most of the artefacts were made from gold with a silver b Figure 1: Qurneh girdle (A.1909.527.17): (a) OM image showing some deformation of barrel beads; (b) X-radiograph showing small barrel beads that have slipped inside a wallet bead. Figure 1 : Ceinture de Qurneh (A.1909.527.17) : (a) image sous loupe binoculaire illustrant la déformation des perles coniques ; (b) radiographie X illustrant les petites perles coniques à l’intérieur d’une perle en portefeuille. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 111-119 Lore G. TROALEN et al. 114 a b Figure 2: PGE inclusions in (a [See colour plate]) pendant (A.1914.1081) from el-Harāgeh by OM; (b) Qurneh child’s necklace (A.1909.527.11) by SEI. Figure 2 : (Voir planche couleur) Inclusions de platinoïdes (a [Voir planche couleur]) du pendantif (A.1914.1081) d’el-Harāgeh sous loupe binoculaire ; (b) du collier d’enfant de Qurneh (A.1909.527.11) au MEB. Figure 3: Ternary diagram (gold, silver, copper) showing the average alloy composition of the jewellery investigated (for some objects, multiple analyses of the diferent components were undertaken). Figure 3 : Diagramme ternaire (or, argent, cuivre) montrant la composition moyenne des alliages des pièces d’orfèvrerie analysées (pour certains objets l’analyse a été efectuée pour leurs diférentes parties). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 111-119 Technological study of gold jewellery pieces dating from the Middle Kingdom to the New Kingdom in Egypt Qurneh Bracelet A199-527-16 standard deviation Earring adult A1909-527-18 standard deviation Necklace A.1909.527.19 Girdle A.1909.527.17 Average on 26 beads Average on 10 beads Ring 1 standard deviation Ring 2 standard deviation Ring 3 standard deviation Wallet Beads (XRF) standard deviation Barrel Beads (XRF) standard deviation Child’s mummy (PIXE) Earring A.1909.527.43 A standard deviation Earring A.1909.527.43 B standard deviation Ring 1 standard deviation Ring 2 standard deviation Fish pendant A.1914.1081 Harāgeh (XRF) Tail standard deviation Body standard deviation Amarna Rings A.1883.49.2 Hoop (XRF) standard deviation Low row Granules (SEM-EDS) standard deviation Upper row standard deviation A.1883.49.8 (XRF) Necklace A.1909.527.11 Fly-shaped pendants 19th Dynasty (XRF) A.1977.140.A A.1977.140.B A.1977.140.C A.1977.140.D A.1977.140.E A.1977.140.F A.1977.140.G A.1977.140.H A.1977.140.I A.1977.140.J A.1977.140.K A.1977.140.L A.1977.140.M A.1977.140.N A.1977.140.O A.1977.140.P Solder Qurneh Bracelet A.1909.527.16 standard deviation Earring adult A.1909.527.18 standard deviation Necklace A.1909.527.19 Earring A.1909.527.43 A standard deviation Earring A.1909.527.43 B Amarna Rings Granules (SEM-EDS) Au % Ag % Cu % 88.2 0.8 95.8 1.0 87.6 1.6 87.7 1.1 88.0 11.6 0.7 4.1 0.9 10.5 1.6 10.7 0.9 10.0 0.3 0.5 0.2 0.1 1.9 0.1 1.5 0.2 2.0 0.8 1.1 0.3 43.8 0.3 42.1 2.7 52.5 0.3 53.1 1.4 3.7 0.1 4.8 1.5 82.5 0.5 83.7 1.2 71.3 1.4 65.9 0.6 14.6 0.5 14.2 1.0 26.8 1.4 32.1 0.8 2.9 0.4 2.1 0.4 2.0 0.1 2.0 0.2 45.0 1.5 82.6 1.1 51.7 1.5 16.3 1.0 3.3 0.1 1.1 0.2 Adult’s mummy (PIXE) 98.2 1.7 0.1 0.1 0.2 0.1 89.4 7.9 2.8 1.0 1.3 0.4 94.0 3.3 2.7 1.3 0.9 0.4 81.1 18.1 0.8 0.6 0.5 0.1 Single analysis for each pendant 74.3 21.7 4.0 70.5 23.4 6.1 67.8 23.3 8.9 74.3 19.7 6.0 77.4 19.9 2.7 68.2 23.5 8.2 74.7 21.6 3.7 74.5 21.7 3.9 73.9 19.6 6.5 81.8 16.2 1.9 71.5 22.9 5.6 72.3 22.6 5.1 68.7 22.6 8.7 77.9 18.3 3.8 76.8 21.4 1.8 73.0 22.3 4.7 Adult’s mummy (PIXE) Ring standard deviation Terminal Part (XRF) Child’s mummy (PIXE) A.1883.49.2 Low row standard deviation Upper row standard deviation 84 3 93 1 87 3 80 10 1 5 2 9 2 15 6 3 2 0 4 0 5 81 0 84 12 1 9 7 1 7 83 5 86 3 9 2 9 3 8 4 5 2 115 content ranging from 2 to 20%. he exceptions from the Qurneh burial are made of electrum, as deined by Pliny, i.e. a gold alloy in which the silver content is greater than 20% (Natural History, XXXIII): in the girdle, the silver content reaches 52.5-53% in the wallet beads and the barrel beads, and in the child’s necklace, the silver content varies between 27 and 32%. he tail of the Harāgeh pendant is also made of electrum containing 51.7% silver. herefore, variation in alloys does not seem to be linked to chronology, but rather to type and presumably use of the objects, as well as perhaps to colour (Schorsch, 2001; Frantz and Schoesch, 1990), considering that the Egyptian words for gold (noub) and silver (hat) are colour-oriented. Copper is generally present in quantities of less than 2%; however, some objects exhibit a copper content higher than 3%. he wallet beads of the Qurneh girdle contain 3.7% copper, while in the barrel beads the copper content varies from 2.8 to 7.4%. We found 3.3% copper in the tail of the ish-pendant, and, inally, the ly-pendants exhibit up to 9% copper. It seems unusual for Egyptian gold deposits to contain more than 2% copper (Ogden, 2000), so most of the jewellery could have been manufactured using native alloys. For the wallet and barrel beads of the girdle, and the tail of the pendant, copper in the range of a few percent could have been added to a silver-rich electrum alloy. For the ly-pendants, the alloys could have been produced by adding copper to a native alloy with a relatively high silver content of 25% (such as the one that is found, for example, in the Qurneh child’s necklace). he addition of copper to debase gold is usually reported from the 18th Dynasty onwards, with copper-contents reaching 75% in some artefacts (Lucas, 1962: 229; Ogden, 2000). However, the amount of copper found in the barrel beads of the girdle clearly indicates that this process was in use since the 17th Dynasty. he Qurneh adult jewellery forms a speciic group in Figure 3, with a relatively high gold content ranging between 87.8% and 95.2%. he child’s earrings show a similar composition, with 82.5% and 83.7% gold, but with a higher level of copper, which may indicate an addition meant to harden the alloy. According to Ogden (2000), Egyptian gold during the 2nd millennium BC typically contained 15-30% silver and copper in amounts lower than 2%, while the use of the cementation method to reine gold was not practised there before the 1st millennium BC (Ramage and Craddock, 2000). Table 2: Compositional results (in wt%) for all the objects analysed by PIXE, XRF and SEM-EDS. Tableau 2 : Compositions (en %) des alliages des objets analysés par PIXE, FX et MEB-EDS. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 111-119 116 Such expected compositions were identiied in the Qurneh child’s necklace, the Harāgeh pendant, and one of the Amarna inger-rings (A.1883.49.8), as well as in published jewellery from the tomb of the foreign wives of hutmose III (Lilyquist, 2003). Indeed, Lilyquist (2003) argues that, in the collections of the Metropolitan Museum, only forgeries have alloys with gold content above 80% 4. Nevertheless, three of four items analysed from the tomb of Yuia and Tjuiu (Tomb KV46, early 14th century BC) are at least 82% gold (Lucas 1962: 545), and one contains 96.4% gold, 1.9% silver, and less than 1% copper (Lucas 1962: 545) 5. At NMS, the Qurneh woman’s earrings contain 95.6% gold, 4.1% silver and 0.2% copper 6, while one Amarna inger-ring (A.1883.49.2) contains 98.2% gold, 1.7% silver and 0.1% copper. Such analyses indicate that Ancient Egyptians had access to gold of exceptional purity, either from alluvial deposits, or by importing from abroad (James, 1992). With regard to the Qurneh jewellery, at least, Nubian gold must be considered a possibility, because the burial itself shows deinite cultural links with Nubia (Eremin et al., 2000; Manley et al., 2002). However, without further studies of contemporary objects, we cannot establish speciic provenances for the placer gold used to manufacture the jewellery. Lore G. TROALEN et al. Qurneh jewellery clearly exhibits the use of hard soldering in the fabrication of the adult’s gold earrings and bracelets, and of the child’s earrings, as well as the necklace. he child’s earrings seem to be melted from over-heating while hard soldering to create joins between the diferent rings (Fig. 4a); Petrie also noted that they “had been over-heated while on a mandril in the furnace for soldering; the solder had stuck them together, and they parted and began to drop away, being half melted” (Petrie, 1909). PIXE analyses of the soldered area show an increase in the copper content to 7%, with a silver content of 12%. In the soldered area of the adult’s bracelets, the copper level reached 6%, with 10% silver (Table 2). he XRF investigation of the terminal part of the Qurneh necklace gave an estimate of 80% gold, 15% silver and 5% a Solder Analysis here are but few published analyses of solders from Egyptian jewellery, with the exception of the study of a necklace from the burial of Wah, dated to the late 11th Dynasty (Schorsch, 1995), and of some objects dated to the 18th Dynasty (Lilyquist, 2003). he common joining technique in Ancient Egypt might have been colloidal hard soldering (or difusion bonding), using a copper salt and organic adhesive. Hard soldering using an alloy with a lower melting point than that of the alloy to be soldered (Maryon, 1949) is considered less usual (Ogden, 1992; Lilyquist, 2003). However, on beads from the Wah burial, Schorsch (1995) found an atypical joining technique, perhaps involving hard soldering by adding diferent amounts of copper to the basic alloy; and the investigation of a ring from the Qurneh necklace has shown the use of a hard solder alloy made by adding copper (Tate et al., 2009). Distinguishing between diferent soldering techniques is diicult when using only non-invasive analyses. However, the 4. In some apparent instances of forgery, the gold composition is exactly 22 carat, corresponding to English gold coinage from the early 20th century (Lucas 1962; Lilyquist 2003). 5. Compositions from assay analysis by W.B. Pollard (1908). 6. hese new analyses conirm the surface analysis undertaken by Eremin et al. (2000). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 111-119 b Figure 4: Detail of solder in (a [See colour plate]) Qurneh child’s earring (A.1909.527.43) by OM and (b) Qurneh adult’s earrings (A.1909.527.18) by SEI. Figure 4 : Détail de la soudure des boucles d’oreilles de Qurneh (a [Voir planche couleur]) de l’enfant (A.1909.527.43) sous loupe binoculaire ; (b) de l’adulte (A.1909.527.18) au MEB. Technological study of gold jewellery pieces dating from the Middle Kingdom to the New Kingdom in Egypt copper. he PIXE investigation of the solder between the tubes of the adult’s earrings indicated about 2% more copper than for the main body (Table 2). All these compositions correspond to a natural alloy with a fusion temperature slightly lower than the composition of the objects (Scott 1990), as is the case with the hard soldering of the beads of the necklace (Tate et al., 2009). Granulation Granulation is obtained by joining spheres to a metal sheet using hard soldering techniques or copper salt (Lilyquist, 1993). One of the irst examples of granulated objects in Egypt is a gold cylinder-pendant, also from el-Harāgeh, which shows the use of hard soldering (Ogden, 1992). A similar morphology in the joining areas is found for the granules of an Amarna inger-ring (A.1883.49.2), details of which can be seen in Figure 5. EDS surface analysis of the granules showed that they are heterogeneous in composition (top: 2.8% copper, 7.9% silver, 89.4% gold; bottom: 2.7% copper, 3.3% silver, 94.0% gold). Although these compositions are unlikely to correspond to the sulk composition – due to surface enrichments this could indicate a decoration undertaken in two stages. Experimentally, it is diicult to obtain an accurate quantitative analysis on such three-dimensional objects, which is probably the reason for the high variability of copper and silver analyses in the soldered areas. However, the indication is that in the top row, the copper content ranges from 5 to 13%, and silver Figure 5: Detail of granulation observed on the bezel of the ingerring A.1883.49.2 by SEI. Figure 5 : Détail de la granulation du chaton de la bague A.1883.49.2 au MEB. 117 from 7 to 12%, while in the bottom row the copper content ranges from 4 to 7%, and silver from 6 to 11%, which could indicate the use of separate hard solders in each row, with diferent melting points. Construction and wirework he Amarna inger-rings show a similar mounting technique, whereby a wire is coiled around the hoop and enters the bezel (Fig. 6a). his may be analogous to technology from late 5th millennium Egypt, when rings could be made with “small strings of beads, gold-foil bands and wires of copper or silver closed by twisting the ends together” (Andrews, 1990). In the case of inger-ring A.1883.49.8, the round section of the hoop is reduced, and has been worked into a wire that crosses the bezel and is coiled upon each extremity. In the case of inger-ring A.1883.49.2, a much thicker hoop has been worked at the terminals into a thinner section and passed through the bezel, while, in addition, a wire is coiled round the hoop, crossing the bezel. Similar techniques may be seen in inger-rings in the British Museum (Andrews, 1991: 164-165, 175) and necklaces from the burial of the foreign wives of hutmose III (Lilyquist, 2003: 196). Hammering, block-twisting, strip-drawing and striptwisting have all been suggested as possible techniques for producing wires in antiquity (Oddy, 1977). Although no speciic work published on Ancient Egyptian wires exists, hammering was usual in Bronze Age jewellery according to Petrie, strip-drawing was identiied by Williams (1924), although this has been signiicantly questioned, while striptwisting was largely discussed by Carroll (1972). As regards Ancient Egyptian objects, Williams (1924) suggests the use of strip-drawing and strip-twisting, while Scheel (1989) suggests that thick wires were produced by strip-twisting and thinner wires by block-twisting. Hammered wire presents facetted surfaces; block-twisted wire has a more or less round section with a solid interior; while both strip-drawing and strip-twisting result in a round section wire with a hollow interior (Oddy, 1977). For the Amarna inger-rings, we identiied the following characteristics: the wire of A.1883.49.2 is relatively thick, with a section that is solid but sometimes not circular, whereas the wire of A.1883.49.8 is solid, rounder in section, and shows some longitudinal lines. In the wires of both rings, we observed laking and overlapping of gold, possibly corresponding to rolling and burnishing treatments (Fig. 6b). Excluding the efects of wear, we observe a homogeneous wire with a near round section and seams on the surface similar to those produced by strip-twisting. Further examination of the wires is in progress at the NMS in order ArcheoSciences, revue d’archéométrie, 33, 2009, p. 111-119 Lore G. TROALEN et al. 118 a turies BC. We have shown that the gold alloys used during this period are preponderantly produced with alluvial gold containing iridium-ruthenium-osmium inclusions. he alloys used to produce the studied jewellery range from high purity ones, with a gold content that can reach 95%, to electrum containing 51 to 53% silver. Apart from objects in which copper was certainly added to the alloy, the copper content was typically found to be below 2%. he use of iligree and granulation in the decoration of the Amarna inger-rings revealed the use of hard soldering, and traces of this technique were found in joints for the majority of the objects studied. he seams in the limited sample set of wires have sections that may indicate the use of strip-twisting. Acknowledgment b he authors would like to thank colleagues at NMS, in particular Henrietta Lidchi and Lesley-Ann Liddiard from the World Cultures Department, for giving us access to the objects and supporting the project; colleagues at C2RMF, especially the AGLAE team and hierry Borel for his help and advice on X-Radiography of gold objects; Katherine Eremin at the Harvard Art Museum for comments on her previous work; Susan La Niece and Nigel Meeks at the British Museum for their comments while writing the article, and inally the Eu-ARTECH project for funding the PIXE analysis. References Figure 6: Amarna inger-rings: (a) X-Radiograph of A.1883.49.8 showing the construction of the bezel and (b) detail of A.1883.49.2 showing the work of the wire coiled upon each end of bezel. Figure 6 : Bagues d’Amarna : (a) radiographie X de A.1883.49.8 montrant le montage du chaton et (b) détail de A.1883.49.2 illustrant le travail du il enroulé autour de l’anneau de chaque côté du chaton. to verify whether the same techniques were generally used throughout the 17th and 18th Dynasties. 4. CONCLUSION his study represents a irst approach to the production techniques of Egyptian gold jewellery by presenting results obtained for a group of items from the collections of the NMS, dated to a period between the 19th and the 13th cenArcheoSciences, revue d’archéométrie, 33, 2009, p. 111-119 AHMED, A.H., 2007. Diversity of platinum-group minerals in podiform chromitites of the late Proterozoic ophiolite, Eastern Desert, Egypt. Ore Geology Reviews 32: 1-19. ANDREWS, C., 1990. Ancient Egyptian Jewellery. London, he British Museum Press. CARROLL, D.L., 1972. Wire drawing in Antiquity. American Journal of Archaeology 76(3): 321-323. DRAN, J.C., SALOMON, J., CALLIGARO, T. and WALTER, P., 2004. Ion beam analysis of art works: 14 years of use in the Louvre. Nuclear Instruments and Methods in Physics Research B 219220: 7-15. EREMIN, K., GORING, E., MANLEY, W.P. and CARTWRIGHT, C., 2000. A 17th dynasty Egyptian Queen in Edinburgh? KMT Modern Journal of Egyptology 11(3): 32-40. FRANTZ, J.H. and SCHORSCH, D., 1990. Egyptian Red Gold. Archeomaterials 4: 133-152. GUERRA, M.F., 2008. An overview on the ancient goldsmith’s skill and the circulation of gold in the past; the role of x-ray based techniques. X-ray Spectrometry 37: 317-327. Technological study of gold jewellery pieces dating from the Middle Kingdom to the New Kingdom in Egypt GUERRA, M.F. and CALLIGARO, T., 2003. Gold cultural heritage objects: a review of studies of provenance and manufacturing technologies. Measurement Science and Technology 14: 15271537. JAMES, T.G.H., 1972. Gold Technology in Ancient Egypt. Mastery of Metal working methods. Gold Bulletin 2(5): 35-42. LILYQUIST, C., 1993. Granulation and Glass: Chronological and Stylistic Investigations at Selected Sites, ca. 2500-1400 B.C.E. Bulletin of the American Schools of Oriental Research 290/291: 29-94. LILYQUIST, C., 2003. he tomb of the three foreign wives of Tuthmosis III. New York, he Metropolitan Museum of Art. LUCAS, A. and HARRIS, J.R., 1962. Ancient Egyptian Materials and Industries. London, Edward Arnold Publishers Ltd. MANLEY, W.P., EREMIN, K., SHORTLAND, A. and WILKINSON, C., 2002. he facial reconstruction of an Ancient Egyptian Queen. Journal of Audiovisual Media in Medicine 25(4): 155159. MARTIN, G.T., 1974. he Royal Tomb at El-‘Amarna, I. he Objects. London: Egypt Exploration Society. MARYON, H., 1949. Metal Working in the Ancient World. American Journal of Archaeology 53(2): 93-125. MEEKS, N.D. and TITE, M.S., 1980. he analysis of Platinumgroup Element inclusions in Gold Antiquities. Journal of Archaeological Science 7: 267-275. MOLLY, E.W., 1959. Platinum deposits of Ethiopia. Economic Geology 54(3): 467-477. ODDY, A., 1977. he production of gold wire in Antiquity. Handmaking methods before the introduction of the draw-plate. Gold Bulletin 3(10): 79-87. OGDEN, J., 1976. he so-called “platinum” inclusions in Egyptian goldwork. Journal of Egyptian Archaeology 62: 138-144. OGDEN, J., 1982. Jewellery of the Ancient world. London, Trefoil Books. 119 OGDEN, J., 1992. Interpreting the past: Ancient Jewellery. London, he British Museum Press. OGDEN, J., 2000. Metals, in P.T. Nicholson, I. Shaw (eds.), Ancient Egyptian Materials and Technology. Cambridge, Cambridge University Press, 148-175. PETRIE, W.M.F., 1909. Qurneh. London: BSAE Publishing. PLINY THE ELDER, Natural History, XXXIII, J. Bostock translation, London, Taylor and Francis. POLLARD, W.B., 1908. Gold, in J.E. Quibell (ed.), Tomb of Yuaa and huiu. Le Caire, Imprimerie de l’Institut français d’archéologie orientale, 77-78. RAMAGE, A. and CRADDOCK, P., 2000. King Croesus’ Gold. Excavations at Sardis and the History of Gold Reining. London, British Museum Press. ROYAL MUSEUM REGISTER OF SPECIMENS. Art and Ethnographical Department. Volume 11 (Eleven). 1908 – 1913 (1908.311 – 1913.401): 82 SCHEEL, B., 1989. Egyptian Metalworking and Tools. Shire: Shire Egyptology, Shire Publications Ltd. SCHORSCH, D., 1995. he gold and silver necklaces of Wah: a technical study of an unusual metallurgical joining method, in C. Brown, F. Macalister, M. Wright (eds.), Conservation in Ancient Egyptian Collections. London, Archetype Publications, 127-135. SCHORSCH, D., 2001. Precious-Metal Polychromy in Egypt in the Time of Tutankhamun. Journal of Egyptian Archaeology 87: 55-71. SCOTT, D., 1990. Soldering with gold alloys in ancient South America. Archaeometry 32: 183-190. TATE, J., EREMIN, K., TROALEN, L., GUERRA, M.F., GORING, E. and MANLEY, W.P., 2009. he 17th Dynasty Gold Necklace from Qurneh, Egypt. ArcheoSciences 33. WILLIAMS, C.R., 1924. Catalogue of Egyptian Antiquities, Gold and Silver Jewelry and Related Objects. New York, he New York Historical Society. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 111-119 he 17th Dynasty gold necklace from Qurneh, Egypt Le collier en or de Qurneh daté de la XVIIe Dynastie, Égypte Jim Tate*, Katherine Eremin**, Lore G. Troalen*, Maria Filomena Guerra***, Elizabeth Goring**** and Bill Manley**** Abstract: In 1908, the archaeologist Flinders Petrie discovered a rich intact burial of an adult and child at Qurneh, near Luxor. Stylistically, the burial has been dated to the late 17th Dynasty, in the 16th century BC. he complete burial group came to Edinburgh in 1909. A recent examination of the rings of the necklace found with the adult burial is presented as part of a wider study of the mummy collections of the National Museums Scotland (NMS). he necklace shows sophisticated workmanship, with 1,699 rings threaded onto four strands, the rings having uniform diameter and thickness, and very few of them showing any visible joints. he necklace rings have been examined by Optical Microscopy, X-radiography, Scanning Electron Microscopy with Energy Dispersive X-ray analysis (SEM-EDS), air-path X-ray Fluorescence (XRF), and proton induced X-ray analysis (micro-PIXE). We summarise these indings and propose the method of manufacture. We also describe an experimental attempt to make joint-less rings in order to compare them with the originals. Résumé : En 1908, l’archéologue Flinders Petrie découvre à Qurneh, près de Louxor, sur la route de la Vallée des Rois, la tombe intacte et richement pourvue d’un adulte et d’un enfant. L’ensemble de la tombe a été stylistiquement daté de la in de la XVIIe dynastie, XVIe siècle av. J.-C., et son contenu fut transféré à Edimbourg en 1909. Dans le cadre d’un projet de recherche sur la collection des momies égyptiennes du Musée National d’Écosse (NMS), le collier trouvé associé à l’adulte a été étudié. Ce collier se compose de 1 699 anneaux enilés sur quatre rangs, chaque anneau présentant une section circulaire très uniforme avec très peu de joints apparents. Comment ses anneaux ont-ils été réalisés ? Le collier a été analysé par microscopie optique, radiographie de rayons X, Microscopie Électronique à Balayage couplée à un système d’énergie dispersive de rayons X (MEB-EDS), Fluorescence de rayons X (FX) et analyse par faisceaux d’ions (micro-PIXE). Les résultats de cette étude sont présentés et discutés en parallèle avec les résultats d’un travail de recréation expérimental, visant à réaliser des anneaux sans joints apparents. Keywords: Qurneh, mummy, necklace, rings manufacture, analysis. Mots-clés : Qurneh, momie, collier, manufacture anneaux, analyse. * National Museums Scotland, Department of Conservation & Analytical Research, NMCC – 242 West Granton Road, Edinburgh EH5 1JA. (j.tate@ nms.ac.uk), (l.troalen@nms.ac.uk) ** Straus Centre for Conservation, Harvard Art Museum – 32 Quincy Street, Cambridge, MA 02138, USA. (keremin@fas.harvard.edu) *** Laboratoire du Centre de Recherche et de Restauration des Musées de France, UMR171 CNRS – 14, quai François-Mitterrand, 75001 Paris, France. (maria.guerra@culture.gouv.fr) **** National Museums Scotland, Department of World Cultures – Chambers Street, Edinburgh EH1 1JF. (b.manley@nms.ac.uk) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 121-128 Jim TATE et al. 122 1. INTRODUCTION he NMS Mummy Project National Museums Scotland (NMS) has extensive Ancient Egypt collections. hey were mainly assembled from 1819 until the early 20th century, and include eleven mummies, over sixty (more or less fragmentary) decorated coins, mummy-boards and masks, and a great number of associated burial goods (Dodson and Manley, 2010). he NMS Mummy Project was initiated in order to raise professional and public understanding about the people whose bodies and belongings had been preserved with such care. Individual mummies and coins have been examined in a programme of whole body and dental X-radiography, CT-scanning, and material analysis and identification (Eremin et al., 2000; MacLeod et al., 2000; Buckley and Evershed, 2001; Manley et al., 2002). Perhaps the greatest treasure of the NMS Ancient Egypt collections is the intact burial of an adult and child in a simple pit-grave at Qurneh, near hebes (modern Luxor), discovered and excavated by Flinders Petrie on 30 December 1908 (Petrie, personal diaries). Petrie sent the entire contents of the burial to the Royal Scottish Museum (a forerunner of NMS) the following year. he grave itself was on the way to the ancient cemetery at the Valley of the Kings, an area containing the tombs of several kings and queens of the 17th Dynasty. he burial was centred on a large anthropomorphic coin with feathered decoration, painted dark blue, and gilded. he ine decoration on the coin (generally ) was known as rishi, from the Arabic for ‘feathered’, a heban innovation dating from the 17th Dynasty, i.e. the early or mid-16th century BC (Miniaci, 2007), surpassed only by those of two 17th Dynasty kings discovered near Qurneh. Only the coins of two heban queens from the early 18th Dynasty, i.e. the late 16th century BC, are larger. herefore, dating the coin to the 16th century BC seems straightforward, and other elements of the burial, such as the ceramics, tend to conirm this. he rishi-coin contained the mummy of a young adult female, unwrapped by Petrie at the time of discovery and now complete, but disarticulated. he grave goods included furniture, ceramics, food-oferings, a sceptre, an inlaid headrest, a decorated oil-horn, and jewellery. A simpler chestshaped coin containing the remains of a young child was also unwrapped by Petrie. he wealth of the burial is evident from the jewellery found with both mummies, which Petrie described as “the largest group of goldwork that had left Egypt” (Petrie, 1932). his wealth is remarkable considering that Egypt was politically divided, and the kings at ArcheoSciences, revue d’archéométrie, 33, 2009, p. 121-128 hebes had no direct access to the wealthy trade routes of the Mediterranean, nor to the gold-rich mines of Nubia. Even the burials of kings of this period were quite simple. he most obvious explanation is that the occupants were members of the 17th Dynasty royal family at hebes, which would certainly accord with the location of the tomb at Qurneh (Eremin et al., 2000; Roehrig, 2007). Nevertheless, the woman’s speciic identity remains a mystery, and none of the studies carried out so far have allowed us to establish whether the child and adult woman were related, although that would be what we may expect of a shared burial in Ancient Egypt. We have not yet undertaken DNA analysis of the human remains, but have concentrated on studying the grave goods, particularly the jewellery, to see what these could reveal. he jewellery is described in Petrie (1909) and Eremin et al. (2000). Both the adult and child had necklaces and earrings: in this study, we concentrate on the adult necklace (NMS A.1909.527.19, Fig. 1a), but refer also to some shared similarities with the child’s earrings (NMS A.1909.527.43). he Gold Necklace he style of the necklace is straightforward: a large number of gold rings strung together to form four decorative bands that combine into each side of a clasp (Fig. 1), the two halves of which are made from four groups of eight Figure 1: (See colour plate) Coin of the adult woman and necklace showing clasp and pin. Figure 1 : (Voir planche couleur) Cercueil de la femme adulte et collier illustrant le fermoir et l’épingle. he 17th Dynasty gold necklace from Qurneh, Egypt rings fused together. he necklace is closed by itting the two parts of the clasp tightly together and inserting a short gold pin through the centre of four overlapping gold loops from either side. Only a small number of necklaces of this type and date are known from museum collections (Roehrig, 2007; Binder, 2008). Petrie noted that the rings were “thick enough not to collapse when squeezed between the ingers” (Petrie, 1909). He measured and weighed them and counted 1,653 rings. A re-count carried out during conservation works in 2006 increased the number to 1,699 (Melville, pers. comm.), possibly by including some of the fused rings from the clasp. here is no record of any additional rings being found and added to the necklace, although Petrie did re-string it because “a very thick pack of ibre illed the rings; so compact that only two rings could be dragged of at a time” (Petrie, 1909). he rings are full rings, that is to say the vast majority are not open, like many of the rings making up parts of other Egyptian jewellery, including a smaller necklace found with the child burial (NMS A.1909.527.11). Petrie stated that “each has been soldered to join it” (Petrie, 1909), but in fact only a very small number of them show external marks that clearly support this statement. 2. METHODS Technical examination Optical and SEM imaging using both secondary electron (SEI) and backscattered electron (BSE) detectors were used to examine the surface appearance and details. he SEM was a Camscan MX2500 operated in both high vacuum and controlled pressure mode. We focused on three of the rings removed during conservation, which were placed on a standard SEM stub and imaged directly. he composition was determined using a hermoNoran Vantage Energy Dispersive (EDS) system (recognising the possibility of slight surface enrichment of gold from the bulk composition). X-radiography was used under various experimental conditions to look for seams or joints. he three rings were analysed along with two earrings from the child burial and other gold jewellery items by XRF and PIXE (see Troalen et al., 2009). he XRF system is an air-path Oxford ED2000SW instrument conigured to provide a collimated X-ray beam of 1.5x2 mm at the object, the precise spot identiied by crossed lasers and using XpertEase fundamental parameters analysis calibrated against internal NMS laboratory gold/silver/copper standards (GCS5 and Rolled 123 Gold). he PIXE 1 at C2RMF used 3 MeV protons (Dran et al., 2004) and experimental conditions for gold (given in Guerra and Calligaro, 2004 and Guerra, 2004) calibrated with in-house gold/silver/copper standards Gold 6917 and the same NMS standards. Methods used to make the rings he question of how the rings were made was asked before it became possible to remove and examine some of the rings in technical detail. We commissioned experimental reconstructions from a modern goldsmith, reported here along with the subsequent technical examination. hree methods that could have been used to make such uniform rings are: – cutting wire segments to uniform lengths, looping into circles and soldering the joints; – casting in a single or multiple ring mould; – punching each ring individually. he irst method would require uniform lengths of wire, thin strips cut from a hammered sheet, or short cast-segments. Wire could also be made by strip- or block-twisting (there is no evidence of Egyptian drawn wire, Andrews, 1990), wound round a uniform diameter hard (bronze?) rod, and cut to produce a collection of open rings which could then be joined by difusion bonding or soldering. For the former, a ine copper-rich powder (possibly with an organic glue) would be applied to a tight joint and strongly heated (Andrews, 1990; Ogden, 2000; Lilyquist, 2003). For the latter, a more open joint could be illed with a lower melting point alloy. Burnishing would hide any visual traces of the joint. Casting would entail a precision two-part mould or a simpler open mould. More complicated lost-wax casting would entail the initial modelling of the rings in wax prior to encasing them in a clay mould, melting the wax, and illing the negative volume with molten gold alloy. Each method could theoretically produce individual rings or multiple groups of rings. Lost-wax cast rings would have no joint, but there would be a sprue; lash lines at the joints might show from two-part moulds. Punching could be performed from a lat sheet or lattened blob of gold, drilling or punching a central hole which could be enlarged to produce rings with no joint. here are several images of craftsmen using single and multiple drills to bore holes in stone beads (Andrews, 1990). 1. hanks to funding from EuArtech. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 121-128 Jim TATE et al. 124 Before we were able to examine individual rings, we investigated the feasibility of punching and commissioned a modern gold worker (Ms Jacqueline Mina) to make ‘joint-less’ gold rings to look like those from the necklace. She made equal weight gold droplets, lattened them, and then formed a central hole with a series of steel punches, annealing the rings throughout to maintain workability. he intention was feasibility rather than reproduction, and the work used 24-carat gold, steel tools, and a gas lame, rather than more archaic materials. 3. RESULTS Examination of the necklace he form of the rings was shown in optical and SEM images to be very regular (Fig. 2). he three rings examined individually measured between 4.2 and 4.5 mm external diameter, 0.8 mm thick, and with a 3.5 mm internal hole. Each ring has a ‘D’ shaped section, so they were not cut from simple lat strips. he outer surface shows tool marks: ine parallel scratches running around the circumference, strong radial gouges, and areas of overlapping metal at the edges (Fig. 2d). We interpret the gouges as signs of the removal of metal using a very ine pointed chisel (the point around 10 microns across) before hammering the edges of a c b d Figure 2: (a) Optical image of rings; (b), (c) & (d) SEM-BSE images showing dimensions and tool marks. Figure 2 : (a) Images sous binoculaire des anneaux (b), (c) & (d) Images MEB-BSE montrant les dimensions et les marques d’outils. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 121-128 the segment lat and smoothing and polishing. he inner surface of each ring is lat, but less smooth. here are no spiral lines from strip or block twisting, nor hammering marks on the lat inner surface, so the rings were not formed from round-section wire. here is no evidence of any lashing line on the inner surface, as expected from a two-part mould; instead, the only concentric internal line is slightly recessed. he rings are symmetric, and their regularity is inconsistent with being cast in an open mould. hese observations conirm the assumption that they were not cast as complete rings. he rough central band on the outer surface of the ‘D’ is similar to the inner surface of the ring: we interpret it as part of the original surface of the metal, which was cast as a wire segment into a ‘D’ shaped mould, the outer surface subsequently being worked to a smooth inish prior to looping each segment to form a ring. However, optical and SEM examination and X-radiography failed to characterise the joints in the majority of the rings on the complete necklace. here are very few open rings, and some with joint-like marks, i.e. the metal is thinner, or thickens into a bulge (e.g., the 4th ring in Fig. 2a). hese could be the result of poor casting or damage rather than fabrication. he joint he experimental reproduction demonstrated that it is perfectly possible to create joint-less rings, but that these did not match several of the features of the originals. It is evident from the SEM images (Fig. 3, a-f ) that the punched rings are less uniform, and that the proile of the inner surface is quite diferent, having a raised inner ridge. While increased symmetry might come with practice, it seems unlikely that either the regularity or the lat, rather porous looking, inner surface of the originals could be achieved. We do not believe therefore that the rings were made using a punch. Following the above examination and analysis, the more damaged of the three rings that had been removed was mounted and polished so that the existence or otherwise of a joint could be determined. he polished and etched ring is shown in Figure 4 (a and b), where the form of the joint is clear. he joint was made with added copper-rich solder (see below), the ends of the metal loop having been thinned to maintain the same thickness as the rest of the ring. he grain structure of the ring itself suggests working and annealing, while the solder shows a dendritic structure, similar to that shown in Scott and Doehne (1990). he thickness, structure and composition of the joint conirm that it is the result of an added solder, not difusion he 17th Dynasty gold necklace from Qurneh, Egypt A D B E 125 C F Figure 3: SEM-BSE images showing stages in the production of reproduction rings (ring diameter 5 mm). Figure 3 : Images MEB-BSE illustrant les étapes de production des recréations des anneaux (diamètre des anneaux 5 mm). Figure 4: (See colour plate) (a) Optical and (b) SEM-BSE images of polished and aqua regia etched area of joint in ring. Scale bar is 200 microns. Figure 4 : (Voir planche couleur) Images (a) sous binoculaire et (b) MEB-BSE de surfaces polies et attaquées à l’eau régal de la jointure de l’anneau. Barre de l’échelle 200 microns. bonding. his use of hard soldering with addition of a distinct metal alloy is considered less common than difusion bonding, but has been found occasionally on pieces from the ancient world (Lilyquist, 2003). Plotting the compositions on the Au-Ag-Cu ternary liquidus diagram provided by Scott and Doehne (1990) indicates a melting temperature of c. 950 °C for the solder, as opposed to just over 1000 °C for the ring, demonstrating a rigorous control of the temperatures during soldering. Comparison with other necklaces While there are many Egyptian gold necklaces from this period containing ring beads (as illustrated in Lilyquist, 2003), most appear to be square sectioned, often with quite clear (and open) joints. We have examined only two other necklaces of the 17th-18th Dynasty; one at the Metropolitan Museum, New York (Object 16.10.314) and one at the British Museum, London (inventory number EA 14693). Binder (2008) also refers to these necklaces and their distinct characteristics of the shebu-type. We are grateful for technical and analytical information regarding the necklaces in the Metropolitan and British Museum to Mark Wypyski and Deborah Schorsch from the Metropolitan Museum, and Nigel Meeks from the British Museum. he Metropolitan necklace is 335 mm long, with approximately 450 ring beads, each 6 mm in diameter (Roehrig 2007). here are hemispherical end-caps, but no clasp. he ring widths vary, and clear joints are present, either square or chamfered. Some of the jointed areas appear as swellings, presumably where there is excessive solder or the metal has begun to melt (Fig. 6). he SEM-EDS analysis shows that the joints have a higher copper content than the rings (4%, as compared to around 2%). he rings have visible platinum group inclusions, but have a higher silver content than those from Qurneh (see below). he SEM examination shows some tool marks and a line around the circumference similar to the Qurneh rings, but the ‘D’ shaped cross-section of the outer side of the rings is less triangular. he British Museum necklace has very inely inished rings with square joints. he clasp has a closing mechanism very similar to that from Qurneh. he rings lack visible PGE group inclusions and are of higher purity gold than those from Qurneh. here were no visible working marks or surface features similar to those seen on the other two necklaces. Compositional analysis he results of each of the experimental methods (Table 1) show that the Qurneh necklace, with its high percentage of gold (86-88%), silver (10-12%), and copper (2%), is in all probability made of alluvial gold. his is supported by the presence of inclusions of osmium-iridium-ruthenium (PGE group elements, Meeks and Tite, 1980; Ogden, 1976) on many of the rings (as well as on other pieces from the burial, see Troalen et al., 2009). A concern with any analysis of ancient gold is the extent of surface change, either deliberate or from corrosion during burial, a problem noted by Lilyquist (2003) for Egyptian ArcheoSciences, revue d’archéométrie, 33, 2009, p. 121-128 Jim TATE et al. 126 Figure 5: Compositions of the Qurneh gold, artefacts in Lilyquist (2003), and comparison of necklaces in the Metropolitan Museum of Art (courtesy of Mark Wypyski and Deborah Schorsch) and British Museum (courtesy of Nigel Meeks). Figure 5 : Compositions des objets de Qurneh, des objets publiés par Lilyquist (2003), et comparaison des colliers du Metropolitan Museum of Art (courtoisie de Mark Wypyski et Deborah Schorsch) et du British Museum (courtoisie de Nigel Meeks). a b Figure 6: (See colour plate) Necklace 16.10.314 in the Metropolitan Museum of Art: (a) optical image; (b) SEM BSC image (Courtesy of Mark Wypyski). Figure 6 : (Voir planche couleur) Collier 16.10.314 du Metropolitan Museum of Art: (a) image sous binoculaire (b) image MEB-BSC. (Courtoisie de Mark Wypyski). gold. he polished section of the necklace ring allowed us to compare directly the surface- and core-gold compositions using SEM-EDS. A small compositional change was found, but the composition determined from the core metal in the section was very close to the PIXE and XRF results (see Table 1). hus, while there may be some surface change resulting from the burial, it represents only a very thin layer which efects the SEM-EDS analysis, and the XRF and PIXE data are essentially corresponding to the bulk composition. Figure 5 compares the Qurneh gold with the published analyses by Lilyquist (2003) for ‘presumed ancient’ mateArcheoSciences, revue d’archéométrie, 33, 2009, p. 121-128 rial from the Metropolitan Museum collection, plus speciic data from necklace (16.10.314) 2 and preliminary SEM-EDS surface analysis data from diferent areas on the British Museum necklace 3. he alloy used for the adult’s necklace and for other pieces from the Qurneh burial (summarised in Troalen et al., 2009) is notably richer in gold and with a lower silver content than the majority of the comparable artefacts presented by Lilyquist (2003). Finally, we compared the adult’s necklace rings with the jewellery from the child’s burial. he child’s necklace difers signiicantly from the adult’s, and resembles the small ring bead strings illustrated in Lilyquist (2003); the rings are smaller than those of the adult, uneven in width, and have open joints; they were clearly formed by rolling metal strips into circular loops. heir composition is also very diferent from the adult’s necklace, with a high but variable silver content (see Troalen et al., 2009). In contrast, the child’s earrings resemble the rings of the adult’s necklace more closely, being ‘D’ shaped in section, with radial tool marks and a lat inner surface. he alloy is also similar, but with slightly higher copper and silver contents (Table 1). While in no way conclusive, this aspect does tentatively support an association between the adult’s and the child’s burials. 2. hanks to Mark Wypyski and Deborah Schorsch. 3. hanks to Nigel Meeks. he 17th Dynasty gold necklace from Qurneh, Egypt 127 Piece Cu (wt%) Au (wt%) Ag (wt%) Adult necklace rings 1-3: surface XRF 1.6 sd 0.1 86.0 sd 0.2 12.0 sd 0.6 Adult necklace rings 1-3: surface PIXE 1.8 sd 0.2 87.8 sd 0.2 10.4 sd 0.3 Child earring A PIXE 2.9 sd 0.4 82.5 sd 0.5 14.6 sd 0.5 Child earring B PIXE 2.1 sd 0.4 83.7 sd 1.2 14.2 sd 1.0 Adult rings 1&2: surface SEMEDS 1.1 sd 0.4 91.7 sd 1.5 7.2 sd 1.2 Adult ring 3: polished section SEMEDS 2.4 sd 0.2 85.5 sd 0.2 12.2 sd 0.4 Adult ring 3: polished solder SEMEDS 8.2 sd 0.3 80.4 sd 0.6 11.5 sd 0.3 Adult ring 3, area near joint PIXE 4.0 86.8 9.1 Table 1: Summary of compositional data from diferent techniques for the three adult necklace rings and the two child earrings (sd is the standard deviation from a number of separate analyses and does not include calibration estimates or counting errors). Tableau 1 : Les compositions obtenues par les diférentes techniques pour les 3 anneaux du collier de l’adulte et pour les deux boucles d’oreille de l’enfant (sd représente l’écart-type d’analyses indépendantes et ne considère pas les mesures de calibration et les erreurs de comptage). 4. CONCLUSIONS he rings of the necklace from the adult burial at Qurneh were most probably made from individual fragments of gold from an alluvial source cast into a ‘D’ shaped wire segments in an open mould, formed into rings, and joined using a metallic solder. Comparison of the surface and polished section of one ring conirmed the composition of the gold and provided clear evidence indicating that the surface composition of the necklace determined by PIXE and XRF is close to that of the bulk metal. Finally, SEM examination and PIXE analysis of the two damaged earrings from the child’s burial have shown that these are similar in form and composition to the adult’s necklace, adding to the likelihood that the two burials are associated. Acknowledgements NMS colleagues, especially Lesley-Ann Liddiard, for making the objects available; Ms Jacqueline Mina for the reproduction experiments; colleagues at the British Museum: Catherine Higgitt, Susan La Niece, Duncan Hook and Nigel Meeks, and at the Metropolitan Museum: Catharine Roehrig, Deborah Schorsch and Mark Wypyski, for making their results available; hierry Borel at C2RMF for radiography and EuArtech for funding for the PIXE analysis. David Scott is acknowledged for metallurgical advice, as are the referees for their helpful comments. References ANDREWS, C., 1990. Ancient Egyptian Jewellery. London, he British Museum Press. BINDER, S., 2008. he Gold of Honour in New Kingdom Egypt. he Australian Centre for Egyptology Studies 8. Oxford, Aris and Philips Ltd. BUCKLEY, S.A. and EVERSHED, R.P., 2001. Organic chemistry of embalming agents in Pharaonic and Graeco-Roman mummies. Nature 413: 837-841. DODSON, A.M. and MANLEY W.P., 2010. Life Everlasting. Ancient Egyptian Coins in National Museums Scotland. Edinburgh: NMS Publishing. DRAN, J.C., SALOMON, J., CALLIGARO, TH. and WALTER, P., 2004. Ion beam analysis of art works: 14 years of use in the Louvre. Nuclear Instruments and Methods in Physics Research B 219220: 7-15. EREMIN, K., GORING, E., MANLEY, W.P. and CARTWRIGHT, C., 2000. A 17th dynasty Egyptian Queen in Edinburgh? KMT Modern Journal of Egyptology 11(3): 32-40. GUERRA, M.F., 2004. Fingerprinting ancient gold with proton beams of diferent energy, Nuclear Instruments and Methods in Physics Research B 226: 185-198. GUERRA, M.F. and CALLIGARO, TH., 2004. Gold traces to trace gold. Journal of Archaeological Science 31: 1199-1208. LILYQUIST, C., 2003. he tomb of the three foreign wives of Tuthmosis III. New York, he Metropolitan Museum of Art. MACLEOD, R.I., WRIGHT, A.R., MCDONALD, J. and EREMIN, K., 2000. Historical Review, Mummy 1911-210-1. Journal of the Royal College of Surgeons of Edinburgh 45(1): 85-92. MANLEY, B., EREMIN, K., SHORTLAND, A. and WILKINSON, C., 2002. he facial reconstruction of an Ancient Egyptian Queen. Journal of Audiovisual Media in Medicine 25(4): 155-159. MEEKS, N.D. and TITE, M.S., 1980. he analysis of platinum group elements inclusions in gold antiquities. Journal of Archaeological Science 7: 267-275. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 121-128 128 MINIACI, G., 2007. Marietta at Dra Abu el-Naga and the Tomb of Neferhotep: a Mid 13th Dynasty Rishi Coin. Egitto e Vicino Oriente 31: 5-25. OGDEN, J., 1976. he so called “platinum” inclusions in Egyptian goldwork. Journal of Egyptian Archaeology 62: 138-144. OGDEN, J., 2000. Metals, in P.T. Nicholson, I. Shaw (eds.), Ancient Egyptian Materials and Technology. Cambridge, Cambridge University Press: 148-175. PETRIE, W.M.F., 1909. Qurneh. London, BSAE Publishing. PETRIE, W.M.F., 1932. Seventy Years in Archaeology. London, H. Holt & Co. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 121-128 Jim TATE et al. ROEHRIG, C.H. (ED.), 2007. Hatshepsut from Queen to Pharaoh. New Haven and London, he Metropolitan Museum of Art/ Yale University Press. SCOTT, D.A. and DOEHNE, E., 1990. Soldering with gold alloys in ancient South America: examination of two small gold studs from Ecuador. Archaeometry 32(2): 183-190. TROALEN, L., GUERRA, M.F., TATE, J. and MANLEY, B., 2009. Technological Study of Gold Jewellery from the 17th and 18th Dynasties in Egypt. ArcheoSciences 33. Preliminary observation on three Late Bronze Age gold items from Ras Shamra-Ugarit (Syria) Observations préliminaires de trois objets en or de l’Âge du Bronze récent de Ras Shamra-Ougarit (Syrie) Romain Prévalet * Abstract: Studies of technical mechanisms of ancient gold work from the Near Eastern Bronze Age still remain rare. he identiication of marks and reconstruction of manufacturing techniques and processes must involve a multidisciplinary approach, where visual inspection, photomicrographs, experimentation, and elementary analyses are considered. Within the framework of my PhD research on the decoration of ancient gold in the Levant, I conducted a preliminary study of three items, decorated with iligreed and granulated designs, which were discovered at Ras Shamra-Ugarit (Syria) and were dated to the Late Bronze Age. In the present paper, I describe the results of my irst observations of the technical characteristics of the iligree and granulation, as well as of the joining processes that were employed by the craftsmen of that famous Levantine kingdom at the end of the second millennium BC. Finally, I also discuss the issues that should be considered in future research. Résumé : Les études sur les techniques du travail de l’or ancien au Proche-Orient à l’âge du Bronze demeurent rares. L’identiication des traces et la reconstitution des techniques et des procédés de fabrication impliquent une approche pluridisciplinaire dans laquelle l’observation, les macrophotographies, l’expérimentation et des analyses élémentaires pourraient être considérées. Dans le cadre de ma thèse de doctorat sur la décoration des ors anciens au Levant, j’ai efectué une étude préliminaire de trois objets en or décorés de motifs iligranés et granulés, découverts à Ras Shamra-Ougarit (Syrie) et datés de l’Âge du Bronze récent. L’article propose les résultats d’une première observation des caractéristiques techniques des procédés de iligrane, de granulation et de soudure employés par les artisans de ce prestigieux royaume levantin de la in du IIe millénaire av. J.-C., ainsi que les problématiques qui devraient être envisagées dans le futur. Keywords: Filigree, Granulation, Joining, Ugarit, Syria, Bronze Age. Mots-clés : Filigrane, Granulation, Soudure, Ougarit, Syrie, Âge du Bronze. 1. INTRODUCTION Filigree and granulation are two techniques of decoration (Ogden, 1982) employed on Near Eastern gold jewellery for creating basic geometric patterns on beads, earrings or pendants as early as the 3rd millennium BC (Maxwell-Hyslop, 1971). According to the archaeological data, a technical system that consists of ixing gold wires or granules on a golden core by a metallurgical process appeared in the Levant at the beginning of the 2nd millennium BC at Byblos, on the Lebanese coast. In western Syria (northern Levant), there is evidence for such techniques from the Middle Bronze Age in the royal tombs of Ebla, and then from the second part of the 2nd millennium BC at Alalakh, Qatna and Ugarit (Nicolini, 1990; Lilyquist, 1993; Prévalet, in press). * Université Paris I Panthéon-Sorbonne, Institut d’Histoire de l’art et d’Archéologie – 3 rue Michelet 75006 Paris. (Romain.Prevalet@malix.univ-paris1.fr) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 129-133 130 Romain PRÉVALET he present paper provides a irst study of the technical characteristics of iligree and granulation at Ugarit, useful for reining our knowledge of gold jewellery manufacture in the workshops of this Levantine kingdom where diverse inluences, cultures and craftsmen interacted at the end of the 2nd millennium BC. 2. METHODOLOGY OF THE STUDY his analysis involved three gold items currently in the collections of the National Museum of Damascus in Syria: a pair of crescent-shaped earrings (RS 20.57, S 2584 and RS 20.26, S 2582 – plate 1), a lat round bead (RS 21.197, S 5744 – plate 2) and a composite bead (RS 21.60, S 5726 – plate 3). he description of the typology, shape and surface of the wires and granules, as well as the hypothesis related to the identiication of the joining processes, are based on direct inspection of the artefacts under a magnifying lens and with micrographs, as the irst step of the multidisciplinary methodology of my research. Plate 2: (See colour plate) Flat round bead. Planche 2 : (Voir planche couleur) Perle circulaire plate. 3. RESULTS OF OBSERVATIONS Wires Filigree is applied on the three items not so much to decorate them, but rather to underline the edges or the han- Plate 3: (See colour plate) Composite bead. Planche 3 : (Voir planche couleur) Perle composite. Plate 1: (See colour plate) Pair of crescent-shaped earrings. Planche 1 : (Voir planche couleur) Paire de boucle d’oreille à navicelle. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 129-133 ging hole, or to support elements. I have observed only one type of wire, round wire, but it was obtained by diferent methods. On the lat round bead, the two holes used for suspension are underlined by a plain hexagonal-section wire almost a millimetre in diameter (micrograph 1, 120X). he longitudinal rectangular facets are clearly imprinted on the wire, and the annealing granular surface shows superimposed layers and creases caused by the repeated impact of the hammer. hese wires were probably produced from a square rod and hammered so that the surface became more regular and rounder. Its function was to reinforce the thin stamped foil where depressions existed, and also to prevent the edges of the foil from cutting the wire from which the bead is sus- Preliminary observation on three Late Bronze Age gold items from Ras Shamra-Ugarit (Syria) 131 In contrast, the wires of the earrings are solid and regularly round, with a diameter ranging between 0.50 and 0.60 millimetres. On the border of the crescent shape, a twist of the rectangular-section strip clearly appears where the rolling was uninished (micrograph 3, 120X), and furrows may be observed on two wires around the barbs of the earrings. Both of these aspects may represent evidence of block-twisting wire. Granules Micrograph 1: Hammered wire. Micrographie 1 : Fil martelé. pended. his aspect explains its thickness and perhaps also the methods of manufacture. he technique employed to produce the wires examined on the composite bead and on the pair of earrings must be block-twisting (Mello et al., 1983). he square-section wires of the bead, from 45 up to 75/100e in diameter, are coarse with a lot of irregularities in shape and rough marks on their surfaces. Nevertheless, it is possible to observe slight twists of a few extremities that may result from the helical seams created while rolling the twisted wires (Oddy, 1977) (micrograph 2, 30X). In addition, it also seems that some parts of the wires bear hammered marks, such as facets and wide grooves resulting from the crushing of the edges of the square wire. Micrograph 2: Twist of a block-twisted wire. Micrographie 2 : Torsion d’un il tors roulé. he earrings and the lat round bead are decorated by granulation. he arrangement of the granules in linear, triangular and circular designs is well known from the Bronze Age in the Near East and adjacent areas (Maxwell-Hyslop, 1971; Nicolini, 1990). At Ugarit, there are medium calibrated granules of 40-60/100e, often regular; a majority of them is lattened by use. he bead has three large granules of 75/100e, and several are irregular in shape and sometimes completely elongated or deformed. In this case, the granules were not well calibrated and graded before their application on the item (cf. micrograph 5). Diferent methods for producing granules are known by alchemists, goldsmiths and specialists, and some of them may have been available to the ancient craftsmen (houvenin, 1973; Eluère, 1993; Nestler and Formigli, 1994; Carroll, 1974). For example, experimentation shows that bits of strip cut to almost the same dimensions and smelted in charcoal powder in a crucible up to the melting point of gold could produce medium and even coarse Micrograph 3: Uninished block-twisted wire. Micrographie 3 : Fils tors roulé non achevé. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 129-133 132 Micrograph 4: Brazing of granules. Micrographie 4 : Brasage de granules. Romain PRÉVALET which melts at a lower temperature than the melting point of the gold base. he use of a brazing alloy, which appeared in the Near East during the 4th millennium BC (Nicolini, 1990), can usually be distinguished by the presence of a thick joint, as for example on the composite bead, where the wires are holding the globules by a reddish bulky trail of soldering (cf. micrograph 2). he wires of the lat bead are ixed to the item by two bits of solder, one at the bottom against the base foil and one at the top. he granule patterns in a circular shape that decorate the bead often look looded on a thick and compact layer of soldering (micrograph 4). At the centre of the bead there is a round line of granules; each sphere is placed on an independent neck-shaped joint resulting from a copper salt joining (micrograph 5, 120X), a process practiced in the Levant at the beginning of the 2nd millennium BC, at Byblos. he association of alloy brazing and copper salt joining on the same object is also visible on the earrings. Most of their granulated designs are ixed on the gold foil with a copper salt join, as the link between granules of the triangles and the thin neck joint (cf. micrograph 3), in spite of a few thickening-outs or traces of solder at the level of the lines and the triangles, as well as the thick joints of the wires around the barbs, which might represent evidence of alloy brazing. Pieces of solder may have been added where the copper salt joining had not fused because of a lack of contact between the parts. 4. CONCLUSION AND DISCUSSION Micrograph 5: Copper salt joining of granules. Micrographie 5 : Soudure de granules aux sels de cuivre. non well-calibrated granules similar to those present on the jewellery of Ugarit. Joining system Microscopic inspection enables us to examine the joining. Two distinct processes were recognized: alloy brazing and copper salt joining, their combination resulting in an alloy ArcheoSciences, revue d’archéométrie, 33, 2009, p. 129-133 Filigree and granulation are not common techniques in Ugarit, as compared to repoussé and incision. Rolled or hammered round wires and medium-calibrated granules characterize those techniques, which are simple processes already known from at least the beginning of the 2nd millennium BC in the Levant. In terms of the joining operations, the craftsmen employed both alloy brazing and copper salt joining, according to their functions: the irst one was probably used more structurally, in the assemblage of the diferent parts of the object, while the second was more appropriate for minute decoration. However, alloy brazing could also have been practiced to join granules, as in the case of the lat bead. In addition, sometimes it seems that complete control of the joining processes, and more speciically a mastery of the temperature, had not been established, an aspect which led to overheating and looding. At this stage of the study, much direct inspection is necessary to sustain the author’s assumptions about these three Preliminary observation on three Late Bronze Age gold items from Ras Shamra-Ugarit (Syria) gold items representing the iligree and granulation techniques employed for the gold jewellery of Ugarit. In future research work, scientiic analysis and examination must be conducted in order to determine the elementary composition of the gold alloy, and to support the identiication of the methods of joining the wires and granules. hen, a multidisciplinary study of all gold jewellery in the Levant dating from the Bronze Age should be undertaken in order to investigate a larger collection of iligreed and granulated items, and to determine the technical choices between alloy brazing and copper salt joining. Acknowledgements I thank the Syro-French archaeological excavation team of Ras Shamra-Ugarit and the Directorate General of Antiquities and Museums, Syrian Arab Republic, for their permission to study and publish the gold inds from Ugarit. I should like to extend my gratitude to Alessandro Pacini for his technical advice. References CARROLL, D.L., 1974. A Classiication for Granulation in Ancient Metalwork. American Journal of Archaeology 78: 33-39. ELUÈRE, C. (ed.), 1993. Outils et ateliers d’orfèvres des temps anciens. Antiquités Nationales Mémoire 2. Saint-Germain-en-Laye, Société des amis du Musée des antiquités nationales et du château de Saint-Germain-en-Laye. 133 LILYQUIST, C., 1993. Granulation and Glass: Chronological and Stylistic Investigations at Selected Sites, ca. 2500-1400 B.C.E. BASOR 290: 29-94. MAXWELL-HYSLOP, K.R., 1971. Western Asiatic Jewellery ca. 3000612 B.C. London, Methuen and Co. Ltd. MELLO, E., PARRINI, P. and FORMIGLI, E., 1983. Etruscan iligree: welding techniques of two gold bracelets from Vetulonia. American Journal of Archaeology, 87, (4): 548-551. NESTLER, G. and FORMIGLI, E., 1994. Granulazione etrusca: un’antica tecnica orafa. Siena: Nuova imagine. NICOLINI, G., 1990. La technique des ors antiques. La bijouterie ibérique du VIIe au IVe siècle. Paris, Picard. ODDY, A., 1977. he Production of Gold Wire in Antiquity. Gold Bulletin 10(3): 79-87. OGDEN, J., 1982. Jewellery of the Ancient World. Trefoil, London. PRÉVALET, R., in press. Techniques de la bijouterie d’or de Ras Shamra-Ougarit: iligrane et granulation, in V. Matoïan, M. Al-Maqdissi, Y. Calvet (eds.), Études ougaritiques II, Ras Shamra – Ougarit XIX. Lyon, Maison de l’Orient et de la Méditerranée. THOUVENIN, A., 1973. La soudure dans la construction des œuvres d’orfèvrerie antique et ancienne. Revue archéologique de l’est et du centre-est 24(1): 11-68. WHITFIELD, N., 1990. Round Wire in the Early Middle Ages. Jewellery Studies 4: 13-28. WOLTERS, J., 1981. he Ancient Craft of Ancient Granulation: A Re-assessment of Established Concepts. Gold Bulletin 14(3): 119-129. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 129-133 La tombe mycénienne de Kazanaki (Volos) et le mythe de la Toison d’or he Mycenaean tomb of Kazanaki (Volos) and the myth of the Golden Fleece Vassiliki Adrimi-Sismani*, Maria Filomena Guerra** et Philippe Walter** Résumé : Parmi les sites mycéniens de hessalie, Dimini, situé près de Volos, pourrait correspondre au fameux centre mycénien d’Iolkos, ville gouvernée par Pélias et qui est à l’origine du mythe de Jason et de la Toison d’or, célèbre peau de bélier reliée à la récupération d’or des alluvions. Une nouvelle tombe mycénienne à tholos, datée de 1 350 av. J.-C., découverte en 2004 à Kazanaki, a livré un nombre très important d’objets en or. L’examen et l’analyse élémentaire de feuilles et disques d’or, colliers et perles de collier au LC2RMF et à Volos avec des équipements portables, nous a permis de faire une première approche des techniques orfèvres de la période mycénienne et de mettre en évidence l’utilisation d’or d’origine alluvionnaire, parfois avec ajout de cuivre, pour la fabrication de l’orfèvrerie. Abstract: Among the Mycenaean sites in hessaly, Dimini, situated close to Volos, could correspond to the famous Mycenaean centre of Iolkos, ruled by Pellias, the site that was at the origin of the myth of Jason and the Golden Fleece, the famed ram’s skin related to the exploitation of alluvial gold. A new Mycenaean tholos tomb discovered in Kazanaki in 2004, and dated to 1350 BC, brought to light a very important number of gold objects. he examination and analysis of gold foils and discs, necklaces, and necklace beads have been undertaken at the LC2RMF laboratory and at Volos with portable systems. In this contribution, we present a irst approach to the Mycenaean goldsmithing techniques and we demonstrate that the goldwork was produced with alluvial gold, sometimes with the addition of copper. Mots-clés : Mycènes, Toison d’or, analyse, orfèvrerie, Kazanaki. Keywords: Mycenae, Golden Fleece, analysis, goldwork, Kazanaki. 1. INTRODUCTION Une tradition mythologique très riche relie la hessalie à la civilisation mycénienne, à la famille royale d’Iolkos (AdrimiSismani, 2007) et au mythe de Jason et des Argonautes, envoyés en Colchide – la région du Caucase en mer Noire – pour chercher la Toison d’or. Environ 200 sites mycéniens ont pu être identiiés en hessalie, parmi lesquels Dimini, situé près du port de la ville moderne de Volos (Adrimi-Sismani, 2004), qui comporte des ateliers d’orfèvrerie, avec des outils en bronze, et d’importants entrepôts (Adrimi-Sismani et Godart, 2005). Ces espaces administratifs, économiques et religieux avec les habitats voisins de Palia/Kastro Volos et Pefkakia et quatre tombes à tholos destinées à l’enterrement de membres de familles royales, semblent correspondre au fameux centre d’Iolkos. * Archaeological Institut of hessalian Studies – 74-76 Gambetta Str., 38221 Volos Greece. (aiths@culture.gr) ** Laboratoire du Centre de Recherche et de Restauration des Musées de France, UMR 171 CNRS – 14, quai François-Mitterrand, 75001 Paris, France (maria.guerra@culture.gouv.fr), (philippe.walter@culture.fr) rec. Oct. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 135-141 136 Pendant les travaux efectués dans les environs nord de la ville moderne de Volos une quatrième tombe à tholos, de la période HR IIIA2 – HR IIIB1 (1350 av. J.-C.), a été mise à jour à Kazanaki, en 2004 (Adrimi-Sismani, 2008). La tombe comporte un long dromos, une entrée et une chambre funéraire circulaire de 6,7 m de diamètre, avec quatre tombes à fosse. Des signes incisés ont été repérés sur le linteau supérieur de la façade monumentale de la tombe. Des ossements de plusieurs corps ont été identiiés (Papanastasiou, 2006). Tous les objets, parmi lesquels un nombre important en or, et les ossements avaient été regroupés dans un seul tas, brûlés in situ et puis remis dans les tombes à fosse, qui ont été ensuite scellées. Pour approcher la question de l’exploitation de gisements d’or alluvionnaire, associés au mythe des Argonautes et de la Toison d’or, 17 feuilles et disques de la tombe de Kazanaki ont été déplacés au C2RMF pour être analysés à l’accélérateur AGLAE. D’autres éléments en or ont été étudiés in situ avec des équipements portables, pour faire une première approche aux techniques orfèvres utilisées dans la production d’objets mycéniens. 2. MÉTHODES Les objets sélectionnés pour ce travail ont été examinés sous loupe binoculaire ainsi que, au C2RMF, par radiographie X et par microscopie électronique à balayage (MEB) avec un Philips XL30 ESEM en mode SE à 20 kV. Leur analyse élémentaire a été efectuée à Volos avec un dispositif d’analyse par spectroscopie de luorescence des rayons X basé sur un tube à rayons X Moxtek à anode ine d’argent et fonctionnant à 35 kV et 95 μA. Le détecteur est une diode SDD AXAS-V de Ketek, refroidie par efet Peltier. Sa résolution en énergie est de 140eV à 5.9 keV dans les conditions usuelles de fonctionnement. La quantiication des données est obtenue avec le logiciel PyMCA (Sole et al., 2007) par la méthode des paramètres fondamentaux, après modélisation précise de la source de rayons X. Les résultats quantitatifs ont été validés dans diférents cas (Gianoncelli et al., 2006 ; Viguerie et al., 2009) et pour les ors, une série de standards ont permis de vériier la justesse des résultats (Tableau 1). Au C2RMF l’analyse élémentaire est réalisée à l’accélérateur AGLAE au moyen des techniques PIXE et PIXE-XRF (particle induced X-ray emission et FX induite par PIXE) avec un faisceau de protons de 3 MeV extrait à l’air. La technique PIXE utilise deux détecteurs de Si(Li), un dédié à la détection des éléments majeurs et l’autre, avec un iltre sélectif de 75 μm de Cu, dédié à la mesure des éléments mineurs et traces (Guerra et Calligaro, 2004). La technique ArcheoSciences, revue d’archéométrie, 33, 2009, p. 135-141 Vassiliki ADRIMI-SISMANI, Maria Filomena GUERRA, Philippe WALTER Lyon Allemand 6917 mesuré certiié Lyon Allemand A mesuré certiié Lyon Allemand 893 mesuré certiié Authentico E-2 mesuré certiié Authentico A-1 mesuré certiié Authentico A-2 mesuré certiié Authentico E-1 mesuré certiié Au % 74,2 76,0 90,1 92,0 72,6 75,2 56,7 56,0 91,0 92,6 73,8 74,6 53,1 50,8 Ag % 18,3 17,0 5,3 4,0 7,4 5,7 42,8 43,5 7,7 6,4 23,4 22,6 46,8 49,2 Cu % 7,5 7,0 4,6 4,0 20,0 19,1 0,5 0,5 1,3 1,0 2,8 2,8 0,1 0,0 Tableau 1 : Valeurs certiiées et mesurées par luorescence à rayons X d’un groupe de standards. Table 1: heoretical values and compositional results for a group of standards analysed by XRF. PIXE-XRF utilise pour la mesure du Pt dans les alliages en or, une cible primaire de As et un iltre de 25 μm de Zn (Guerra et al., 2005 ; Guerra, 2004). 3. RÉSULTATS ET DISCUSSION Nous avons étudié 22 fragments de bandes d’or décorées en ondulé et d’épaisseur et tailles diférentes, 14 disques, 13 perles de collier à doubles spirales et à papyrus, et 2 colliers à rosettes et leurs de lis (types de perles selon Higgins, 1980). La composition élémentaire des objets se trouve sur le Tableau 2. Les feuilles et les disques Les feuilles et les disques sont fabriqués par martelage, des oriices de 130-140 μm de diamètre permettent leur application sur des vêtements (Higgins, 1980). Leur morphologie, l’absence de traces d’usure et leur ine épaisseur indiquent qu’il s’agit d’une production à but funéraire. Certaines feuilles, très ines et froissées, ne révèlent leur décoration ondulée qu’à la radiographie X. Leur composition est homogène (Tableau 2), seules deux feuilles possèdent des teneurs en Cu supérieures à 2 %, et correspond à l’électrum naturel (selon la déinition de Pline, Zehnacker, 1983). D’autres objets possèdent des teneurs identiques, comme une épingle à cheveux du IIIe millénaire, de typologie proche de l’épingle de Troie (Swann et al., 1997). 137 La tombe mycénienne de Kazanaki (Volos) et le mythe de la Toison d’or Feuilles (PIXE) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 en vague Feuilles (XRF) 15 16 17 18 19 20 21 moyenne (sans n°3) écart-type Disques (PIXE) 1 2 Disques (XRF) 3 4 5 6 7 8 9 10 11 12 13 14 Collier BE36161 (XRF) lis 1 2 3 4 5 rosette 1 2 3 4 5 moyenne écart-type Collier BE36137 (XRF) 1 2 3 4 5 moyenne écart-type Perles BE 30000 (XRF) 1 2 3 4 5 Perles BE36111 1 2 3 4 5 6 7 8 moyenne écart-type Tableau 2 : Résultats obtenus par PIXE et FX portable pour tous les objets de Kazanaki analysés et pour des objets mycéniens analysés par d’autres auteurs. Table 2: Compositional results for all the analysed objects from Kazanaki by XRF and PIXE and for a few Mycenaean objects analysed by other authors. Autres auteurs Schliemann 1878 Hackens 1993 Hartmann 1970 Vavelidis and Andreou 2008 Au % Ag % Cu % Pd ppm Sn ppm Sb ppm 76,1 75,4 87,1 79,2 76,2 71,5 76,0 74,8 76,5 75,4 70,7 73,7 76,4 75,4 75,7 21,7 22,8 12,0 19,0 21,7 25,0 22,7 22,9 21,5 22,8 25,3 24,2 21,5 22,8 21,2 1,4 1,4 0,6 1,5 1,5 2,9 0,9 1,4 1,5 1,4 2,7 1,3 1,5 1,4 2,0 100 120 70 90 100 130 140 100 100 100 130-320 <30 100 100 120 750 650 100 40 700 500 700 770 600 650 600 700 850 700 600 50 50 25 <30 40 50 50 50 60 100 30-800 50 50 75 50 78,0 78,0 78,3 81,0 78,2 79,2 77,7 76,4 2,5 20,6 20,5 20,2 17,2 20,1 19,2 20,6 21,6 2,0 1,5 1,5 1,5 1,8 1,7 1,6 1,7 1,6 0,5 71,7 63,6 23,1 31,4 4,8 4,3 70 200 150 70-350 <30 <30 83,9 76,5 84,8 70,8 82,2 75,0 83,4 83,0 76,4 85,3 68,0 68,2 14,7 19,7 14,0 25,1 16,4 19,4 15,2 15,3 20,8 11,7 29,9 30,9 1,4 3,8 1,3 4,2 1,4 5,6 1,4 1,6 2,8 3,0 2,1 0,9 73,4 74,4 73,7 74,0 74,7 73,0 71,4 73,8 75,0 75,2 73,9 1,1 25,7 24,8 25,4 25,5 24,8 25,9 28,2 25,2 24,4 24,2 25,4 1,1 0,9 0,8 1,0 0,5 0,6 1,1 0,4 1,0 0,6 0,6 0,8 0,2 80,0 80,3 83,2 81,9 81,1 81,3 1,3 18,3 18,4 15,8 17,3 17,6 17,5 1,0 1,7 1,3 1,0 0,9 1,3 1,2 0,3 82,2 74,2 76,1 78,1 80,2 15,4 16,9 21,3 17,1 15,9 2,3 8,9 2,6 4,8 3,9 87,6 86,5 86,1 86,9 84,0 87,4 88,8 88,6 87,0 1,5 9,2 10,9 11,8 10,0 13,6 10,5 9,6 9,3 10,6 1,5 3,2 2,6 2,1 3,1 2,4 2,2 1,7 2,1 2,4 0,5 Tombe 4 Tombe 4 coupe moyenne écart-type bande diadème feuille perle il globule Au % 89,36 73,11 82,45 0,78 87,85 75,1 90,3 77,5 99,8 78,8 Ag % 8,55 23,37 17,47 0,76 12,17 24,9 9,25 22 0,15 20,35 Cu % 0,57 2,22 0,08 0,04 0,5 0,55 <0,10 3,5 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 135-141 138 Vassiliki ADRIMI-SISMANI, Maria Filomena GUERRA, Philippe WALTER a c Les disques se séparent en trois groupes chimiques (Fig. 2), dont un s’approche de celui des feuilles. La couleur des feuilles se situe dans le vert-jaune du diagramme ternaire de couleur (Rapson, 1990), alors que la couleur des disques se situe dans la région jaune. Les colliers BE36137 et BE36161 et les perles BE36111 Le collier BE36137 est constitué de perles creuses en forme de rosette à six pétales. Ces mêmes perles intercalées de perles en forme de leur de lis constituent le collier BE36161. Les perles BE36111 sont à double spirale. Toutes les perles sont fabriquées de façon identique : une feuille lisse à l’arrière (parfois pliée sur la feuille avant) et une feuille avant (Fig. 3a), dont le motif est obtenu par pression sur une matrice (Treister et Hargrave, 2001). Les perles sont percées de chaque côté (vers l’intérieur) d’un oriice pour BE36137 (environ 230 μm de diamètre), de deux oriices pour BE36161 (environ 200μm de diamètre, Fig. 3b) et de trois oriices pour BE36111 (200-300 μm de diamètre). Malgré l’absence de soudures visibles sous binoculaire, il est possible que cette technique ait été utilisée (Schorsch, 1992). L’absence de traces d’usure indique que ces objets sont une production à but funéraire. La composition de chaque collier est très homogène (Tableau 1). BE36161 est réalisé avec un alliage très proche ArcheoSciences, revue d’archéométrie, 33, 2009, p. 135-141 b Figure 1 : (Voir planche couleur) Les perles : (a) en rosette et lis du collier BE36161 ; (b) à papyrus BE30000 ; (c) à double spirale BE36111. Figure 1: (See colour plate) he beads: (a) rosette and lily from necklace BE36161; (b) papyrus BE30000; (c) double volute BE36111. de celui des feuilles alors que BE36137 a une composition très proche de celle des disques de meilleure qualité. Malgré des teneurs en cuivre qui varient de 1,7 à 3,2 %, la teneur en argent de BE36111 est homogène mais inférieure aux teneurs mesurées pour les autres objets de Kazanaki. Les teneurs en argent et en cuivre des alliages se situent dans les valeurs espérées pour l’or natif et leurs couleurs se partagent entre les régions vert-jaune et jaune du diagramme ternaire de couleur (Rapson, 1990). Les perles BE30000 Cinq perles creuses à papyrus BE30000 (motif utilisé à l’époque minoenne en orfèvrerie et céramique, Evans, 1921), présentant des signes d’usure, sont de couleur, d’épaisseur de feuille et de facture distinctes des autres perles étudiées dans ce travail. Deux feuilles estampées sont soudées pour former la perle, qui est percée en haut et en bas, vers l’intérieur, d’oriices de 300 μm de diamètre. Les alliages possèdent des teneurs en argent de 15 à 21 % et en cuivre de 2 à 9 %. Leur couleur va du jaune rougeâtre au jaune verdâtre du diagramme ternaire de couleur (Rapson, 1990). Les alliages et l’origine de l’or La concentration d’or est inférieure à 89 % pour tous les objets analysés, ceci concordant avec la composition des feuilles de la tombe 4 de Mycènes (Schliemann, 1878), d’une coupe mycénienne dite de Syrie (Hackens, 1983), de deux objets mycéniens publiés par Hartmann en 1970, de feuilles et globules de creuset trouvés dans une tombe mycénienne en hessalie et des objets de Kastanas (Vavelidis et Andreou, 2007) (voir Tableau 2). La tombe mycénienne de Kazanaki (Volos) et le mythe de la Toison d’or 139 a b Figure 2 : Diagramme ternaire représentant les éléments Au-Ag-Cu (en %) pour les objets de la tombe de Kazanaki analysés par PIXE et par FX portable. Figure 2: Ternary diagram for the base-alloys of the objects from Kazanaki, Au-Ag-Cu (in %), analysed by PIXE and by portable XRF. Les deux colliers et les perles à double spirale possèdent une composition élémentaire très homogène pour chaque ensemble, ce qui permet de suggérer une fabrication unique par ensemble. Si l’on en exclue deux, les feuilles d’or possèdent aussi une composition homogène, contrairement aux disques (l’analyse d’un plus grand nombre pourrait révéler trois groupes distincts) et aux cinq perles à papyrus. Tous les alliages contiennent des teneurs en argent typiques des ors alluvionnaires (Raub, 1995), néanmoins les teneurs en cuivre de certains de ces objets ne peuvent être obtenues que par ajout volontaire (Guerra et Rehren, 2009 ; Hauptamann et Klein, 2009). La teneur en cuivre de certains disques et de trois perles à papyrus peut atteindre, respectivement, 6 et 9 %. Rappelons que la cémentation est un procédé métallurgique qui n’apparaît qu’au Ier millénaire av. J.-C. (Ramage et Craddock, 2000) ; la couleur et les propriétés des alliages à l’époque mycénienne dépendent alors de la teneur en argent des ors disponibles et ne peuvent être modiiées que par ajout de cuivre. Les concentrations en cuivre mesurées pour nos objets varient entre 0,5 et 9 %. Les concentrations mesurées par Vavelidis et Andreou (2007) pour des pépites provenant de gisements alluvionnaires en Grèce, pour Kastanas et pour les feuilles d’or ne dépassent pas 0,5 % (alors que l’argent varie de 1 à 22 %) ; des valeurs identiques ont été mesurées pour les objets mycéniens publiés par Schliemann (1878), Hartmann (1970) ; Figure 3 : Détails sous loupe binoculaire du montage des perles à : (a) double spirale BE36111 ; (b) rosette et lis du collier BE36161. Figure 3: Details under the stereomicroscope of the mounting of: (a) double volutes beads BE36111; (b) rosette and lily beads from necklaces BE36161. Hackens, 1983. Néanmoins, le globule de creuset publié par Vavelidis et Andreou, 2007 possède 3,5 % de cuivre. L’addition de cuivre à des ors alluvionnaires est une pratique qui semble avoir été employée aussi à la même époque en Égypte et en Géorgie (respectivement Troalen et al., 2009 ; Hauptmann et Klein, 2009) et en Anatolie dès le milieu du IIIe millénaire (Swann et al., 1997). En ce qui concerne la question de l’origine de l’or, l’analyse de 17 feuilles et disques par PIXE montre des teneurs en Sn qui peuvent atteindre 850 ppm (Tableau 2). Un groupe chimique avec de fortes teneurs en Pd et Sn est accompagné d’une droite (Fig. 4) où s’alignent les autres objets, notamment les trois disques analysés. La présence de Sn et/ou Pt dans les ors est en général liée à l’exploitation de gisements ArcheoSciences, revue d’archéométrie, 33, 2009, p. 135-141 140 Vassiliki ADRIMI-SISMANI, Maria Filomena GUERRA, Philippe WALTER Figure 4 : Représentation des teneurs en Pd et en Sn (en ppm) mesurées par PIXE. Figure 4: Pd and Sn contents (in ppm) measured by PIXE. alluvionnaires (Dube, 2006). Néanmoins, aucune inclusion de platinoïdes (Meeks et Tite, 1980) n’a pu être décelée sur la surface des objets. Malgré la taille et l’épaisseur des feuilles d’or, nous avons estimé par PIXE-XRF les concentrations de Pt pour six d’entre elles. Les résultats montrent qu’une partie des objets possède des teneurs inférieures à la limite de détection (80 ppm) alors que les autres possèdent des concentrations estimées à 200-500 ppm. Seul un plus grand nombre d’analyses permettrait de vériier l’existence de deux groupes. La présence de Sn et de Pt dans les alliages conforte l’hypothèse de l’utilisation d’ors alluvionnaires à Volos pendant la période mycénienne. Néanmoins, le manque de données sur les caractéristiques géochimiques des sources d’or à l’Age du Bronze en Méditerranée, en Égypte et dans le Levant – nous rappelons les circuits commerciaux à cette époque, notamment le lapis-lazuli qui devrait provenir d’Afghanistan (Hughes-Brock, 1999) – nous empêche de proposer à présent une provenance pour ces ors. Il faudra néanmoins remarquer que les ors analysés par Hauptmann et Klein (2009) de mines en Géorgie exploitées à l’Âge du Bronze présentent de faibles teneurs en Sn et Pt. 4. CONCLUSION L’étude analytique de feuilles, disques et perles de colliers de la tombe à tholos de Kazanaki nous a permis de réaliser une première approche aux pratiques orfèvres et à l’origine de l’or en hessalie à l’époque mycénienne. Si l’on excepte les perles à papyrus, tous les objets ont vraisemblablement été produits dans un but funéraire, à partir de ines feuilles d’or, parfois décorées par estampage. Les perles sont creuses, obtenues par assemblage de deux coques et ensuite percées. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 135-141 Les disques et les feuilles sont percés pour application sur des vêtements. Les teneurs en argent varient entre 9 et 31 % et en cuivre entre 0,5 et 9 %. Si les teneurs en argent sont typiques des ors natifs, les teneurs en cuivre dépassant 2 % correspondent à des alliages volontaires. Nous pouvons suggérer la recherche d’efets polychromes par simple ajout de cuivre à des ors natifs, dont la quantité d’argent est variable, ce qui laisse un large choix à l’orfèvre. Les ensembles sont homogènes, sauf les perles à papyrus et les disques. L’analyse de la totalité des disques pourrait montrer trois groupes homogènes correspondant à trois séries de fabrication distincte. L’homogénéité des alliages pourrait, après analyse de tous les objets de la tombe, séparer les productions par enterrement. L’analyse PIXE et PIXE-XRF de feuilles et disques révèle la présence de Sn et Pt, éléments caractéristiques des ors alluvionnaires. Les ors mycéniens de Volos peuvent ainsi avoir été exploités grâce à l’utilisation de peaux de béliers, ou toisons d’or. Si l’origine alluvionnaire de ces ors est clairement prouvée, leur provenance reste non identiiée par manque de données géochimiques. Seule l’analyse d’objets provenant d’autres sites mycéniens et de sites en Géorgie chronologiquement proches, permettrait de montrer le rapport entre la hessalie et la Colchide prouvant que les ors de Kazanaki sont ceux de la mythique Toison d’or. Bibliographie ADRIMI-SISMANI, V., 2004-2005. Le palais de Iolkos et sa destruction, BCH 128-129, 1. Études, 1-54. ADRIMI-SISMANI, V., 2007. Iolkos: Myth, Archaeology and History, in Gordeziani R. et al. (eds), he Argonautica and World Culture (Phasis, Greek and Roman Studies 10,1), Tbilisi: 20-32. La tombe mycénienne de Kazanaki (Volos) et le mythe de la Toison d’or ADRYMI-SISMANI, V., 2008. The Mycaenean tholos tomb at Kazanaki, Annual Report of the British School at Athens : 59-61, pl. 103-105. ADRIMI-SISMANI, V. et GODART, L., 2005. Les inscriptions en Linéaire B de Dimini/Iolkos et leur contexte archéologique, Annuario della Scuola Archeologica di Atene LXXXIII, Serie III, 5- Tomo I: 47-69. DUBE, R.K., 2006. Interrelation between gold and tin: a historical perspective, Gold Bulletin 39(3) : 103-113 EVANS, A. 1921. he Palace of Minos at Knossos, vol. 1 : he Neolithic and Early and Middle Minoan ages, London, MacMillan and Co. Limited. GIANONCELLI, A., CASTAING, J., BOUQUILLON, A., POLVORINOS, A., et WALTER, P., 2006. Quantitative elemental analysis of Della Robbia glazes with a portable XRF spectrometer and its comparison to PIXE methods, X-ray spectrometry 35(6) : 365-369. GUERRA, M. F., CALLIGARO, T., DRAN, J.-C., MOULHERAT, C. et SALOMON, J., 2005. Development of a PIXE, PIGE and PIXEXRF combination for the analysis of the gold from the First Empire of the Steppes, Geoarcheological and Bioarchaeological Studies 3 : 343-346. GUERRA, M.F. 2004. Fingerprinting ancient gold with proton beams of diferent energy, Nucl. Instrum. and Methods B 226 : 185-198. GUERRA, M. F. et CALLIGARO, T., 2004. Gold traces to trace gold, Journal of Archaeological Sciences 31, 1199-1208. GUERRA, M. F. et REHREN, Th., 2009. In-situ examination and analysis of the gold jewellery from the Phoenician tomb of Kition (Cyprus), ArchéoSciences 33. HACKENS, T., 1983. Gold Jewellery: craft, style & meaning from Mycenae to Constantinopolis, Louvain-la Neuve, BelgiumArt and Archaeology Publications. HARTMANN, A., 1970. Prähistorische Goldfunde aus Europa, Band 2: Spektralnalytische Untersuchungen und deren Auswertung, Berlin, Gebr. Mann Verlag. HAUPTMANN, A. et KLEIN, S., 2009. Bronze Age gold in southern Georgia, ArchéoSciences 33. HIGGINS, R.A., 1980. Greek and Roman jewellery, 2nd edition, University of California Press, Berkeley and Los Angeles. HUGHES-BROCK, H., 1999. Mycenaean beads: Gender and social contexts, Oxford journal of archaeology 18(3) : 277-296. MEEKS, N.D. et TITE, M.S., 1980. he analysis of Platinumgroup Element inclusions in Gold Antiquities. Journal of Archaeological Science 7: 267-275. 141 PAPANASTASIOU, A., 2006. To osteologiko yliko apo to tholoto tapho ste thesi Kasanaki, Proceedings of the 2nd Conference “Archaeologiko Ergo hessalias kai Stereas Ellados”, Volos-University of hessaly, 151-161. RAMAGE, A. et CRADDOCK, P.T., 2000. King Croesus’ gold: Excavations at Sardis and the History of Gold Reining. London, British Museum Press. RAUB, Ch. J., 1995. he metallurgy of gold and silver in prehistoric times, in Prehistoric Gold in Europe, G. Morteani & J. P. Northover eds, 243-259. RAPSON, W. S., 1990. he metallurgy of the coloured carat gold alloys. Gold Bull. 23(4) : 125-133. SCHLIEMANN H., 1878. Mykenae und Tiryns, Leipzig, F. A. Brockhaus. SCHORSCH, D., 1995. he gold and silver necklaces of Wah: a technical Study of an unusual metallurgical joining method, in Conservation in ancient Egyptian collections, Brown C., Macalister F., Wright M. (editors), Archetype Publications, London: 127-135. SOLÉ, V. A., PAPILLON, E., COTTE, M., WALTER et P., SUSINI, J., 2007. A multiplatform code for the analysis of energy-dispersive X-ray luorescence spectra, Spectrochimica Acta Part B-Atomic Spectroscopy 62(1) : 63-68. SWANN, C.P., BETANCOURT, P.P., FLEMING, S. et FLOYD, C.R., 1997. PIXE analysis of Trojan gold jewelry, Nucl. Instrum. and Methods B130 : 320-323. TROALEN, L., GUERRA, M.F., TATE, J., MANLEY, W.P., 2009. Technological study of gold jewellery pieces dated from Middle Kingdom to New Kingdom in Egypt. ArchéoSciences 33. VAVELIDIS, M. et ANDREOU, S., 2008. Gold and gold working in Late Bronze Age Northern Greece, Naturwissenschaften 95 : 361-366. TREISTER, M Y. et HARGRAVE J., 2001. Hammering Techniques in Greek and Roman Jewellery and Toreutics, Colloquia Pontica 8, Brill Academic Publishers, Leiden. VIGUERIE, L. de, DURAN, A., BOUQUILLON, A., SOLÉ, V. A., CASTAING, J. et WALTER, P., 2009. Quantitative X-ray luorescence analysis of an Egyptian faience pendant and comparison with PIXE, Anal Bioanal Chem, DOI 10.1007/s00216-0092974-7. ZEHNACKER, H., 1983 (traduction). Pline l’Ancien, Histoire Naturelle, livre XXXIII, Paris, Belles Lettres. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 135-141 he Phoenician gold jewellery from Kition, Cyprus L’orfèvrerie phénicienne en or de Kition, Chypre Pavlos Flourentzos* and Maria Luisa Vitobello** Abstract: In 1998, a built tomb was excavated on the southern outskirts of the town of Larnaca. he tomb was labelled MLA 1742 and included a unique collection of jewellery that likely adorned the deceased. heir assemblage totalled twenty three pieces belonging to various jewellery typologies, manufactured in gold and ornate with semi-precious stones. he tomb shows several ritual similarities with the famous Royal Tombs of Salamis. It can be suggested that a member of the Kition Phoenicians’ upper classes was buried here, possibly even a member of the royal family of this city. he tomb is dated to a period around the end of the 8th century BC, when Cyprus was under a strong orientalising inluence, due in all probability to the presence of Phoenician settlers in Kition. Various goldsmithing techniques, such as granulation, cloisonné, etc., were applied in the production of artefacts. Some of the jewellery typologies are quite rare or even unique. he rings and swivel rings found in the tomb show a strong Egyptian inluence, revealed in the decorative representation of various Egyptian deities and other Egyptian motifs. he variety of the jewellery and their well-known ind spot provided the opportunity to study the hoard and gather more precise data that will facilitate the investigation of similar artefacts in the future. A pilot study, reported in this volume by Guerra and Rehren, within the AUTHENTICO research project activities was carried out on some of the artefacts, showing the high caratage of the gold alloys used in the fabrication of the jewellery. Résumé : En 1998 une tombe a été fouillée dans la banlieue sud de la ville de Larnaca. La tombe, référencée MLA 1742, comprenait un ensemble unique d’orfèvrerie de parure du défunt. L’ensemble était constitué de vingt-trois pièces de typologies diverses, fabriquées en or et ornées de pierres semiprécieuses. La tombe présente plusieurs signes de rituels comparables à ceux des fameuses Tombes royales de Salamis. Nous pouvons suggérer qu’un membre de la haute société phénicienne de Kition y était enterré, voire même un membre de la famille royale de la cité. La tombe est datée de la in du VIIIe siècle av. J.-C. quand Chypre se trouvait sous une forte inluence orientalisante, attribuable selon toute probabilité à la présence de colons Phéniciens à Kition. Plusieurs techniques orfèvres, telles la granulation, le cloisonné, etc., ont été utilisées pour la production de ces pièces d’orfèvrerie. Certaines typologies de ces pièces sont très rares, voire uniques. Les anneaux et les bagues pivotantes trouvés dans la tombe présentent une forte inluence égyptienne, révélée par les représentations de diférentes divinités Egyptiennes et de motifs égyptiens. La variété de l’orfèvrerie et leur célèbre lieu de découverte ont fourni l’opportunité d’étudier ce dépôt et de rassembler des données exactes qui faciliteront les recherches dans l’avenir sur des objets similaires. Une étude pilote, dans le cadre des activités de recherche du projet AUTHENTICO et présentée dans ce volume par Guerra et Rehren, a été menée sur cette orfèvrerie permettant de montrer la haute qualité des alliages d’or utilisés pour fabriquer les pièces. Keywords: cloisonné, Cyprus, ibula, granulation, jewellery, Phoenicia, Kition. Mots-clés : cloisonné, Chypre, ibule, granulation, orfèvrerie, Phénicie, Kition. 1. INTRODUCTION Kition was one of the most important kingdoms of Cyprus, located in the south-eastern part of the island (Fig. 1) and hosting one of the most signiicant hoards in Cyprus dating from Antiquity. he ancient town has a Mycenaean pre-history; however, it also lourished since the 9th century BC, when an important tribe of traders and * Department of antiquities – 1, Museum Avenue, 1516 Nicosia, Cyprus. (plourentzos@da.mcw.gov.cy) ** EJTN GEIE – 124, Rue du Commerce, B-1000 Brussels, Belgium. (info@ejtn.org) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 143-149 144 Pavlos FLOURENTZOS, Maria Luisa VITOBELLO Figure 1: Map of Cyprus with the localisation of Larnaca. Figure 1 : La carte de Chypre avec la localisation de Larnaca. seafarers came to colonize the area. hose seafarers were of course Phoenicians, originally from the famous city of Tyre in the Levant (Geling and Viemer, 1990). Since the Ancient and Classical periods (8th to 5th century BC), a Phoenician dynasty ruled Kition. In 1998, an isolated built tomb – as it was not part of a wider necropolis – of monumental architecture was excavated in the south-western outskirts of modern Larnaca (Fig. 2). he tomb had a chamber and a dromos where sacriices of horses were exposed, with bronze components from the horses’ harnesses. Inside the tomb, due to great amount of water lowing in throughout the centuries, completely deteriorated bones were found (Hadjsawas, 2000; Karageorghis, 1976). Apart from the jewellery that is presented below, various vessels imported from Phoenicia were also found. An amphora showing a unique decoration, which was closed with a lid, deserves special attention. he jewellery consists of various types of pendants and beads (part of a necklace or necklaces), earrings, bracelets, rings and a unique gold ibula (Pierides, 1971). 2. EARRINGS Five earrings were found: all of them belong to the boat or leech shape typology, and are forged from a single piece of solid cast gold, hammered and shaped as a crescent. One end of the crescent extends to form the hanging hoop from which the hook is tapered, smoothed out into a ine needle for insertion into the ear. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 143-149 3. RINGS hree rings of solid gold were discovered in the tomb. he irst one (MLA 1742/1) is a gold ring with an ivory intaglio bezel. he hoop and the bezel were worked separately; the ring is made of a single piece of solid gold, cast in a mould, while the hoop’s surface is roughly inished by burnishing. he upper side of the hoop has an elliptical bezel, bearing a separate seal device. he sphragistic device, broken into three pieces, is engraved with the representation of an ibis standing on its side, while the background is illed with crossed lines representing bamboo or papyrus. he second one (MLA 1742/2) is a gold ring with an indigo (sodalite) intaglio gem. he artefact shows an identical structure and manufacturing techniques to the previous ring, also having an elliptical bezel with a removable seal device. he sphragistic device is engraved and depicts the Egyptian goddess of life Maat with a feather. he goddess Maat was the personiication of the basic laws of all existence; she embodied the concepts of law, truth and world order. In the Egyptian tradition, judges were regarded as priests of Maat. he third object (MLA 1742/3) is also a gold ring with faience intaglio. he manufacturing techniques are similar to those already mentioned above; the seal, however, is not removable, possibly to withstand daily use. Another interesting feature of this ring is that the intaglio, broken and partly worn, possibly depicts an ibis and a feather. he Phoenician gold jewellery from Kition, Cyprus 145 Figure 2: The localisation of the Phoenician tomb of Kition in modern Larnaca. Figure 2 : La localisation de la tombe phénicienne de Kition dans la ville moderne de Larnaca. 4. BRACELETS he tomb also yielded two unique bracelets. he irst one (MLA 1742/18) (Fig. 3) consists of two parts: a ribbon bracelet and a circular box setting, holding a discoid cabochon agate, having at the centre an eye reminiscent of the eye of Horus, having an apotropaic signiicance in ancient tradition. he stone is set into a box setting, the sides of which and top frame are granulated. he strap consists of a ribbon obtained from four parallel double loop-in-loop chains, threaded through by additional loops linking the chains and forming the braided strap up to the desired length. At both ends of the ribbon, the inials have their loops extended by approximately 25 mm to provide hinges through which pivots connect the ribbon to the box setting. he box setting is decorated with a row of granules and alternating granulated triangles; a row of granules runs along the full length of the side; a border of round wire encircles the entire side below, with an additional double braided wire above the round wire. A thin top plate is fused to a cylinder forming the outer wall of the box; starting from the outer edge, the centre frame is decorated with a circle of round wire and a ArcheoSciences, revue d’archéométrie, 33, 2009, p. 143-149 146 Pavlos FLOURENTZOS, Maria Luisa VITOBELLO Figure 3: (See colour plate) he bracelet from the tomb of Kition with circular box setting holding a discoid cabochon agate, reference MLA 1742/18. Figure 3 : (Voir planche couleur) Le bracelet de la tombe de Kition possédant une monture circulaire contenant une agate en cabochon, référence MLA 1742/18. row of granules alternating with granulated triangles pointing inward towards the centre of the box. A second row of granules, showing granulated triangles pointing outward from the centre and a circlet of round wire, delimitates the inner border of the frame. On both sides of the box, diagonally opposed, tiny cylinders obtained from coiled wires form the hinges through which the pivots are inserted to connect the ribbons’ hinges to the box setting. From a technical and an artistic point of view, this bracelet is extremely attractive and rare; if considering the manufacturing techniques, the combination and reinement of the applied competences and skills of the ancient craftsman that fabricated this artefact render it more than exceptional worldwide. he second bracelet (MLA 1742/19) (Fig. 4) is manufactured with an entirely diferent technique. It consists of a quadruple loop-in-loop chain, with a wire diameter of 0.4 mm and a chain length of 180 mm. At both ends of the chain, an inner tapering wire pivot is inserted through the scarab’s suspension-hole to allow swivelling; the wire is wound into and around the bracelet’s inal loops, coiled fourfold around itself to secure the scarab. he scarab is set inside an elliptical bezel, with its sides smoothly rounded of at both ends, two globules (recalling dung balls) holding the scarab. he globules are embossed, and joined to the bezel by fusing. Both globules show a round hole in the lower part, to allow the connecting tapered wire to pass through. he scarab is carved both on the soft round top and lat on the backside, and its surface is decorated with an inscription showing Egyptian hieroglyphics: on its lower part, two goddesses seated side by side and three ‘beathyls’ topped by a cartouche. he cartouche is typically evidence of objects often associated with the Pharaoh. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 143-149 Figure 4: (See colour plate) he bracelet from the tomb of Kition with a scarab and a quadruple loop-in-loop chain, reference MLA 1742/19. Figure 4 : (Voir planche couleur) Le bracelet de la tombe de Kition avec un scarabée et une chaîne en quadruple loop-in-loop, référence MLA 1742/18. 5. THE NECKLACE Another important ind is an extraordinary necklace consisting of seventeen beads and pendants of various shapes and signiicances. he gold elements of the necklace are obtained from high carat natural gold alloy, an aspect conirmed by further characterization (Guerra and Rehren, 2009). One of these is a gold sheet disc (MLA 1742/25) (Fig. 5) made of beaten gold sheet and framed by a round wire edge fused onto it. A large granule is positioned in the centre of the disc. Such a disc is a diagnostic symbol of the goddess Astarte, often represented ornate with a crescent, the goddess’ symbol. hree beads of this necklace (MLA 1742/34, 6, 7) are spherical, made of high carat gold alloy, and obtained by joining two hemispheres, each one embossed out of thick plate. he joining of the two halves is clearly outlined by magniication. he upper hemisphere is pierced at the top, where a small round suspension loop is inserted and fused. In addition, there are also two acorn-shaped beads made of high carat gold alloy (MLA 1742/30, 9), obtained by the same technique described above; the upper hemisphere is 147 he Phoenician gold jewellery from Kition, Cyprus Figure 5: (See colour plate) Mounting proposal for the necklace of Kition, containing seventeen beads and pendants of various shapes. Figure 5 : (Voir planche couleur) Proposition de montage du collier de Kition, consistant en dix-sept perles et pendentifs de formes variées. bordered by a row of granules from which three granulated triangles pointing downwards provide additional decoration. he necklace also consists of four cylindrical beads (MLA 1742/17, 28, 31, and 32). hree beads (numbers 28, 31, 32) are made from marble and agate, pierced longitudinally and itted at both ends with gold sheet inials. he other one, no. 17, is made from ivory, is plain and does not bear golden sheet inials. Another element of the necklace is a plain spherical agate pierced through (MLA 1742/33). An additional outstanding bead (MLA 1742/29) is executed in a stone – probably marble – representing a bull’s protome, delicately carved to enhance the eyes, nose, and ears: the forehead is decorated with the Astarte sun disc, represented by carved dots. he bezel which surrounds the bead is obtained from gold sheet; two decorative rows of round wire and a central row of braided wire are positioned and fused along the lower side of the bezel, encircling the seal; both bead and bezel are pierced through in the top part to allow threading, while the two opposed holes are bordered by two round loops. he height of the pendant allows for an engraved back side seal representing a goddess with uplifted arms. Two scarabs (MLA 1742/4, 5) represent further elements of the necklace. No. 4 consists of an engraved faience scarab, pierced longitudinally for suspension. On each side of the opening, two globules (dung balls), obtained by embossing thin gold plate, are fused to the bezel, holding the jewel. he high bezel is made of gold sheet; the scarab’s backside seal is engraved, showing centred on the surface an alabaster bottle, lanked by two C-shaped ornaments and, on the outer edge, on both sides, two ankhs symbols are engraved. he other scarab (no. 5) is quite unique in its conception; it is composed of two superimposed elements: a scarab, obtained from embossed gold sheet, with delicate features reproducing the animal’s details, is positioned over a back seal stone, in turn set into a golden bezel. he entire upper gold surface of the embossed scarab has a patina, the origin of which needs further investigation. he back side seal, positioned under the golden carapace, is also carved, showing probably a hieroglyphic symbol lanked by two feathers. Finally, the necklace has two other golden beads. he irst (MLA 1742/16) is made of gold alloy, possibly representing a vase; the body is topped by a cylindrical neck, its upper part decorated with one loop of round wire, one loop of braided wire (two twisted wires facing opposite direction), and another loop of round wire; the wires are fused onto the neck surface. he second bead (MLA 1742/8) has a lentoid shape, and is obtained from two hemispheres of embossed gold sheet; the bead is open on both sides to allow threading; both sides of the opening are bordered by rows of granules. Along the circumference, two rows of granules deine each hemisphere; along each row, six granulated triangles, alternating with six granulated lozenges, decorate both the upper and the lower hemisphere of the bead. 6. THE FIBULA he gold ibula (MLA 1742/20) (Fig. 6) is obtained from a single gold rod, skilfully forged to the desired triangular shape, which recalls an inverted bow. he bow is pinched at the top into a loop from which both sides extend and are elegantly shaped into inverted curves; on one side, the body lows into the coiled spring, and then tapers into the needle; on the other side, the body is forged shaped into the foot and catch plate. his typology of ibula is deined as an inverted bow-shaped ibula (Gjerstad, 1948). As decorative elements, three cloisonné rosettes are positioned on the top and on both sides of the ibula. he Egyptian inluence of these particular elements is worth mentioning: similar rosettes decorate the head-dress of an Egyptian lady from the court of Tuthmosis III, 18th Dynasty, 1450 BC, hebes (Heiniger and Heiniger, 1974). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 143-149 148 Figure 6: (See colour plate) he unique inverted bow-shaped gold ibula of Kition, reference MLA 1742/20. Figure 6 : (Voir planche couleur) L’unique ibule en or à l’arc inverse de Kition, référence MLA 1742/20. he top rosette is eleven-petalled, the right rosette (facing the coil) is ten-petalled, while the left rosette (facing the catch-plate) is nine-petalled. Each rosette is obtained from gold sheet, strips of which form the cloisons. he cloisons outline the petals, which are illed with polychrome enamels of various colours. Additional decorative elements of the artefact are double loop-in-loop chains linked to the coiled spring: from each chain, connected by means of suspension rings, three delicately chiselled bells, recalling stylized lotus buds, are hanging. A further extraordinary feature of this ibula is an alef (a Phoenician letter) engraved on the front side, above the rosette, probably representing the initial letter of the owner’s name. During the Late Cypro-Geometric and Early CyproArchaic period (circa 850-700 BC) this type of ibulae were mostly made of bronze; the shape is typically and diagnostically Cypriot, often found in tombs of the above mentioned period all over the island. Very rarely, this type of ibula is made of silver: the ibula from the Kition-Larnaca built tomb no. 1 thus represents a golden work of art of exceptional importance for Cypriot and world heritage. Preliminary to a inal publication which will reconstruct the excavation’s context and methodology, some introducArcheoSciences, revue d’archéométrie, 33, 2009, p. 143-149 Pavlos FLOURENTZOS, Maria Luisa VITOBELLO tory content is provided in this paper, aiming to position the artefact within its historical and archaeological context. he ibula’s main body, of irregular triangular shape, can easily be referred to Type 4b (Fig. 7) of the Cypriot ibulae, according to the criteria already presented by the Swedish Cyprus Expedition (Gjerstad, 1948: 145, Fig. 26, Bronze, 4a and 4b) and further conirmed by the typological analysis (Stronach, 1959); it represents an original Cypriot type. he chronological horizon appears to be clearly established around the 8th-7th century BC (Cypro-Geometric period), tending to a lower dating of such a chronology. his typology is common in the Syro-Palestinian area, and in the Anatolian and Aegean environments, especially Rhodos and Aegina. Characteristic features of the Kition ibula are the addition of three chains and lotus bud pendants, and, above all, the exceptional enrichment of three cloisonné rosettes to the main body, enhanced by polychrome enamel cloisons, which are placed two at the centre of each of the ibula’s arms, and one at the top of the ibula’s apex, respectively. With their curved shape, both of the ibula’s arms appear to recall the Egyptian solar boat: this appears evident in the bows’ shapes, which meet towards the apex with an elegant decorative efect, while the two rosettes, topping both arms, are placed in the area where the cabin is usually located. Such a tendency to use Egyptian or Egyptianizing elements or symbolism in such an eclectic and audacious way is a typical and founding aspect of Phoenician art, of which the Kition ibula is certainly one of the most original and uncommon examples (Ciafaloni, 1995). Not only in jewellery, but also in glyptic arts (i.e. on a chalcedony scarab at the Louvre Museum) and in ivories (Herrmann, 1986: nn. 989-994, Pls. 2255-2257), the solar boat symbolism is widely asserted, and represented in numerous and ingenuous variations with respect to the classical Egyptian interpretation. Phoenician sumptuous objects had a widespread difusion with all Mediterranean elites, in this particular instance in Figure 7: he elbow ibula of type 4 from Stronach (1959). Figure 7 : La ibule à l’arc coudé de type 4 de Stronach (1959). 149 he Phoenician gold jewellery from Kition, Cyprus its oriental area, in the chronological span already identiied, and in particular in Cyprus, as shown by the example of the so-called Cypro-Phoenician bowls (Markoe 1985), which constitute the most evident manifestation of such an eclectic trend. Within the Levantine and Cypriot environments, the Phoenician craft was permeated with Mesopotamian and Near-Eastern inluences; in fact, it can probably be inferred that it shares an ample aesthetic and technical Koiné only partially known. Notwithstanding this aspect, the extraordinary recent discovery of the Nimrud Queens Tombs (Damerji, 1999) has added an impressive quantity of artefacts to the meagre evidence related to Mesopotamian jewellery from the irst millennium BC. As an example, cloisonné and champlevé techniques are widely exempliied by the application of enamels and semiprecious stone inlays on sumptuous bracelets (Damerji, 1999: Abb. 27-30), known until now only from relief reproductions of neo-Assyrian palaces. It then appears that not only Egypt should be taken into consideration as a source of the decorative technique applied on the rosettes, even if, since the New Kingdom period, there are undoubtedly copious testimonials of loral elements characterized by these techniques in Egypt (see the head-dress of one of Tutmosis III’s wives from hebes, now at the Metropolitan Museum of Art, New York), as well as in a Mycenaean setting (Demakopoulou, 1988). To conclude, it could be stated that, in the Kition ibula, the corpus of Egyptianizing elements generally recalling the religious world steers towards an eschatological and symbolic interpretation of the artefact, possibly destined, since its creation, for a real use in everyday life, to be worn by a prince or a member of the Kition aristocracy, as well as for funerary use. Furthermore, the rosette motive in the Mesopotamian area appears linked to the major oriental female goddess Ishtar/ Astarte, i.e. in neo-Assyrian setting, chronologically parallel (8th century BC) to the Kition ibula: cloisonné rosettes appear on a bracelet from the above mentioned Nimrud Queens Tombs (Damerji, 1999: Abb. 30), where the same goddess is represented in the bracelet centre-plate. Taking into account the eclectic and syncretic orientation of the Phoenician culture already identiied above, it is possible to assume that both sources, the Egyptian and Mesopotamian, eiciently merge into a certainly unique artefact. Acknowledgements With thanks to Prof. Davide Ciafaloni, Università di Milano Bicocca, Italy, for his collaboration and contribution concerning the humanistic aspects of the research on the Kition ibula. References CIAFALONI, D., 1995. Iconographie et Iconologie, in V. Krings (ed.), La civilisation phénicienne et punique. Manuel de recherche. Leiden: Brill, 535-549. CULICAN, W., 1966. he irst merchant venturers: the ancient Levant in history and commerce. London, hames & Hudson. DAMERJI, M.S.B., 1999. Gräber assyrischer Königinnen aus Nimrud. Mainz, Verlag des Römisch-Germanischen Zentralmuseums. DEMAKOPOULOU, K. (ed.), 1988. he Mycenaean World: Five Centuries of Early Greek Culture, 1600-1100 BC. Athens: Ministry of Culture, National Hellenic Committee (ICOM). GELING, U. and VIEMER, H.G., 1990. Die Phoenizier in Zeitalter Homer. Mainz. GJERSTAD, E., 1948. he Cypro-Geometric, Cypro-Archaic and Cypro-Classical Periods. he Swedish Cyprus Expedition, Vol. IV, part II, Arts & Crafts. Stockholm, he Swedish Cyprus Expedition. GUERRA, M. F. and REHREN, Th. 2009. In-situ examination and analysis of the gold jewellery from the Phoenician tomb of Kition (Cyprus), ArcheoSciences 33. HADJISAVVAS, S., 1999. Chronique de fouilles et decouvertes archeologique à Chypre. In BCH (123). HEINIGER, E.A. and HEINIGER, J., 1974. he Great Book of Jewels. Boston, New York Graphic Society. HERRMANN, G., 1986. Ivories from room SW37 Fort Shalmanesher. London, British School of Archaeology in Iraq. KARAGEORGHIS, V., 1976. Kition Mycenaen and Phoenician discoveries in Cyprus. London, hames & Hudson. MARKOE, G., 1985. Phoenician Bronze and Silver Bowls from Cyprus and the Mediterranean. Berkeley, CA, University of California Press. MOSCATI, S., 1988. he Phoenicians, Milan. PIERIDES, A., 1971. Jewellery in the Cyprus Museum. Nicosia, Republic of Cyprus, Ministry of Communications and Works, Department of Antiquities. STRONACH, D., 1959. he Development of the Fibula in the Near East. Iraq 21: 180-206. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 143-149 In-situ examination and analysis of the gold jewellery from the Phoenician tomb of Kition (Cyprus) Examen et analyse in situ d’orfèvrerie en or de la tombe phénicienne de Kition (Chypre) Maria Filomena Guerra* and hilo Rehren** Abstract: he Phoenician tomb discovered in Kition in 1998 yielded some of the most impressive gold jewellery items recovered by archaeological excavations in Cyprus. Seal inger-rings, lunate earrings, necklace pendants, bracelets and a richly decorated elbow ibula exemplify the goldsmith’s skill in the 8th-7th century BC. he aim of this work is to describe and illustrate the goldsmith’s techniques with low magniication micrographs and to determine the variety in composition of the gold alloys used in the manufacture of 25 items. he study, integrated in the European project AUTHENTICO, was carried out at the National Museum of Archaeology of Nicosia with a portable video camera equipped with a zoom and a handheld XRF spectrometer. Résumé : Parmi les objets d’orfèvrerie les plus impressionnants livrés par les fouilles archéologiques menées à Chypre, se trouvent ceux découverts en 1998 dans la tombe phénicienne de Kition. Bagues à sceau, boucles d’oreille, pendants de colliers, bracelets et une ibule à l’arc coudé richement décorée, mettent en évidence la virtuosité de l’orfèvre au VIIIe-VIIe siècle av. J.-C. Le but de ce travail est de décrire et illustrer les techniques employées par l’orfèvre et de déterminer la variété des alliages en or employés. L’étude analytique a été réalisée dans le cadre du projet européen AUTHENTICO, au Musée national d’archéologie de Nicosie, à l’aide d’une camera vidéo portable équipée d’un zoom et d’un analyseur de luorescence X portable miniaturisé. Keywords: Jewellery, analysis, Kition, Phoenician, gold. Mots-clés : Orfèvrerie, analyse, Kition, phénicien, or. 1. INTRODUCTION he Phoenician tomb MΛA 1742, with two chambers and a step dromos excavated near Larnaca in 1998 (Hadjisawas, 1999), delivered Phoenician pottery and jewellery of high quality and exceptional iconography described in a publication by Flourentzos and Vitobello in this issue of ArcheoSciences. Similar costume jewellery was found in other Phoenician sites of the Mediterranean (Carpenter, 1958). he aim of this work is to describe the diferent steps of production of some of the 25 gold items found in the tomb and to determine the base alloys used in the manufacture of their diferent parts. Very few analytical studies address Cypriot gold jewellery, and the number of gold objects generally found in the archaeological excavations from before * Laboratoire du Centre de Recherche et de Recherche et de Restauration des Musées de France, UMR171 CNRS – 14, quai François-Mitterrand, 75001 Paris, France. (maria.guerra@culture.gouv.fr) ** UCL Institute of Archaeology – 31-34 Gordon Square, London WC1H 0PY, United Kingdom. (th.rehren@ucl.ac.uk) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 151-158 152 Maria Filomena GUERRA, hilo REHREN the Late Cypriot period is small, most likely due to the absence of gold sources on the island (Maxwell-Hyslop, 1971). he jewellery inds in Cyprus are often stylistically connected to Syria-Palestine – ropes of twisted wire, rosettes, granulated triangles, etc. – and Minoan-Mycenae – granulations (Lilyquist, 1993), pins, pennanular and twisted earrings, etc. – gold work. he rich Late Cypriot II inds of Enkomi (Myres, 1914 ; Murray et al., 1970 ; Åström, 1972) demonstrate the existence of relations with Tell Ajjul and Meggido in Palestine, with Ras Shamra in Syria, with Troy in Greece, etc. From the 8th to the 7th century BC, the Mesopotamian inluence on the various jewellery working centres of the Levant persists (Maxwell-Hyslop, 1960). he excavations of Amathus, Salamis and Kition in Cyprus show that rich Phoenician goods reached Cyprus during this period (Bikai, 1989). The jewellery items from the Kition tomb that are most representative of the relation of Cyprus with other Mediterranean civilisations are the solid lunate earrings (Fig. 1a) and the elbow triangular ibula (Fig. 1b); see also the contribution by Flourentzos and Vitobello in this issue. a he main focus of this paper is on the massive gold elbow triangular ibula of Kition, decorated with three rosettes in inlaid cloisonné and hanging down chains with lotus lower-shaped bells. he arrival of this type of ibula in the Mediterranean region is related to the expansion of Mycenaean trade, which brought new types of artefacts to Cyprus towards the end of the 13th century BC (Stronach, 1959): along these lines, we can refer to the chamber tomb with a dromos inds from Enkomi (Courtois, 1982), Kato Alonia/Kouklia (Karageorghis, 1963), Amathus (Karageorghis, 1988), Ayios Athanasios (Karageorghis, 1986) in Kourakou/Nicosia (Karageorghis, 1965), and Alassa-Kandou/Limassol (Karageorghis, 1988). Usually in bronze, triangular bow (or ‘knee’ or ‘elbow’) ibulae are very popular from the 8th century BC to the 1st century AD in Cyprus, Egypt and Persia, but also in France (Mohen et al., 1971), Portugal (Ponte, 1999), the Caucasian regions (Muscarella, 1965) and Spain (Chamorro, 1987). he ibula from Kition has inlaid cloisonné rosettes and lower bell-shaped pendants. he use of decorative rosettes dates back to the 3rd millennium BC in Mesopotamia and Egypt (Tait, 2006; Cline and O’Connor, 2006), and can b c + ArcheoSciences, revue d’archéométrie, 33, 2009, p. 151-158 153 In-situ examination and analysis of the gold jewellery from the Phoenician tomb of Kition (Cyprus) also be seen in a pin from Troy (Schliemann, 1880). hose in inlaid cloisonné with red and blue glass paste can be seen in the 15th century BC Egyptian Queen Iahhotep’s collar (Collective, 2004) and in the Mycenaean inger ring from Vapheio (Collective, 1988). he use of chains and bells is common in Phoenician jewellery, and examples of it are represented by the earrings from Camyrus (Salzmann, 1863), harros (Acquaro, 1984) and Liomilya/Kouklia (Brunet and Moretti, 1990). Solid lunate earrings in gold, bronze and silver, characterized by a solid crescentic body from one end of which the pin for suspension rises, were found as early as the 3rd millennium BC at Ur and took various forms during the Iron Age (Hawkes, 1961). In this context, we are reminded of those from Ajjul (10th century BC), Nimrud (7th century BC) and Tell Fara (11th-10th century BC) (Curtis and Maxwell-Hyslop, 1971). In Cyprus, leech-shaped earrings – an elongated version of the tapered hoop (Higgins, 1980), but with elongated loop – were found in Kouklia (Catling, 1968), Mesayitonia – Limassol (Karageorghis, 1963b), and in Amathus (Karageorghis, 1979). Other inds from the tomb include a necklace with stone inlay and a bracelet with a scarab inlay, both with delicate loop-in-loop wirework; several beads or pendants made entirely from gold, some with ine iligree work, and several stone or faience beads with gold inials; and several metal signet rings with cartouche-shaped bezels containing faience, glass or stone inlays. A full presentation of the inds is in preparation (Flourentzos forthcoming). 2. RESULTS For the study of the jewellery items from Kition, we used a high resolution Flexia video microscope with a zoom reaching x80, a built-in LED illumination and a Bersoft image measurement software. he analysis of the alloys was carried out with a handheld XRF spectrometer, Innov-X Systems Model Alpha 8000 LZX, with a silver tube, and operating at 40 kV, 29 μA, and employing a quantiication method optimised for alloys. he XRF analyses were carried out with an acquisition time of 25 sec, resulting in lifetimes of 19-22 sec. Precision and accuracy were tested by repeated analyses of a range of both certiied and internal reference materials with compositions comparable to the gold alloys studied (Table 1). Precision was found to be always better than 5% relative, and accuracy was typically better than 10% relative. he majority of the results reported are averages of 2-3 measurements, and they have been normalised to 100 wt%. In Au wt% Ag wt% Cu wt% 1814 coin given measured balance 4.8 4.3 90.8 4.3 4.9 A1 EJTN given measured balance 5.1 2.2 92.4 6.1 1.5 A2 EJTN given measured balance 20.5 4.2 75.1 21.2 3.8 E2 EJTN given measured balance 41.5 0.5 59.6 39.6 0.7 Table 1: Internal reference analyses by XRF. Tableau 1 : Résultats des analyses de standards d’or par FX. some instances, low concentrations of iron were found, but disregarded as potential surface contaminations. Earrings he ive solid gold lunate-shaped earrings with a bent over the hoop are cast close to shape. he hoop is hammered to form the crooked pin (Fig. 2), and the other end is inished by polishing. he compositions of the gold alloys used in the fabrication of these ive items have silver contents ranging from about 11% to 21%, and copper contents from about 1% to just over 2% (see results on Table 2). hus, despite their stylistic and technical similarity, and internal consistency, they are made from a rather wide range of gold alloys. Interestingly, the silver to copper ratio is about 10 for all analyses. he variety of the Phoenician gold alloys was shown for the Gadir region by Ortega-Feliu et al. (2007). Figure 2: Both ends of earring 1742/24, with the hammer facets on the surface typical of hammering. Figure 2 : Les deux terminaux de la boucle d’oreille 1742/24 avec les facettes typiques du martelage. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 151-158 154 Maria Filomena GUERRA, hilo REHREN region of analysis Au wt% Ag wt% Cu wt% Earrings 1742/21 body 83.8 14.5 1.7 1742/22 body 78.7 19.0 2.3 1742/23 body 77.3 20.7 2.0 1742/24 body 88.2 10.9 0.9 1742/25 body 77.9 20.0 2.1 2 bells on pendant 1 82.7 15.8 1.5 2 bells on pendant 2 81.7 16.0 2.3 2 bells on pendant 3 82.3 15.4 2.3 3 chains jointly 82.0 16.4 1.6 spiral 84.1 14.9 1.0 shield of pin 82.6 16.6 0.9 side of lower 83.3 15.7 1.0 1742/1475 hoop 97.7 2.1 0.2 1742/1476 hoop 97.9 1.9 0.2 1742/1477 hoop 99.9 0.0 0.2 Fibula 1742/20 Finger-rings he ibula he massive gold elbow triangular-shaped ibula is decorated with rosettes in cloisonné, inlaid with red and blue glass paste. he plain triangular pin is approximately 2 mm in diameter and shaped by casting and hammering – one soldering point on the top (Fig. 3a) reinforces the item. It is decorated with engraved lines produced with a triangular burin. he cloisonné rosettes consist of a base gold foil, cut and polished on the borders, where strips around 400 μm thick and 1000 to 1200 μm in length, cut with a chisel, are hard-soldered. We note that this technique was found previously in Tartesic (Ontalba-Salamanca et al., 2001) and Phoenician (Ortega-Feliu et al., 2007) jewellery. he internal surfaces of the petals are scratched (Fig. 3b), potentially to improve the adherence of the glass paste, either directly, when applied as a hot semi-molten mass, or to provide more grip for any adhesive that might have been used if the inlays were applied cold. On the back-side of the gold foil, one wire covers the joining with the pin. he loop-in-loop chains are obtained with strip-twisted wires of 450-500 μm in diameter (Fig. 3c), and the bells are made of rolling gold foils in repoussé adorned with wires of about 600 μm in diameter (Fig. 3d). he suspension rings are of about 700μm in diameter, and the round ring of which diferent pendants are suspended has a diameter of about 950-1000 μm. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 151-158 Table 2: Results obtained by XRF for the lunate earrings, the elbow triangular ibula, and the signet inger-rings. Tableau 2 : Résultats obtenus par FX pour les boucles d’oreille en lune, pour la ibule triangulaire coudée et pour les bagues à sceau. All the elements of the ibula were produced with a relatively similar base alloy, containing 82 to 83 wt% gold, 15 to 16 wt% silver, and 1 to just over 2 wt% copper. Within this already relatively narrow range two discrete subgroups emerge, with their main diference residing in the copper content. he chains and bells forming the pendants are made from an alloy containing a slightly higher copper content (1.5 to 2.3 wt% Cu), while the body of the ibula has consistently only 1 wt% copper. In contrast to the earrings, where the copper and silver contents were directly correlated, no such correlation is apparent here. he silver content is almost the same for all parts of the ibula, while the copper content difers by a factor of two between the pendant parts and the main body of the ibula. Whether the diferent concentrations of copper were used in order to change the hardness, or simply relect the assemblage of parts obtained from diferent sources or even workshops, is impossible to decide on the basis of these analyses alone. he remaining gold artefacts were also studied and a full publication is forthcoming. Here, it may suice to say that the workmanship is consistent with the details described above for the ibula and earrings, and that the majority of the other items are made from gold of similar or even baser composition than the one used for the ibula pendant parts, that is to say of gold with 15 to 25 wt% silver and 2 to 3 wt% copper. he bracelet with scarab and the necklace with stone inlay are made from higher-quality gold, with 15 In-situ examination and analysis of the gold jewellery from the Phoenician tomb of Kition (Cyprus) 155 a b c d Figure 3: (See colour plate) he ibula construction: (a) the two branches are hard-soldered; (b) cloisonné rosette – the internal surfaces of the petals are scratched; (c) lotus lower bells hoop; (d) the loop-in-loop chains. Figure 3 : (Voir planche couleur) La construction de la ibule : (a) les deux branches soudées par brasure ; (b) la rosette en cloisonné – les surfaces internes sont grattées ; (c) les anneaux des cloches en forme de leur de lotus ; (d) les chaînes de suspension. to 18 wt% silver and 1 to 2 wt% copper. Remarkably, the three signet inger-rings with cartouche-shaped bezels (Fig. 4 shows that the internal surface of the bezel is also scratched) form a compositionally very diferent group, being made of high-carat gold. heir silver content is consistently below 2 wt%, and copper below 0.5 wt% in almost all analyses (Table 2). he microscopic study of the objects did not reveal any hard grey inclusions in the gold, and the XRF analyses did not indicate the presence of platinum group elements in any of the analysed areas. Tin was found only in one of the XRF spectra, in one of the inger rings, at a level too low to reliably quantify. 3. DISCUSSION he technical study of the inds shows the use of casting and hammering as well as of hard-soldering for the production of the jewellery from the Kition tomb. All these techniques could be identiied on Phoenician, Punic and Tartesic jewellery. Similarly, the techniques of decoration by application of lower bells, glass paste in cloisonné, loopin-loop chains, ropes made with twisted wires, etc. are also expected for this type of jewellery. Traces of use-wear could be seen on the surface of all objects. his indicates that the objects were used for some period of time, and not only made for funerary purposes. Phoenician jewellery is frequently accepted as the precursor of Etruscan gold work; however, the diameters of the wires, ArcheoSciences, revue d’archéométrie, 33, 2009, p. 151-158 156 the thickness of the gold foils, the wire and iligree patterns, and so forth seen in our objects are far from the delicate craftsmanship of Etruscan gold. Chemically, there appear to be two diferent compositional groups represented among the gold objects; both are consistent with unalloyed native gold. With the exception of the three inger rings, the silver content varies from 15 to 25 wt%, while the copper content is mostly between 1 and 2.5 wt%. he silver content is what one would expect from mined natural gold, neither reined nor alloyed. Hough et al. (2009) report typical silver values for primary or hypogene gold of 5 to 20 wt%, only occasionally exceeding 50 wt%. Analytical data for geological gold suggest that most individually analysed gold particles have between 10 and c. 40 wt% silver, but almost always less than 1% copper (e.g. Chapman and Mortensen, 2006). However, most gold deposits also have small amounts of copper and silver minerals associated, and during the concentration of the gold from the mined rock ore, some of these heavy minerals will inevitably be retained in the resulting concentrate (see e.g. Hauptmann and Klein, 2009) in this volume of ArcheoSciences). hus, during the smelting of the gold concentrate, this copper will be partly alloyed into the bullion, resulting in the slightly elevated levels reported here. In modern gold mining, using advanced mineral concentration methods and aiming for complete gold recovery, this results in much higher silver and copper concentrations in the raw bullion than are found in the individual gold grains from the same deposit. Gold obtained from secondary placer deposits, in contrast, difers systematically in its composition from the gold in the primary deposits. During transport, the silver content of the primary gold is more and more reduced due to leaching (e.g. Krupp and Weiser, 1992; Knight et al., 1999; Hough et al., 2009), while the primary accessory minerals such as chalcopyrite and various silver minerals are quickly dissolved and hence efectively separated from the placer gold. Depending on the make-up of the geological hinterland of the river system from which the placer gold is panned, it can be enriched in platinum group elements (Meeks and Tite, 1980) even though gold and platinum rarely occur together in the same primary deposits. his is noteworthy for example in the gold from Sardis in Turkey (Meeks et al., 1996). hus, placer gold will have less copper and lower silver on average than freshly mined gold, but potentially more PGE inclusions. It is of course possible that the copper was added intentionally; the efects of even small amounts of copper on hardness are relatively high, due to the c 12% diference in atomic radii between gold and copper (Hough et al., 2009), while the efect on the colour is probably too low to have ArcheoSciences, revue d’archéométrie, 33, 2009, p. 151-158 Maria Filomena GUERRA, hilo REHREN Figure 4: he bezel of gold signet inger-ring 1742/1476: the internal surfaces of the bezels are scratched. Figure 4 : Le chaton de la bague à sceau 1742/1475 : les surfaces internes du chaton ont été grattées. been manipulated at such a ine level. Equally, it is possible that a batch of gold of a certain composition, and hence property, was chosen for a speciic task, that is, a coppercontaining gold might have been preferred for objects that could exploit the improved hardness, such as thin sheet or wire making. he three inger rings with their exceptionally high gold content of 98 % or more form a very distinct group among the material excavated from the tomb; such a highpurity gold could indicate that it was reined and parted. However, current knowledge suggests that gold parting, that is the intentional removal of silver from native gold, was only introduced sometime in the mid-1st millennium BC, in the context of the development of gold coinage in Sardis (Ramage and Craddock, 2000). herefore, it is more likely that these rings were made from a particular gold source rather than from fully reined gold. Such pure gold most likely comes from a secondary or supergene deposit. Supergene gold forms under speciic environmental conditions, such as the warm and humid paleoclimate that resulted in the lateritic soils of SE Europe and Turkey to form (Herrington et al., 2007). During lateritic weathering, primary argentiferous gold is temporarily brought into solution and then re-deposited as almost pure gold grains and nuggets, typically containing 2 wt% silver or less, and nearly no copper (Freyssinet et al., 1989, 2005; Hough et al., 2009). On balance, we assume that the inger rings were made from supergene gold rather than from gold that had been subjected to parting. In-situ examination and analysis of the gold jewellery from the Phoenician tomb of Kition (Cyprus) 4. CONCLUSION he investigation of the Kition jewellery has given detailed insight into the manufacturing skills of the Phoenician jewellers producing these outstanding objects, as well as showing their competent selection of natural gold alloys for diferent applications. he level of craftsmanship is consistent with the best of the known jewellery from the period, and the assumption that these objects were made by Phoenicians with some Egyptianizing inluence on the iconography and design. While we can not exclude the possibility that the low copper concentrations in the gold containing 15 to 20 wt% silver are due to the alloying of a purer gold with a copper-silver alloy, we believe that the compositional characteristics of mined gold and supergene gold, respectively, are suicient for an interpretation based on selection of natural gold varieties, rather than active manipulation of the gold composition. References ACQUARO, E., 1984. Sardegna Archeologica, Studi e Monumenti 2, arte e cultura punica in sardegna, Carlo Delino editore. ÅSTRÖM, L., 1972. he Swedish Cyprus Expedition Vol. IV/1D. he Late Cypriote Bronze Age. Other arts and crafts Lund. BIKAI, P.M., 1989. Cyprus and the Phoenicians. he Biblical Archaeologist 52(4): 203-209. BRUNET, M. and MORETTI, J.-C., 1990. Chronique des fouilles et découvertes archéologiques à Chypre en 1989. Bulletin de correspondance hellénique 114(2): 941-985. CATLING, H.W., 1968. Kouklia : Evreti Tomb 8. Bulletin de correspondance hellénique 92/1: 162-169. CARPENTER, R., 1958. Phoenicians in the West, American J. Archaeology 62/1: 35-53. CHAMORRO, J.G., 1987. Survey of Archaeological Research on Tartessos. American J. Archaeology 91/2: 197-232. CHAPMAN, R.J. and MORTENSEN, J.K., 2006. Application of microchemical characterization of placer gold grains to exploration for epithermal gold mineralization in regions of poor exposure. Journal of Geochemical Exploration 91: 1-26. COLLECTIVE, 2004. Pharaon, Paris, Institut du Monde Arabe/ Flammarion. COLLECTIVE, 1988. Das Mykenische Hellas, Kulturministerium Griechenlands/ICOM – Freie Universität Berlin, Athens. COURTOIS, J.-C., 1982. L’activité métallurgique et les bronzes d’Enkomi au Bronze Récent, in J.D. Muhly and V. Karageorghis eds. Early metallurgy in Cyprus, 4000-500 BC, Nicosia, 155-175. CURTIS, J.E. and MAXWELL-HYSLOP, K.R., 1971. he gold jewellery from Nimrud, Iraq 33/2: 101-112. 157 CLINE, E.H. and O’CONNOR, D.B., 2006. hutmose III: a new biography, he University of Michigan Press. FREYSSINET, Ph., ZEEGERS, H. and TARDY, Y., 1989. Morphology and geochemistry of gold grains in lateritic proiles of southern Mali. Journal of Geochemical Exploration 32: 17-31. FREYSSINET, Ph., BUTT, C.R.M., MORRIS, R.C. and PIANTONE, P., 2005. Ore-forming processes related to lateritic weathering, Economic Geology 100th Anniversary Volume, 681-722. HADJISAWAS SOPHOKLIS, 1999. Chronique des fouilles et découvertes archéologiques à Chypre en 1998, Bulletin de correspondance hellénique 123(2): 599-633. HAUPTMANN, A. and KLEIN, S., 2009. Bronze Age gold in Southern Georgia. ArcheoSciences, 33. HAWKES, C., 1961. Gold ear-rings of the Bronze Age East and West, Folklore 72/3: 438-474. HERRINGTON, R., BONI, M., SKARPELIS, N. and LARGE, D., 2007. Palaeoclimate, weathering and ore deposits – a European perspective. Proceedings of the Ninth Biennial SGA Meeting, Dublin, 1373-1376. HIGGINS, R., 1980. Greek and Roman Jewellery, 2nd ed., London. HOUGH, R., BUTT, C. and FISCHER-BÜHNER, J., 2009. he crystallography, metallography and composition of gold. Elements 5: 297-302. KARAGEORGHIS, V., 1963. Chronique des fouilles et découvertes archéologiques à Chypre en 1963, Bulletin de correspondance hellénique 88(1): 289-379. KARAGEORGHIS, V., 1963b. Une tombe de guerrier à Palaepaphos, Bulletin de correspondance hellénique 87(1): 265-300. KARAGEORGHIS, V. 1965. Chronique des fouilles et découvertes archéologiques à Chypre en 1964, Bulletin de correspondance hellénique 89(1): 231-300. KARAGEORGHIS, V., 1979. Chronique des fouilles et découvertes archéologiques à Chypre en 1978, Bulletin de correspondance hellénique 103(2): 671-724. KARAGEORGHIS, V., 1986. Chronique des fouilles et découvertes archéologiques à Chypre en 1985, Bulletin de correspondance hellénique 110(2): 823-880. KARAGEORGHIS, V., 1988. Chronique des fouilles et découvertes archéologiques à Chypre en 1987, Bulletin de correspondance hellénique 112(2): 793-855. KNIGHT, J.B., MORTENSEN, J.K. and MORISON, S.R., 1999. he relationship between placer gold particle shape, rimming and distance of luvial transport as exempliied by gold from the Klondike, Yukon Territory, Canada. Economic Geology 94: 635-648. KRUPP, R.E. and WEISER, T., 1992. On the stability of gold-silver alloys in the weathering environment. Mineralium Deposita 27: 268-275. LILYQUIST, C., 1993. Granulation and glass: chronological and stylistic investigations at selected sites, ca. 2500-1400 B. C. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 151-158 158 E., Bulletin of the American Schools of Oriental Research 90-91: 29-94. MAXWELL-HYSLOP, K.R., 1960. he Ur jewellery, Iraq 22: 105115. MAXWELL-HYSLOP, K.R., 1971. Western Asiatic Jewellery c 3000612 B.C., London, Methuen & Co. MEEKS, N. and TITE, M.S., 1980. he analysis of platinum-group element inclusions in gold antiquities. Journal of Archaeological Science 7: 267-275. MEEKS, N., CRADDOCK, P.T., GECKINLI, A., HOOK, D., MIDDLETON, A. and RAMAGE, A., 1996. he scientiic study of the refractory remains and gold particles from the Lydin gold reinery at Sardis. Archaeometry ’94, ed. S. Demirci, A.M. Özer and G.D. Summers, Ankara, 461-482. MOHEN, J.-P., NICOLARDOT, J.-P. and CUNISSET-CARNOT, P., 1971. Une ibule « chypriote » trouvée en Côte-d’Or, , Bulletin de la Société préhistorique française, 68(2): 602-609. MURRAY, A.S., SMITH A.H. and WALTERS, H.B., 1900.Excavations in Cyprus, Bequest of Miss E.T. Turner to the BritishMuseum, [London, 1900; reprinted 1970], p. 1-54, pls. I-XII. MUSCARELLA, O.W., 1965. A ibula from Hasanlu, American J Archaeology 69/3: 233-240. MYRES, J.L., 1914. Handbook of the Cesnola Collection, New York. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 151-158 Maria Filomena GUERRA, hilo REHREN ONTALBA-SALAMANCA, M.A., AGER, F.J., YNSA, M.D., GÓMEZTUBÍO, B.M., RESPALDIZA, M.A., GARCÍA LÓPEZ, J., FERNÁNDEZGÓMEZ, F., DE LA BANDERA, M.L. and GRIME, G.W., 2001. External microbeam set-up at the CNA (Sevilla) and its application to the study of Tartesic jewellery, Nucl. Instrum. Meth. B 181: 664-669 ORTEGA-FELIU, I., GOMEZ-TUBIO, B., ONTALBA SALAMANCA, M.A., RESPALDIZA, M.A., DE LA BANDERA, M.L. and OVEJERO ZAPPINO, G., 2007. Gold and electrum jewellery in the strategic area of Gadir in Phoenician period, Nucl. Instrum. Meth. B 260: 329-335. PONTE, M.S., 1999. As ibulas do Bronze Final do Norte e Centro de Portugal, Revista de Guimarães II: 539-560. RAMAGE, A. and CRADDOCK, P., 2000. King Croesus’ Gold. Excavations at Sardis and the History of Gold Reining, London, he British Museum Press. SALZMANN, M., 1863. Bijoux phéniciens trouvés dans la nécropole de Camiros, Revue archéologie VIII, 1-6. SCHLIEMANN, H., 1880. Ilios. he city and country of the Trojan, John Murray, London. STRONACH, D., 1959. he development of the ibula in the Near East, Iraq 21/2: 180-206. TAIT, H., 2006. 7000 years of jewellery, London, British Museum. New finds from an old treasure: the archaeometric study of new gold objects from the Phoenician sanctuary of El Carambolo (Camas, Seville, Spain) 1 Nouvelles trouvailles pour un ancien trésor : l’étude archéométrique de nouveaux objets en or du sanctuaire phénicien de El Carambolo (Camas, Séville, Espagne) Alicia Perea* and Mark A. Hunt-Ortiz** Abstract: In this paper, we present the results of an archaeometric study by SEM-EDS, carried out on six gold objects originating from the historical site where, 50 years ago, an important gold treasure, belonging to the Orientalizing-Tartesian period, was found. The new finds came to light recently, during rescue excavations that unearthed a monumental complex described as a Phoenician sanctuary. Résumé : Nous présentons les résultats d’une étude archéométrique par MEB-EDS effectuée pour six objets en or provenant du site historique où, 50 ans auparavant, un trésor important, appartenant à la période orientalisante-tartésique, a été trouvé. Les nouvelles découvertes ont été mises à jour récemment, pendant des fouilles de sauvetage qui ont exhumé un complexe monumental décrit comme un sanctuaire phénicien. Keywords: Archaeometry, archaeometallurgy, goldwork, SEM-EDS, Orientalizing period, Southern Spain, Tartesian archaeology. Mots-clés : Archéométrie, archéométallurgie, Travail de l’or, MEB-EDS, période orientalisante, Espagne méridionale, Tartessique, archéologie. 1. INTRODUCTION On 30 September 1958, a treasure of gold objects was found by workmen refurbishing a public building at the top of a hill overlooking the fertile Guadalquivir river valley. The Carambolo gold hoard consists of two sets of objects, representing a total of 21 ornaments, including 16 rectangular plaques, two pendants, a pair of bracelets, and a pendant necklace. They have recently been dated to a period around the 8th century BC, except for the necklace, which is prob- * Arqueometal Research Group. CCHS, CSIC, Albasanz 26-28, 28037-Madrid, Spain. (alicia.perea@cchs.csic.es) ** ATLAS Research Group. Prehistory and Archaeology Department. Universidad de Sevilla. María de Padilla S/N, 41004-Sevilla. Spain. (mhunt@us.es) 1. This paper is part of the Research Project entitled: “Grounds for an archaeometric and technomic research on metallurgy during Prehistory and Antiquity. The Iberian Peninsula” (Ref. nº HUM2006-06250), within the Programme CONSOLIDER INGENIO 2010 (CSD-TCP), funded by the Ministry of Science and Innovation, Spain. rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 159-163 Alicia PEREA, Mark A. HUNT-ORTIZ 160 Figure 1: Loop-in loop chain. Figure 1 : Chaîne en loop-in-loop. ably Cypriot in origin and of a later date, around the 6th century BC, the period when the treasure was concealed. From a technological point of view, it has been interpreted as the joint production of two different workshops, one indigenous and the other of Mediterranean origin, suggesting a peaceful coexistence between local people and the Phoenicians, although no evidence for technological transmission has been detected (Perea and Armbruster, 1998). Rescue excavations at the time brought to light some structures that were interpreted as a ‘pit dwelling’. The pottery was said to be found in mixed archaeological levels containing both local and Phoenician types (Carriazo, 1973). For many years, El Carambolo was the only site to produce a stratigraphy for the period when natives and Phoenician colonists first came into contact. Years later, a bronze statue of Astarte, the Phoenician goddess, was found nearby. This and other finds alerted archaeologists to the possibility that the site may have been more than a simple settlement, and that it might have had a religious function. Rescue excavations carried out at this site from 2002 to 2004, due again to building works in the expanding city of Seville, unearthed a monumental sanctuary, dated to a period between the 8th and the 6th centuries BC. The complete stratigraphic sequence of the site extends from the final Bronze Age to the end of the Orientalizing period in the 6th century BC, when the site was abandoned (Fernández and Rodriguez 2005; 2007).The discovery of new gold objects during the most recent excavation has reopened the old debate on native versus Phoenician workshops. ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? Figure 2: Cylindrical lining. Figure 2 : Revêtement cylindrique. 2. THE NEW FINDS AND THE METHOD OF RESEARCH The new find consists of six gold objects: – a 42 cm long fragment of a double loop-in-loop chain. It displays the marks of an ancient repair and several broken links. – a 1.3 cm long cylindrical bead or lining made from grooved sheet, misshapen and incomplete. – four identical discoidal appliqués, 0.2 cm in diameter, each with a central hemispherical boss and a wire looped attachment on the back. These objects were sent to the Laboratorio de Microscopía Eletrónica CENIM, CSIC, in Madrid, for topographic examination using a Scanning Electron Microscope (SEM), and were initially analysed by Energy Dispersive X-ray Spectroscopy (EDS) in order to determine the elemental composition of the alloys, and also the type of solder used, if any. 3. TOPOGRAPHIC STUDY The cross-section of thewire employed for making the chain links is circular, with traces of a helicoidal line that has largely disappeared due to wear. These features indicate, firstly, a prolonged use, and also the fact that the wire was manufactured by twisting a thin strip of gold (also known as strip-twisting). It also shows the marks of an ancient repair. At the midpoint of the chain, an interconnected link appears which could represent the beginning of a second branch of the chain, or a link for suspending a pendant. New finds from an old treasure: the archaeometric study of new gold objects… 161 Figure 3: Discoidal apliqué (left) with the wire loop at the back (right). Figure 3a, b : Applique discoïde (gauche) avec la boucle sur le revers (droite). The cylindrical piece seems to be the lining of a cord or thick wire, rather than a separate bead. It is made from grooved foil rolled into a tube, which is considerably worn at the edges. The four discoidal appliqués are real miniatures made by stamping small circular pieces of sheet, and had a wire loop soldered onto the back so that they could be sewn onto cloth. The wire loop was made by the same system of twisting as the one used for the chain. They are so small that the soldering process which had to avoid fusing the tiny objects was difficult. Traces of these problems can be seen in the dendritic structures throughout the surface. 4. MICROANALYTICAL STUDY Table 1 records the complete results of the analysed areas. The chain is made from an alloy of 12% silver and 1% copper. The cylindrical lining has a silver content of 16% and about 1% copper. We cannot provide an approximate composition for the four appliqués, because all the measurements are affected to a greater or lesser extent by the process of soldering the loop to the back. However, they seem to be of fairly pure gold, with a silver content that cannot have exceeded 8%. In the case of the solder, we think it was carried out using a ternary alloy as a soldering material, with a silver content of around 20-25% and about 2% copper. There is no indication that an autogenous welding method or a bonding using copper salts was used, since the curve of the silver and copper measurements increases notably at the interfaces. 5. THE IBERIAN DISCOIDAL GOLD APPLIQUÉS The objects that provide most information for interpreting this find are the discoidal appliqués. We are aware of a small but significant number of these objects, which always appear in relatively numerous sets. All the finds are distributed in the southwest of the Iberian Peninsula, and their chronological contexts point to the final Bronze Age / Early Iron Age transition period, which encompasses the so-called Orientalizing period. The hill fort (castro) of Ratinhos (Moura, Portugal) presents signs of continuous occupation from the Final Bronze Age to the 6th century BC (Silva and Berrocal-Rangel, 2005). In the higher part of the castro, a large building, described as a palace, was located, where a set of seven discoidal appliqués with a diameter of about 0.8 cm was found. The edges of the appliqués were decorated with a square cross-section twisted wire. EDXRF and PIXE analyses (Valerio et al., 2007) showed that the fabrication material was a gold alloy with 10% silver. The authors think that solid phase welding was used to attach the loop to the back by heating below the melting point without using a solder. A second set of two appliqués, with a diameter of approximately 1.6 cm, is currently on display in the Lisbon Museum. It originates from Outeiro da Cabeça (Torres Vedras, Lisbon, Portugal) and was associated with earrings, beads and double spiral pendants, but it has no contextual data (Pingel, 1992: 295-296, nº 277). A third set consists of 39 items from Fortios (Portalegre, Portugal), again lacking an archaeological context, and is also held in the Lisbon Museum (Pingel, 1992: 298, ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? Alicia PEREA, Mark A. HUNT-ORTIZ 162 Item & analysis Nº Area Au Ag Cu Others Item & analysis Nº Area Solder loop/sheet 73.10 Au Ag 24.96 1.94 Cu 25.25 2.05 CRB01 201 Chain Link 88.91 11.09 - CRB04207 CRB01202 Link 86.06 13.94 - CRB04208 Solder loop/sheet 72.70 CRB01203 Link 87.44 12.56 - CRB04209 Solder loop/sheet 74.40 CRB01204 Repair 75.67 22.55 0.66 Fe, Ca CRB05 201 Discoid Sheet back 80.39 19.27 0.34 CRB01205 Repair 77.97 19.09 0.70 Fe, Ca CRB05202 Sheet back 92.50 7.03 Fe, Ca CRB05203 Sheet back 79.86 19.23 0.91 23.23 2.37 0.48 CRB01206 Repair 72.90 24.42 0.67 CRB02 201 Lining Sheet 83.14 16.17 0.70 CRB05204 Loop 83.08 16.78 0.13 CRB02202 Sheet 82.45 16.70 0.85 CRB05205 Loop 87.84 11.10 1.05 CRB02203 Sheet 65.44 17.09 0.71 CRB05206 Loop 91.50 8.14 Solder loop/sheet 72.88 26.01 1.11 26.24 1.51 Si,Al, K 0.36 92.26 7.25 0.48 CRB05207 CRB03 201 Discoid Sheet 100 - - CRB05208 Solder loop/sheet 72.25 CRB03202 Sheet 90.34 8.81 0.85 CRB05209 Solder loop/sheet 72.64 25.45 1.65 CRB03203 Sheet 99.19 - 0.81 CRB05210 Sheet 100 - CRB03204 Sheet back 90.88 9.12 - CRB05211 Sheet 88.46 11.03 0.52 0.55 CRB05212 Sheet 89.44 10.56 - CRB02204 Used edge 89.69 9.76 Fe - CRB03205 Sheet back CRB03206 Sheet back 88.91 10.68 0.41 CRB06 201 Discoid Sheet back 75.46 24.16 0.38 CRB03207 Loop 87.59 11.38 1.03 CRB06202 Sheet back 81.93 18.07 - CRB03208 Loop 89.69 10.31 - CRB06203 Sheet back 80.37 19.48 0.15 CRB03209 Loop 80.61 18.41 0.98 CRB06204 Loop 82.67 17.33 - CRB03210 Solder loop/sheet 75.14 22.63 2.23 CRB06205 Loop 77.55 22.19 0.27 CRB03211 Solder loop/sheet 73.52 24.27 2.20 CRB06206 Loop 77.17 22.83 - CRB03212 Solder loop/sheet 76.10 22.10 1.79 CRB06207 Solder loop/sheet 73.07 25.82 1.11 CRB06208 Solder loop/sheet 73.57 25.91 0.51 Fe Others CRB04 201 Discoid Sheet back 87.41 12.35 - CRB04202 Sheet back 71.26 26.99 1.76 CRB06209 Solder loop/sheet 74.38 25.18 0.44 CRB04203 Sheet back 74.65 23.27 2.08 CRB06210 Sheet 82.81 16.74 0.45 CRB04204 Loop 84.60 14.97 0.43 CRB06211 Sheet 77.10 22.90 - CRB04205 Loop 81.60 17.01 1.39 CRB06212 Sheet 78.22 21.19 0.59 CRB04206 Loop 90.38 9.62 - Table 1: Elemental spot composition (SEM-EDS, wt%). Tableau 1 : Composition élémentaire ponctuelle (MEB-EDS en %). nº 287). The objects have a diameter of about 2 cm, wire loops on the back, and are decorated with concentric circles. This is the largest set of such artefacts, and it also contains the largest examples known to date. Other comparable objects have been found, rather different morphologically from those under discussion here, but which could have been used in a similar manner to decorate garments, although they do not have the loop on the back, the defining feature of the sets described earlier. These are the appliqués from Sâo Martinho (Alcácer do Sal, Setúbal, Portugal) and El Castañuelo (Huelva, Spain), which have holes on the outer edge, probably for sewing or fixing. The association of the find of Sâo Martinho (Armbruster and Parreira, 1993: 188-191) with spirals, and the context of the items from El Castañuelo – a cist necropolis (Schubart, 1975: 95-96, lám. 54; Perea, 1991: 107) – seem to indicate ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? an earlier date than that of the items with a loop on the back instead of holes (Perea, 2005). They could be dating to the Final Bronze Age of the southwest. As a hypothesis, we propose that these ornaments originated from garments belonging to the local elites of the Final Bronze Age. The subsequent adoption of the new Mediterranean technology brought by the Phoenician colonists included soldering. This enabled the wire loops to be attached to the back, so that they could be sewn onto cloth – something which had been previously accomplished by using holes. Soldering also enabled the appliqués to be adorned with twisted wires around the edges. New finds from an old treasure: the archaeometric study of new gold objects… 6. CONCLUSION The great treasure of El Carambolo and this new find must be compared. The first one was hidden intentionally so that the jewels could be recovered. The objects were in a perfect state of preservation and showed few signs of wear. The new find, however, consists of fragmentary, incomplete objects that display traces of having been used for a long time, an aspect which seems to reflect an unintentional loss. As we can observe from the finds of discoidal appliqués that appear from the Final Bronze Age onwards, the adornment of garments with gold must have been relatively common in the southwest of the Iberian Peninsula. This represents new evidence for the same process, i.e. the coexistence of native and Phoenician workshops, or the presence of goldsmiths with different technical knowledge. This is an interesting case of technological transmission: the Atlantic versus the Mediterranean. Acknowledgements We are very grateful to Álvaro Fernández Flores, director of the excavations at El Carambolo, for entrusting the study of the gold finds to us. References ARMBRUSTER, B. and PARREIRA, R. (eds.), 1993. Inventario do Museu Nacional de Arqueologia. Colecçâo de Ourivesaria. 1º volume, Do Calcolítico à Idade do Bronze. Lisboa, Secretaria de Estado da Cultura. CARRIAZO, J.M., 1973. Tartessos y El Carambolo. Investigaciones arqueológicas sobre la protohistoria de la Baja Andalucía. 163 Madrid, Dirreción General de Bellas Artes, Ministerio de Educación y Ciencia. FERNÁNDEZ, A. and RODRIGUEZ, A., 2005. El complejo monumetal del Carambolo Alto, Camas (Sevilla). Un santuario orientalizante en la paleodesembocadura del Guadalquivir. Trabajos de Prehistoria 62(1): 111-138. FERNÁNDEZ, A. and RODRIGUEZ, A., 2007. Tartessos Desvelado. La Colonización Fenicia del Suroeste Peninsular y el Origen y Ocaso de Tartessos. Córdoba, Editorial Almuzara. PEREA, A., 1991. Orfebrería Prerromana. Arqueología del Oro. Caja de Madrid: Comunidad de Madrid, Consejería de Cultura, Dirreción General de Patrimonio Cultural. PEREA, A., 2005. Mecanismos identitarios y de construcción de poder en la transición Bronce-Hierro. Trabajos de prehistoria 62(2): 91-103. PEREA, A. and ARMBRUSTER, B., 1998. Cambio tecnológico y contacto entre Atlántico y Mediterráneo: el depósito de El Carambolo, Sevilla. Trabajos de Prehistoria 55(1): 121-138. PINGEL, V., 1992. Die Vorgeschichtlichen Goldfunde der Iberischen Halbinsel. Eine Archäologische Untersuchung zur Auswertung der Spektralanalysen. Madrider Forschungen, Band 17. Berlin, Walter de Gruyter und Co. SCHUBART, H. 1975. Die Kultur der Bronzezeit im Südwesten der Iberischen Halbinsel. Madrider Forschungen 9. Berlin, Walter de Gruyter und Co. SILVA, A.C. and BERROCAL-RANGEL, L., 2005. O Castro dos Ratinhos (Moura), povoado do Bronze final do Guadiana: 1ª campanha de excavaçôes (2004). Revista Portuguesa de Arqueología 8(2): 129-176. VALÉRIO, P., CERQUEIRA, L., MONGE, A., BERROCAL-RANGEL, L., SILVA, A. and ARAÚJO, F., 2007. Early Iron Age gold artefacts from Castro dos Ratinhos (Moura, Southern Portugal), in Archaeometallurgy in Europe 2007, (Aquileia-Grado, 17-21 June). CDRom. Milano, Associazione Italiana di Metallurgia. ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? Two small orientalising spirals (Rome, 10th-9th century BC): common objects – precious jewels Deux petites nattes en spirale orientalisantes (Rome, Xe-IXe siècle avant J.-C.) : objects du quotidien – bijoux précieux Daniela Ferro*, Alessandro Bedini** and Ida Anna Rapinesi*** Abstract: The finding of hair decorations is not rare in archaeological excavations. It is well known that pins, gold nets, gold wires and threads embellished noble ladies’ heads, but it is less common to find small accessories meant to bind braids created as small jewels, such as the 7th century BC spirals found in Lazio. Two hair coils of great refinement, probably used as ‘braid binders’, with terminals decorated by filigree have been found in two funeral sets from two princely tombs of the Laurentina Acqua Acetosa Necropolis (n. 74 and n. 133). The objects present similarities with other items of the same typology from Amendolara (CS), Narce, Marsiliana and Vetulonia. The item from tomb 133 is made of silver covered by electrum, while the second one, from tomb 74, was made entirely in gold. Their analytical study allowed the determination of peculiar details concerning their fabrication, from the composition of the constituent materials to the technological solutions used for their production. The scientific observations, carried out by scanning electron microscopy (SEM) and energy dispersion spectroscopy (EDS), furnished important data for the characterization of the single object, to be used also as a base for comparison with the few homologous jewels known to date. Résumé : La découverte de décorations de cheveux dans les fouilles archéologiques n’est pas rare. Il est bien connu que les têtes des dames nobles étaient ornées de coiffures d’or, mais il est peu fréquent de découvrir de petits accessoires, réalisés comme de petits bijoux, servant à coiffer les cheveux. Les pièces concernées par ce travail sont datées du VIIe siècle avant J.-C. et ont étés découvertes dans le Lazio. Deux spirales très raffinées, probablement utilisées comme décoration de nattes et dont les extrémités sont ornées de filigranes, ont été trouvées dans les fouilles de deux tombes princières de la nécropole Laurentina (n.74 et n.133). Les objets présentent des similitudes avec d’autres de même typologie provenant d’Amendolara (CS), Narce, Marsiliana et Vetulonia. Le bijou de la tombe 133 est en argent couvert d’électrum alors que la deuxième pièce de la tombe 74 est entièrement réalisée en or. Leur étude analytique a permis de déterminer les détails particuliers de leur fabrication, la composition des matériaux constitutifs et les escamotages technologiques utilisés pour leur création. Les observations scientifiques, réalisées à l’aide d’un système de microscopie électronique à balayage et par microanalyse électronique, fournissent des éléments importants pour la caractérisation des objets. Keywords: gold hair decoration, filigree, electrum-silver, production technology, SEM-EDS. Mots-clés : coiffures d’or, filigrane, électrum-argent, technologie de fabrication, MEB-EDS. * Consiglio Nazionale delle Ricerche – Institute for Nanostructured Materials (CNR-ISMN), c/o Department of Chemistry, Sapienza University of Rome, P.le A. Moro 5, 00185 Rome, Italy (daniela.ferro@cnr.it) ** Soprintendenza Speciale per i Beni Archeologici, Rome, Italy *** Soprintendenza Speciale per i Beni Archeologici, Rome, Italy rec. Aug. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 165-168 166 1. INTRODUCTION Among personal female objects from archaeological contexts, one of the most frequent is the hair coil, linked to a particular hair-style that seems to be a constant throughout the entire Iron Age (10th-9th century BC) in the Tyrrhenian area. This style was characterized by two plaits of hair on each side of the head, fixed by spirals made of a simple wire wrapped more or less tightly, with diameters ranging from 2-2.5 to 4 cm. This type of spiral appears in the tombs of the first Iron Age (11th-10th century BC), both in the necropolises of the Latium area, as well as in the Faliscan and Veio areas. The spirals are probably a translation in metal of strings of organic fibres but, in any case, it is not possible to separate their decorative function from the more functional properties of the spring spiral. Specific metal alloys had to be used, and a particular metal treatment had to be chosen as a function of the ‘elastic’ effect of the ornamental object. In general, the spirals were composed of two or more turns of double wires with the terminal part often flat and undulate, with a diameter varying from 2 to 4 cm. In the richest tombs of the 8th and 7th centuries BC, representative of a more differentiated society, these simple coils evolved into more precious objects through the use of noble metals finely decorated by elements in filigree or gold lamina. These last two variants characterize the jewels of the orientalising period (Formigli, 1976). The shape of the spirals varies: some exemplars present a double torsion, while others are decorated all along the wire; however, in all cases, they represent a concentration of style, jewellery, technology and science containing important diagnostic information. Previously, two exceptional hair coils were found by Alessandro Bedini in a princely tomb (101) of the Laurentina Necropolis in Castel di Decima (Rome) and dated to the end of the 7th century BC (Bedini, 1977; 1985). In the 7th century BC, found always only in princely tombs, the spirals have a small diameter of about 1 cm, perhaps indicative of a new fashion of hair style, with the two terminals decorated by a horizontal and undulating motif in filigree between two series of horizontal parallel threads, as visible in the exemplar from tomb 74 described in the present work. This object can be compared with other hair coils of silver covered by electrum of the ‘pseudo-camera’ tomb (133) of the same necropolis (Bedini and Cassotta, 2006). This new typology is also present in the Faliscan area (Narce) and in Etruria (Marsiliana and Vetulonia), and is probably of Greek inspiration, as the exemplars from Amendolara (Cosenza) indicate (De la Genière, 1973). ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? Daniela FERRO, Alessandro BEDINI, Ida Anna RAPINESI The analytical study by non-destructive methods is aimed at gathering information on the ancient goldsmiths’ degree of knowledge and ability to optimize the elastic properties of metals through specific treatments. Scanning electron microscopy (SEM) and X-ray energy dispersive microanalysis (EDS) have been employed for the purpose of determining the micro-morphology of the metal surface and the chemical composition of the areas that present some details of interest for the technological definition of processes. From the interpretation of the morphological and analytical data, it is possible to derive information on micro-diagnostic markers for the characterization of the spring metal treatment, to be compared to similar objects. 2. METHODS The use of scanning electron microscopy (SEM) and energy dispersion spectroscopy (EDS) has been employed to investigate both the technological peculiarity of the item, as well as the possible transformation of the precious metal into an elastic spring. In the year 1678, the physicist Robert Hooke formulated a method useful for the description of the macroscopic action of a spring, by studying the mechanical behaviour of a metallic wire wrapped in a cylindrical-shaped spiral operating by compression or strength. In both cases, the microscopic effect of the elongation/compression motion is the stretching of the intermolecular bonds, within the limits of elastic deformation, after which the atomic bonds rearrange, causing irreversible deformation. The archaeological evidence indicates that the existence of springs dates back to the Bronze and Copper Ages. Examples of metal artefacts functioning as springs or containing spring actions are: fibulae widely used in the Mediterranean area, certain fastenings of necklaces, bracelets, earrings, and diverse typologies of hair binders, the subject of the present work. The origin of the spring is beyond the scope of this article, but we can start analyzing some peculiar aspects of these items, useful for the identification of the intentional creation of an elastic system. A metal presents a crystalline structure constituted by grains of different shape and dimension, dependant on a series of factors, including mechanical and thermal stresses. Particular working processes are necessary in order to exploit the cold plastic properties of any metallic material. In fact, only by employing cold-working processes, such as hammering or drawing, is it possible to directionally elongate the grains. Not all metals are suitable for making springs, and specific compositions of alloys, as well as particular mechanical procedures, are necessary to enhance characteristics of Two Small orientalising spirals (Rome, 10th-9th century BC)… elasticity. The impossibility of any metallographic sampling of the precious object only allowed an analytical study based on the morphology of the surface and on the microstructure and chemical composition of small areas. Anticipating more detailed studies, the information derived from the understanding of the working process suggests that spirals are not to be considered solely as simple hair decorations, but mainly as a technological achievement. 3. RESULTS Laurentina Necropolis tomb 133 – The analyzed object is a silver strip spiral (ø 0.750 mm) covered by an electrum sheaf, with a high Ag content, circa 40 wt%, as revealed by EDS analysis. An electrum wire, produced by torsion of a small lamina with a composition similar to the lamina of the spiral body, makes up the filigree decoration on the terminal parts (Fig. 1). Beyond the possible economic reasons for the choice of a two body system consisting of an inner strip of silver mechanically enclosed in a lamina of electrum, it is suggested here that the artefact’s system was also developed for its elastic properties and general wear resistance to further unintentional cold plastic deformation. Longitudinal lines all along the strip’s length are observable on the internal side of the lamina, suggesting the use of a burnishing tool to further cold harden the electrum lamina and ‘iron out’ eventual buckles and kinks that tend to form on the inside surface and lateral inner edges while forming the spiral. Figure 1: Spiral in silver/electrum, from Laurentina Necropolis tomb 133. Figure 1: Spirale en argent/électrum provenant de la tombe 133 de la Nécropole Laurentina. 167 The formation of Ag corrosion compounds, such as AgCl and Ag2S, has been detected by SEM-EDS analysis. The natural formation of Ag corrosion products on the internal wire is the likely cause of electrum lamina detachments. Furthermore, the preferential corrosion of the silver alloyed in the electrum produces a black patina on the gilded coating (Carraro et al., 2005). The irreversibility of these physical-chemical processes involving the metal microstructure allows their consideration as diagnostics in authenticity tests. Concerning the soldering process employed to fix the filigree elements, EDS analysis revealed the presence of silver soldering, where the Ag content was 100%. The use of silver soldering is typical for filigree work from the Classic and Imperial Roman periods (Ferro et al., 2008), owing to its high degree of wettability, leaving intact the decoration’s tridimensionality. In this case, in which the filigree was made with an electrum alloy with a low melting point, the excessive heating caused the loss of relief definition. Laurentina Necropolis tomb 74 – A simple gold wire forms a regular spiral with the terminals decorated with filigree. The filigree wire (ø 0.2 mm) was produced by torsion of a gold strip (Fig. 2). The precious metal composition of the coil and of its decorations shows a high gold content (Au 94 wt%), with silver and copper as the other major elements, a composition typical for precious objects from the Roman period. The use of high gold is not optimal for the production of an ‘elastic’ metal, despite cold work hardening procedures. Figure 2: Spiral in gold from Laurentina Necropolis tomb 74. Figure 2 : Spirale en or provenant de la tombe 77 de la Nécropole Laurentina. ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? 168 Longitudinal and parallel traces all along the wire’s length are well visible also in this coil, suggesting the use of a burnishing tool employed to further surface harden the wire while creating a polished finish. From the verified invariance of the spring wire’s diameter (0.4 mm), it is possible to assume the use of a procedure employing a calibration system. The filigree soldering was carried out with silver, found in the form of snippets, as observed in the false colour EDS map, with silver in red, copper in green and gold in blue (Fig. 3). The fact that the choice of silver as a solder is verified once more suggests an improvement of the scientific knowledge derived from empirical experience: in fact, the wettability of silver is higher than that of most brazing alloys, due to its superficial tension properties. With respect to the spring from tomb 133, the filigree appears more detailed, because the heat induced to melt silver in this case had no influence on the nearly pure gold, with a much higher melting point. 4. DISCUSSION AND CONCLUSION The investigation of the two hair coils is part of a more complex study of the technology applied in their production. The results obtained on the hair coils from Via Laurentina, compared with those obtained on other similar items originating from Etruscan areas (Carraro and Ferro, 2008), allows the identification of different fabrication procedures; however, all of them were carried out according to Figure 3: (See colour plate) EDS false colour X-ray map of the filigree soldering area in the spirals from tomb 74: Ag – red, Au – blue, Cu – green. Figure 3 : (Voir planche couleur) Cartographie de rayons X en fausses couleurs de la soudure du filigrane sur la spirale de la tombe 74: Ag-rouge, Au-bleu, Cu-vert. ArcheoSciences, revue d’archéométrie, 33, 2009, p. ??-?? Daniela FERRO, Alessandro BEDINI, Ida Anna RAPINESI common rules and design. Further analyses on such samples, reproduced by experimental archaeology, will be necessary in order to identify distinctive elements for this type of common, yet precious, objects. Acknowledgments The authors are grateful to David Loepp for the editing and to Vania Virgili for her collaboration. References BEDINI, A., 1977. L’ottavo secolo nel Lazio e l’inizio dell’Orientalizzante antico, alla luce di recenti scoperte nella necropolis di Castel di Decima. La Parola del Passato 32: 274-309. BEDINI, A., 1985. Tre corredi protostorici dal Torrino. Osservazione sull’affermarsi e la funzione delle aristocrazie terriere nell’VIII secolo nel Lazio. Archeologia Laziale 7 (QAEI 11): 44-64. BEDINI, A. and CASSOTTA, A., 2006. Technical reports, in M.A. Tomei (ed.), Memorie dal Sottosuolo-Ritrovamenti Archeologici 1980-2006. Milano, Electa Press, 467 and 479. CARRARO, A. and FERRO, D., 2008. Binomio archeometria e restauro nello studio delle oreficerie etrusche, in Lo Stato dell’Arte congresso nazionale IGIIC. Spoleto: IGIIC publisher, 521-527. CARRARO, A., FERRO, D., MARGHELLA, G. and ZOCCOLILLO, L., 2005. Multidisciplinary diagnostic approach to the degrade state of the jewel exposed in a showcase of Villa Giulia museum in Rome, in Acta 8th International Conference on Non destructive testing and microanalysis for diagnostic and conservation of the cultural and environmental heritage - May, 15th-19th Lecce, Italy. CD-ROM. DE LA GENIÈRE, J., 1973. A propos de quelques mobiliers funéraires d’Amendolara, Mélanges de l’École Française de Rome Antiquité 85/1: 7-53. FERRO, D., FORMIGLI, E., PACINI, A. and TOSSINI, D., 2008. La saldatura nell’oreficeria antica. Roma, Ed. Kappa. FORMIGLI, E., 1976. L’antica tecnica dei bracciali a filigrana. Studi Etruschi 44: 203-210. Investigation of ancient gold objects from Artemision at Ephesus using portable µ-XRF Enquête sur les objets anciens en or de l’Artemision à Ephèse au moyen d’un équipement de µFX portable Michael Melcher*, Manfred Schreiner*, Birgit Bühler**, Andrea M. Pülz*** and Ulrike Muss**** Abstract: During two measurement campaigns in the Archaeological Museum in Istanbul and the Ephesus Museum in Selçuk, Turkey, more than 80 ancient gold objects (e.g. appliqués, statuettes and coins) from the 8 th-6th centuries BC were analyzed using portable XRF instruments in order to obtain information on the chemical composition and homogeneity of the artefacts discovered during excavations in the sanctuary of Artemis at Ephesus. hese results and complementary studies on the manufacturing techniques employed for making these objects will contribute to our knowledge of metalworking, trade, as well as the transfer of ideas and technologies at Ephesus and in Western Asia in the Archaic period. he objects can be characterized as homogeneous alloys, mainly consisting of gold (Au), silver (Ag) and copper (Cu). A high compositional variation of Au and Ag between diferent objects could be determined (Au between 48.7 and 99.9%, Ag between <detection limit (dl) and 50.9%, Cu between <dl and 4.4%). hese results demonstrated the range of gold alloys represented within the collection and allowed a discrimination between natural and artiicial gold alloys (including the identiication of ‘reined’ gold). In some cases, objects which were closely related from a typological, stylistic and/or technological point of view were also similar in composition, difering only in the sub-percent range. Résumé : Plus de 80 objets anciens en or (p.ex. appliques, statuettes et monnaies) datés du VIIIe-VIe siècle av. J.-C. et trouvés lors des fouilles du sanctuaire d’Artémise à Ephèse, ont été analysés au moyen d’un équipement de FX portable pendant les deux campagnes de mesure réalisées au musée archéologique d’Istanbul et au musée d’Ephèse à Selçuk, Turquie, de façon à obtenir des informations sur leur composition chimique et sur leur homogénéité. Ces résultats, complémentés par des études sur leurs techniques de fabrication, contribuent à la connaissance du travail du métal, de son commerce, ainsi que du transfert d’idées et technologies à Ephèse et dans l’Asie de l’Ouest pendant la période Archaïque. Les objets ont été fabriqués avec des alliages homogènes, se composant en majorité d’or (Au), argent (Ag) et cuivre (Cu). Une forte variation de teneurs a pu être mise en évidence pour l’Au et l’Ag dans le cas de diférents objets (Au entre 48,7 and 99,9 %, Ag entre < limite de détection (dl) et 509 %, Cu entre < dl et 4,4 %). Ces résultats montrent la gamme d’alliages d’or de la collection et permettent de discriminer entre alliages d’or naturels et artiiciels (en incluant l’identiication d’or « ainé »). Dans certains cas d’objets de typologie, style et/or techniques de fabrication proches, leurs compositions sont identiques, la diférence n’apparaissant qu’au niveau du sous-pourcent. Keywords: Gold, Ephesus, Artemision, XRF. Mots-clés : Or, Éphèse, Artemision, FX. * Institute of Science and Technology in Art – Academy of Fine Arts, Schillerplatz 3, 1010 Vienna, Austria. (m.melcher@akbild.ac.at) (m.schreiner@ akbild.ac.at) ** Vienna Institute for Archaeological Science (VIAS) – University of Vienna, Franz-Klein-Gasse 1, 1190 Vienna. (birgit.buehler@univie.ac.at) *** Österreichisches Archäologisches Institut – Franz-Klein Gasse 1, 1190 Vienna. (andrea.puelz@oeai.at) **** Institut für Klassische Archäologie – University of Vienna, Franz-Klein-Gasse 1, 1190 Vienna. (ulrike.muss@univie.ac.at) rec. Aug. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 169-175 170 1. INTRODUCTION AND HISTORICAL BACKGROUND About 1.500 gold objects were found in the sanctuary of the goddess Artemis at Ephesus, Turkey (Pülz, 2009). Most of these occurred in strata located under the Archaic dipteros – the so-called “Temple of Croesus” – and may thus be dated to the second half of the 7th or the irst half of the 6th century BC (Hogarth, 1908; Bammer and Muss, 1996). Within the Archaic period, the Artemision is unique in terms of the large number and variety of gold jewellery. No other sanctuary from this period has yielded a similar wealth of gold objects. During the British and Austrian excavation campaigns at this site, a small number of igural objects, such as statuettes in human or animal form and miniature objects, were found. hey were deposited in the sanctuary as votive gifts. he vast majority of gold objects, however, are dress ornaments and jewellery. Among the gold inds, approximately 530 appliqués were found in the sanctuary of Artemis, representing at the same time the major group of artefacts (Pülz and Bühler, 2006). Most of them show holes on the edges and can therefore be addressed as garment decorations (e.g. Selçuk, inv. no. 1/71/89, Fig. 1). Because of the lack of parallels which would point to individual appliqués votive oferings in the Artemision, an interpretation of these objects as representing decorations of the cult image (Romano, 1988) or ritual garment oferings seems likely. Figure 1: (See colour plate) Appliqué from the sanctuary of Artemis (Selçuk, inv. no. 1/71/89). Figure 1: (Voir planche couleur) Appliqué du sanctuaire d’Artemise (Selçuk, inv. no. 1/71/89). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 169-175 Michael MELCHER et al. It is interesting, however, that none of the gold statuettes found in the Artemision from the 7th or 6th centuries BC which represent the adored goddess show garments or dresses with appliqué decorations, like later copies. A solemn igurine wears a dress whose border is decorated with meander patterns (Selçuk, inv. no. 2/59/80). he second most important group of gold jewellery items consists of spherical and drop-shaped pendants and beads (some of which look like fruit), pins with loral heads, ibulae (predominantly of the Phrygian ‘Asia Minor’ type), brooches in the shape of birds of prey, and boat-shaped earrings with, in some cases, elaborate decoration (Bühler and Pülz, 2009). he vast majority of gold earrings found in the Archaic temple of Artemis at Ephesus belong to the so-called ‘boat-shaped’ type (sometimes also referred to as ‘leech-shaped’), with a swollen body, which may be plain or decorated. Earrings of this type are very common in Western Anatolia. he best parallels for the examples from Ephesus were found in the Güre region (Özgen and Öztürk, 1996), Sardes, Lydia (Waldbaum, 1983), the Bayındır grave mound in northern Lycia (Özgen and Öztürk, 1996) and among the objects from the Norbert Schimmel Collection (Muscarella, 1974). Most of the motifs or igurines refer to Artemis as the goddess of fertility, or her representation as the mistress of animals. For example, the bird of prey (Istanbul, inv. no. 3093, Fig. 2) is considered to be sacred to the goddess, and Figure 2: (See colour plate) Bird of prey (Istanbul, inv. no. 3093). Figure 2: (Voir planche couleur) Rapace (Istanbul, inv. no. 3093). Investigation of ancient gold objects from Artemision using portable µ-XRF it is found in the form of brooches, pendants or statuettes in the sanctuary (Bühler and Pülz, 2008). he bee, a symbol of reincarnation and fertility which is closely connected with the cult of Artemis, also appears in many variations, from naturalistic to very abstract forms (Pülz, 2001). Various pendants, beads and pin heads resemble fruits or seed and can therefore be interpreted as symbols of fertility. However, there are also individual inds, like a gold object in the form of a barley seed, with a real seed inside, which links it directly to the cult of Artemis. here is no other sanctuary in the Archaic period which is, both in terms of quantity and quality, comparable to the Artemision in terms of the assemblage of gold objects. Only collections of artefacts originating from graves, especially gold jewellery from Rhodes (650-600 BC) (Laineur, 1978), and from the Lydian Tumuli (Özgen and Öztürk, 1996) are similar in numbers. he latter, of course, are dated to a much later period (the end of the 6th and the beginning of the 5th century BC), but, in terms of their forms, motifs, and techniques, they strongly resemble the Ephesian inds. When considering the artefacts found in the Artemision, it can be shown that the majority of them are related to forms indigenous to Ionia, and that they were created in local workshops. he unique ibulae decorated with a combination of lion heads and falcon heads embedded in a lower (Selçuk, inv. no. 1/43/94, Fig. 3) have no parallels Figure 3: Fibulae decorated with a combination of lion heads and falcon heads embedded in a lower (Selçuk, inv. no. 1/43/94). Figure 3 : Fibule décorée d’une combinaison de têtes de lions et de faucons entourée de motifs loraux (Selçuk, inv. no. 1/43/94). 171 in the Archaic world, and can therefore be addressed as a speciic Ephesian product (Freiberger and Gschwantler, 2008). On the other hand, there are a few objects, such as, for example, a richly decorated head of a grifon (Selçuk, inv. no. 117/61/90) or a piece of jewellery in the form of a blossom, whose style and technology is strikingly similar to Rhodes jewellery of the Orientalizing period, so that the hypothesis of an import can be considered. Statuettes and miniature objects (such as human extremities and small vessels) in particular were not objects intended for daily use, but designed for the purpose of giving. hrough precious votive oferings, such as gold objects, one could show not only one’s religious faith but also demonstrate one’s status as compared to other individuals. However, votive oferings were obviously even more than that: they were also prestige objects for the recipient, that is the deity or its mortal representative (i.e. the priest), because, through precious gifts, status and power over other sanctuaries were strengthened (Godelier, 1999). 2. METHODS 83 of these objects, among them 11 coins, were selected for material analysis using two diferent self-built portable X-ray luorescence (XRF) instruments (henceforth referred to as XRF I and XRF II), according to the diferent requirements of the areas on the objects’ surfaces to be analyzed. Instrument XRF I (COPRA, Compact Portable Roentgen Analyzer, Fig. 4), constructed within the EU-Project No. SMT4-CT98-2237 and used for the measurement campaign in the Istanbul Archaeological Museum, is characterized by a very small diameter (half-value width of 126 µm at 40 kV, 0.4 mA, Fe-Kα) of the primary X-ray beam (anode material: molybdenum), which is achieved by a polycapillary. As a result, this instrument is particularly suited for investigations of micro-domains on the surfaces of (gold) objects. A proper focusing on the desired domain is achieved by an external object positioning system (allowing for a positioning of the object with an accuracy of 0.01 mm in all 3 spatial directions), a microscope and a CCD-camera. For the detection of the luorescence radiation, a drift chamber detector (Röntec X-lash L) with an energy resolution of <170 eV is used. Instrument XRF II (Fig. 5), used for investigations in the Ephesus Museum in Selçuk, is equipped with a Rhodium (Rh) X-ray tube (Oxford XTF5011), a silicon drift detector (Röntec XFlash 1000), and a positioning system consisting of two laser beams (Desnica, 2005; Desnica and Schreiner, 2006). he diameter of the primary beam is approximately ArcheoSciences, revue d’archéométrie, 33, 2009, p. 169-175 172 Michael MELCHER et al. set on an object’s surface. he spectra were quantiied using 18 appropriate standard materials in the systems Au-Ag-Cu and Ag-Cu, as well as the software WinAxil (Win Axil X-ray Analysis Software, Canberra Benelux, Belgium). he crosschecking of the quality of the quantiication in order to estimate analytical errors was carried out by the quantiication of these standard materials. Relative errors of less than 3% for the main components Au and Ag (compositional range ≥30 wt%), and less than 20% for the minor component Cu (compositional range ≤5 wt%) were veriied. Figure 4: Scheme of the instrument XRF I used for the investigation of objects in the Archaeological Museum, Istanbul. Figure 4: Schéma de l’équipement de FX-I utilisé dans l’étude des objets du musée archéologique d’Istanbul. Figure 5: Scheme of the instrument XRF II used for the investigation of objects in the Ephesus Museum, Selçuk. Figure 5 : Schéma de l’équipement de FX-II utilisé dans l’étude des objets du musée d’Ephèse à Selçuk. 1 mm. Due to their relatively low weight, both instruments are especially suited for investigations of objects on site (i.e. in museums, collections, or near excavations) and enable a qualitative and quantitative determination of major, minor and trace elements in a non-destructive manner. For all measurements, a tube voltage of 35 kV, a tube current of 0.8 mA, and a measurement time of 200 s were used. Additionally, the tube was equipped with a 12.5 µm palladium (Pd)-ilter for suppression of difraction peaks in the resulting spectrum. Typically, 3-4 measurement points were ArcheoSciences, revue d’archéométrie, 33, 2009, p. 169-175 3. RESULTS AND DISCUSSION Table 1 presents the results of the analyses in tabular form. In addition to the major and minor elements Au, Ag and Cu, in some cases Fe was also detected in small amounts. As the Fe intensities also showed signiicant variations between diferent measurement points, the presence of Fe might be due to supericial contaminations of the objects. he most important result of these XRF analyses is the identiication of 8 objects that were made of a gold alloy with very high Au content, of more than 97%, that is of almost pure – and therefore reined – gold. Four of these objects have an Au content of more than 99%, containing only traces of Cu and/or Ag: one is a cast miniature pendant in the shape of a human being (Istanbul, inv. no. 3071), while the other three objects (Istanbul, inv. nos. 3175 a, 3175 b and Selçuk, inv. no. 42/41/86) are appliqués of identical type (type b.5variant 10, according to A.M. Pülz; see e.g. Pülz and Bühler 2006). he other four objects in this group are high-quality jewellery items which consist of a large number of individual sheet-gold components and ine decorative elements (mainly granulation, in some cases also beaded wires): a brooch (Istanbul, inv. no.3157) and an earring (Istanbul, inv. no. 3033), which may have been products of the same workshop, as well as an insect-shaped miniature pendant (Istanbul, inv. no. 3087) and brooch (Istanbul, inv. no. 3088). he Au content of these four objects varies between 97.2% and 98.9%, the Cu content between 1.1% and 2.8%, and they all contain no Ag. As these four objects consist of a large number of components, it is possible that at least some of the copper detected on the surface of these objects is due to the process of reaction soldering with copper salts, which may have been used to join these components. If this assumption is correct, the actual gold content of the sheetgold components used to produce these objects may have been even higher. Signiicantly, all eight jewellery items with an Au content of more than 97% are products of outstanding quality. 173 Investigation of ancient gold objects from Artemision using portable µ-XRF Object inv. no. Au Ag Cu Object inv. no. Au Ag Cu Object inv. no. Au Ag Cu 3175 99.9 (0.0) 0.0 (0.0) 0.1 (0.0) 105/30/77 70.0 (2.5) 28.2 (2.5) 1.9 (0.0) 184/54/88 58.0 (1.5) 40.5 (1.6) 1.6 (0.2) 3176 99.9 (0.0) 0.0 (0.0) 0.1 (0.0) 50/41/86 68.6 (0.3) 28.7 (0.1) 2.9 (0.3) 18/41/86 57.7 (0.7) 41.0 (0.9) 1.3 (0.3) 3071 99.5 (1.4) 0.0 (0.0) 0.6 (1.4) 53/32/85 68.6 (3.4) 29.4 (3.5) 2.1 (0.4) 3079 57.6 (5.0) 40.4 (5.4) 2.0 (0.4) 42/41/86 99.5 (0.1) 0.4 (0.1) 0.1 (0.0) 115/61/90 68.1 (0.4) 31.1 (0.5) 0.8 (0.0) 17/41/86 57.0 (1.0) 41.5 (1.0) 1.6 (0.1) 3157 98.9 (0.7) 0.0 (0.0) 1.1 (0.7) 1/42/93 67.5 (2.3) 31.9 (2.1) 0.6 (0.2) 58/38/81 56.6 (2.6) 41.4 (2.4) 2.0 (0.2) 3087 98.4 (1.1) 0.0 (0.0) 1.6 (1.1) 47 67.3 (0.2) 32.0 (0.3) 0.8 (0.1) 29/68/89 56.2 (1.3) 42.2 (1.7) 1.7 (0.4) 3033 98.3 (1.7) 0.0 (0.0) 1.7 (1.7) 37/41/86 67.0 (3.4) 31.7 (3.3) 1.4 (0.1) 3078 55.7 (5.9) 42.4 (6.4) 1.9 (0.5) 3088 97.2 (0.5) 0.0 (0.0) 2.8 (0.5) 12/42/93 66.2 (1.3) 31.3 (1.2) 2.6 (0.1) 3106 55.3 (4.4) 42.3 (4.4) 2.4 (1.5) 133/61/87 95.3 (0.0) 4.0 (0.2) 0.8 (0.1) 1/59/80a 66.0 (0.7) 33.0 (1.0) 1.0 (0.7) 2/59/80 55.1 (2.0) 43.8 (2.3) 1.7 (0.3) 2/42/93 92.4 (1.7) 3.8 (0.8) 3.9 (1.0) 45/41/86 65.9 (5.7) 32.9 (5.8) 1.3 (0.1) 3061 52.6 (1.8) 45.4 (2.7) 2.1 (0.9) 12/41/86 92.2 (1.3) 6.0 (0.9) 1.9 (0.4) 1/59/80b 65.8 (0.7) 33.2 (0.7) 1.1 (0.1) 3062 52.2 (1.3) 45.6 (1.9) 2.2 (0.6) 3/42/93 91.3 (2.5) 4.3 (0.9) 4.0 (1.3) 13/42/93 64.2 (2.2) 33.5 (2.3) 2.4 (0.1) 3077 51.5 (1.2) 45.7 (1.4) 2.9 (0.4) 4/42/93 91.1 (3.4) 4.6 (0.7) 4.4 (3.2) 3/51/91 62.9 (0.0) 35.1 (0.1) 2.0 (0.2) 3118 50.2 (2.1) 47.8 (2.0) 2.0 (0.4) 1/43/94 91.0 (1.0) 4.8 (0.5) 4.3 (1.1) 3159 62.8 (3.6) 34.9 (3.1) 2.3 (1.4) 32/68/89 50.0 (0.7) 46.8 (0.8) 3.2 (0.2) 3084 84.2 (2.9) 14.6 (3.1) 1.1 (0.2) 38/68/89 62.6 (0.5) 36.2 (0.4) 1.2 (0.0) 183/54/88 49.0 (4.7) 49.6 (4.1) 1.4 (0.8) 40/41/86 48.7 (9.8) 50.9 (9.7) 0.4 (0.0) 3083 84.1 (2.2) 14.9 (2.9) 0.8 (0.7) 50/32/85 62.4 (2.8) 34.3 (3.0) 3.3 (0.1) 117/61/90 83.4 (1.4) 15.4 (0.5) 1.2 (0.9) 7/43/94 62.3 (2.1) 35.6 (2.0) 2.0 (0.6) Coins 3090 80.6 (8.7) 16.0 (7.4) 3.4 (3.2) 15/43/94 61.9 (2.8) 36.3 (2.8) 1.8 (0.1) 101/41/86 83.2 (4.2) 15.7 (4.2) 1.2 (0.1) 1/41/86 78.7 (0.6) 18.5 (0.3) 2.9 (0.9) 52/32/85 61.3 (1.3) 37.3 (1.2) 1.5 (0.1) 99/43/94 66.5 (7.0) 31.9 (7.3) 1.6 (0.3) 111/61/87 78.1 (0.8) 20.3 (0.7) 1.7 (0.1) 75/32/85 61.2 (1.5) 36.6 (1.6) 2.2 (0.4) 99/41/86 64.5 (1.2) 34.7 (1.2) 0.8 (0.1) 63/32/85 77.5 (0.2) 19.9 (0.4) 2.6 (0.2) 29/51/91 61.2 (0.7) 36.1 (0.6) 2.8 (0.1) 100/41/86 62.2 (2.0) 36.7 (2.1) 1.2 (0.1) 135/61/87 77.2 (-) 21.0 (-) 1.8 (-) 91/32/85 60.8 (3.0) 36.9 (3.3) 2.4 (0.3) 98/43/94 60.7 (1.3) 37.7 (1.2) 1.7 (0.1) 182/54/88 74.4 (0.9) 23.3 (1.1) 2.3 (0.2) 3093 60.8 (1.0) 36.3 (1.8) 2.9 (0.8) 318/61/87 60.7 (4.4) 37.0 (4.1) 2.4 (0.5) 3076 74.2 (7.3) 24.2 (7.2) 1.7 (0.3) 174/38/81 60.6 (1.6) 35.8 (1.7) 3.6 (0.2) 96/41/86 60.6 (1.9) 38.4 (1.7) 1.1 (0.2) 110/61/87 73.9 (0.8) 24.0 (1.0) 2.2 (0.1) 3039 60.1 (5.0) 35.7 (5.6) 4.3 (3.2) 95/41/86 60.5 (3.2) 38.3 (3.2) 1.2 (0.0) 112/61/87 72.5 (0.4) 25.1 (0.5) 2.5 (0.0) 6/19/81 59.7 (0.7) 37.9 (0.6) 2.5 (0.0) 317/61/87 57.1 (3.8) 41.1 (3.4) 1.8 (0.4) 17/51/91 71.4 (3.9) 27.0 (3.6) 1.6 (0.3) 1/71/89 58.8 (2.3) 37.9 (2.9) 3.3 (0.7) 98/41/86 56.4 (0.3) 42.1 (0.3) 1.6 (0.0) 8/59/80 70.1 (1.4) 27.6 (1.1) 2.3 (0.4) 10/43/94 58.5 (1.5) 39.2 (1.4) 2.4 (0.1) 23/51/91 43.4 (6.8) 51.1 (6.9) 5.6 (0.1) Table 1: Composition (average values of typically 3-4 measurements) of all 83 gold objects (72 jewellery items of diferent types, appliqués and statuettes, as well as 11 coins) examined in the Archaeological Museum, Istanbul and in the Ephesus Museum, Selçuk. he data are sorted by descending Au content. he numbers in brackets indicate the range (maximum value minus minimum value) observed in the single measurements. Tableau 1: Composition (teneurs moyennes de typiquement 3-4 mesures) de tous les 83 objects en or étudiés au musée archéologique d’Istanbul et dans le musée d’Éphèse à Selçuk. Les 11 derniers objets sont des monnaies. Les données sont répertoriées par valeur décroissante de la teneur en Au. Les valeurs entre parenthèses correspondent aux gammes (valeur maximale et valeur minimale) observées pour les mesures. he use of gold alloys with a high gold content for the production of a number of jewellery items from the sanctuary of Artemis at Ephesus is signiicant because this gold was obviously reined. Considering the evidence from the gold reinery at Sardis, Lydia (see Ramage and Craddock, 2000), which is dated to “no later than the middle of the sixth century BC” (Ramage and Craddock, 2000: 95), it is possible that either the raw material (the reined gold – perhaps in the form of gold bars or coins?) or the inished jewellery items were imported from Sardis. Provided this assumption is true and the date suggested for the Sardis reinery is correct, the use of reined gold could be employed as a chronological criterion, which would allow us to ascribe the objects in question to the 6th century BC. At present, most gold objects from the sanctuary of Artemis at Ephesus are ascribed to the second half of the 7th or the irst half of the 6th century BC. An alternative hypothesis would be that the gold was reined in Ephesian workshops. However, to date, there is no evidence available in favour of this theory, and it is not very likely that a gold reinery was active at Ephesus before the workshops recently excavated at Sardis, which provide so far the only certain evidence for the use of the cementation process (separation of silver and gold using salt) known from the ancient world. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 169-175 174 In this context, it is important to note that the Lydians are credited with the invention of struck coinage in the 7th century BC, and that “the irst coins were of native alluvial gold, with the composition carefully adjusted by the addition of small quantities of silver to bring the gold content down to a ixed composition, just below the minimum gold content found in the metal from the Pactolus” (Ramage and Craddock, 2000: 212-213). Analyses of Lydian coins of this type from the collections of the British Museum have shown that the majority of them contain about 54% Au (with a variation of about 2%), 44% Ag (with a variation of about 2%) and 2% Cu (with a variation of 0.5%), as well as traces of Pb (up to 0.2%) and Fe (0.1-0.2%) (Ramage and Craddock, 2000: 172). Two of the ‘Lydian’ coins from the sanctuary of Artemis at Ephesus which were analyzed in the context of the present study have a similar composition (Selçuk, inv. nos. 317/61/87 and 98/41/86), and the same is true for a considerable number of gold objects from the same site. During the reign of the last king of Lydia, Croesus (561547 BC), the irst bimetallic currency system of pure gold and silver was introduced, and this required the ability to separate and reine gold and silver. All analyses of Lydian gold and silver coins have shown them to be of very high purity (98% or more), and gold of similar purity or higher was found on the sherds used to melt the reined metal in the Sardis workshops (Ramage and Craddock, 2000: 169174). Although native gold usually contains silver, typically between 5 and 40%, its copper content rarely exceeds 1% (Ramage and Craddock, 2000: 254). herefore, if a gold alloy contains more than 1-2% Cu, it is likely to be an intentional alloy. Alluvial gold extracted from the river Pactolus varies in composition and is said to contain about 17-30% Ag (Ramage and Craddock, 2000: 172). he area analysis of the cross-section of a natural granule of alluvial gold found during the excavation of the Lydian gold reinery at Sardis has shown it to contain 69.6% Au, 29.8% Ag and 0.6% Cu (Ramage and Craddock, 2000: 148). his is particularly interesting in comparison with a gold globule (Selçuk, inv. no. 42/68/89) found in the sanctuary of Artemis at Ephesus, which contains 67.3% Au, 32.0% Ag and 0.8% Cu. It is thus possible that this is indeed a droplet of unreined native gold, possibly alluvial gold from the river Pactolus or other Lydian deposits. A number of inished gold items found in the sanctuary of Artemis at Ephesus have a similar composition, so that it is possible that they were made of unreined native gold, possibly from Lydian deposits. Signiicantly, all these objects seem to be dated to the 7th century BC. Examples of such ArcheoSciences, revue d’archéométrie, 33, 2009, p. 169-175 Michael MELCHER et al. artefacts are: a high-quality, massive human statuette produced by casting, with extensive cold-working (Selçuk, inv. no. 1/42/93 – 67.5% Au, 31.9% Ag and 0.6% Cu) and three ibulae of the Phrygian ‘Asia Minor’ type (one pair: Selçuk, inv. no. 1/59/80a – 66.0% Au, 33.0% Ag, 1.0% Cu and inv. no. 1/59/80b – 65.8% Au, 33.2% Ag, 1.1% Cu; small single ibula: Selçuk, inv. no. 115/61/90 – 68.1% Au, 31.1% Ag, 0.8% Cu). At least one of the coins of ‘Lydian’ type analyzed as part of this study (Selçuk, inv. no. 99/41/86) also belongs to this category, as it contains 64.5% Au, 34.7% Ag and 0.8% Cu. 4. CONCLUSION he XRF analyses presented in this study have shown that a wide range of gold alloys, both natural and intentional, were used to produce the gold objects from the temple of Artemis at Ephesus (second half of the 7th – irst half of the 6th century BC). Signiicantly, the composition of these gold objects relects important cultural and technological innovations of the period in question, such as the invention of the cementation process to achieve a complete separation of gold and silver, and the invention of coinage (irst, a gold alloy with high silver content in the 7th century BC, then pure gold and silver coins in the 6th century BC). As these innovations took place at Sardis (Lydia), the data provide further evidence for a strong link between the sanctuary of Artemis at Ephesus and the Kingdom of Lydia. A small number of inished gold objects, as well as a piece of workshop debris, seem to consist of unreined native gold with 32-33% Ag and approximately 1.0% Cu or below. Another small group of objects was made of virtually pure (97-98% Au or above) – and therefore reined – gold. he vast majority of objects, however, in particular most of the appliqués, have a gold content in the range of 55-70%, and a copper content of more than 1.5-2.0%, suggesting that they may have been alloys formed by adding additional silver (which presumably also contained some copper) to a natural, unreined gold alloy. Acknowledgement he results presented in this paper were obtained during a research project inanced by the Austrian Science Fund (FWF) and directed by PD Dr. Ulrike Muss. he authors also would like to thank the Turkish Department of Antiquities and Museums (Ankara) and the directors and staf of the Istanbul Archaeological Museum and the Ephesus-Museum Selçuk for the permission to examine the gold objects from the Artemision at Ephesus. Investigation of ancient gold objects from Artemision using portable µ-XRF References BAMMER, A. and MUSS, U., 1996. Das Artemision von Ephesos. Das Weltwunder Ioniens in archaischer und klassischer Zeit. Mainz, Verlag Philipp von Zabern. BÜHLER, B. and PÜLZ, A.M., 2008. Typologie und Technologie der Raubvogeldarstellungen aus Gold, in U. Muss (Hrsg.), Die Archäologie der ephesischen Artemis. Gestalt und Ritual eines Heiligtums. Wien, Phoibos Verlag, 173-184. BÜHLER, B. and PÜLZ, A.M., 2009. Technological and typological studies on gold jewellery from the Archaic Artemision at Ephesos: the ‘boat-shaped’ earrings, in Archaeometallurgy in Europe: 2nd International Conference, Aquileia, Italy, 17-21 June 2007. Milan, AIM, 377-386. DESNICA, V., 2005. Development, construction and application of a portable x-ray luorescence device for analysis of art objects. PhD thesis, Vienna University of Technology, Austria. DESNICA, V. and SCHREINER, M., 2006. A LabVIEW-controlled portable x-ray luorescence spectrometer for the analysis of art objects. X-Ray Spectrometry 35: 280-286. F RE I B E R G E R, V. and G S C H W A N T L E R, K., 2008. Goldschmiedetechnische Beobachtungen zu den Löwenkopibeln aus dem Artemision von Ephesos, in U. Muss (Hrsg.), Die Archäologie der ephesischen Artemis. Gestalt und Ritual eines Heiligtums. Wien, Phoibos Verlag, 185-197. GODELIER, M., 1999. Das Rätsel der Gabe. Geld, Geschenke, heilige Objekte. München, Beck Verlag. HOGARTH, D.G., 1908. Excavations at Ephesus. he Archaic Artemisia. London, British Museum. LAFFINEUR, R., 1978. L’orfèvrerie Rhodienne orientalisante. Paris, De Boccard. 175 MUSCARELLA, O.W., 1974. Ancient Art. he Norbert Schimmel Collection. Mainz, Verlag Phillip von Zabern. ÖZGEN, I. and ÖZTÜRK, J., 1996. Heritage Recovered. he Lydian Treasure. Istanbul, Published by Ugur Okman for Republic of Turkey. PÜLZ, A.M., 2001. Zur Interpretation von Motiven: Die Goldappliken aus dem Artemision von Ephesos, in U. Muss (Hrsg.), Der Kosmos der Artemis von Ephesos. Sonderschriften ÖAI 37. Wien, Verlag des Österreichischen Archäologischen Institutes, 221-232. PÜLZ , A.M., 2009. Goldfunde aus dem Artemision von Ephesos, Forschungen in Ephesos XII/5. Wien, Verlag der Österreichischen Akademie der Wissenschaften (with a contribution from B. Bühler). PÜLZ, A.M. and BÜHLER, B., 2006. Die Goldappliken aus dem Artemision von Ephesos. Studien zur Typologie und Technologie. Österreichische Jahreshefte 75: 229-245. RAMAGE, A. and CRADDOCK, P., 2000. King Croesus’ Gold. Excavations at Sardis and the History of Gold Reining. London, British Museum Press in association with Archaeological Exploration of Sardis, Harvard University Art Museums, Monograph 11. ROMANO, I.B., 1988. Early Greek cult images and cult practices, in R. Hägg, N. Marinatos, G.C. Nordquist (Hrsg.), Early Greek Cult Practice. Proceedings of the 5th International Symposium at the Swedish Institute at Athens, 26-29 June 1986. Stockholm, P. Aströms Förlag, 127-133. SAGONA, C., 2002. Archaeology of Punic Malta. Peters ed., Leuven. WALDBAUM, J.C., 1983. Metalwork from Sardis: he inds through 1974. Sardis Monographs 8. Cambridge, MA, Harvard University Press. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 169-175 Analytical study of the manufacturing techniques of Kushan gold jewellery (National Museum of Antiquities of Tajikistan) Étude analytique des techniques d’orfèvrerie Kouchane (musée national des Antiquités du Tadjikistan) Maria Filomena Guerra*, Guy Demortier**, Maria Luisa Vitobello***, Saidmurod Bobomulloev****, Dominique Bagault*, hierry Borel* and Ikhom Mirsaidov***** Abstract: Little is known about goldsmithing during the period of the Great Kushans. In order to undertake a irst approach to the technological choices and the gold alloys used in the fabrication of the diferent parts of Kushan objects, ive gold jewellery items from the collections of the National Museum of Antiquities of Tajikistan containing wires and granulation were transported from Dushanbe to Paris to be studied within the European research project “AUTHENTICO” with non-destructive scientiic-based techniques at the C2RMF. All the items were found in archaeological excavations carried out in the 1970s in several regions of Tajikistan and were examined by optical microscopy, scanning microscopy, and X-ray radiography. he compositions of the alloys were determined by ion beam analysis (PIXE) at the AGLAE accelerator. Résumé : Les techniques orfèvres de la période des Grands Koushans sont très méconnues. Ain de décrire les choix techniques de l’orfèvre et comprendre le choix des alliages d’or utilisés pour fabriquer des diférentes parties constituant les bijoux, cinq objets d’orfèvrerie de la période Kouchane, appartenant aux collections du musée National des Antiquités du Tadjikistan, ont été transportés de Douchanbe à Paris, dans le cadre du projet européen « AUTHENTICO », pour être étudiés au C2RMF au moyen de techniques scientiiques non-destructives. Tous les objets proviennent de fouilles archéologiques menées dans les années 1970 dans diférentes régions du Tadjikistan. Ils ont été observés par microscopie optique, microscopie électronique et radiographie X. Leur composition a été déterminée par analyse par faisceau d’ions (PIXE) à l’accélérateur AGLAE. Keywords: Kushan, jewellery, analysis, PIXE, gold. Mots-clés : Koushan, orfèvrerie, analyse, PIXE, alliages d’or. * Laboratoire du Centre de Recherche et de Recherche et de Restauration des Musées de France, UMR171 CNRS – 14, quai François-Mitterrand, 75001 Paris, France. (maria.guerra@culture.gouv.fr) ** LARN, University of Namur – 61, Rue de Bruxelles, 5000 Namur, Belgium; Center for Dating and Diagnostics, Department of Engineering of Innovation, Faculty of Cultural Heritage, University of Salento, Lecce, Italy. (guy.demortier@tvcablenet.be) *** National Museum of Antiquities of Tajikistan, Academic Rajobov – 5, Dushanbe, Tajikistan. (said@ac.tajik.net) **** EJTN GEIE, European Jewellery Technology Network – 124 rue du Commerce, Brussels, Belgium. (info@ejtn.org) ***** Nuclear and Radiation Safety Agency – 33 Rudaki Avenue, 734025, Dushanbe. (agentilhom@mail.ru) rec. Sept. 2007 ; acc. Nov. 2007 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 177-185 178 Maria Filomena GUERRA et al. 1. INTRODUCTION he Kushan Empire had its origins in the 2nd century BC, when a confederation of nomadic Indo-European Yuezhi tribes invaded and settled in the Sogdiana and the Bactria areas in Central Asia (Puri, 1999; Rapin, 2007). his branch of the Yuezhi had their origins in north-western China, where they lived until being pushed back by the Xiongnu. he Xiongnu, a pastoral nomadic tribal confederation from the Ordos (horley, 1971) who produced a large quantity of gold work (Bunker, 1993), occupied a large territory, from the Yellow River to the North of Siberia. Only Sima Qian (145-86 BC) provides in his Shiji (Record of the Historian) some details about this civilisation and about their military superiority, which allowed them to defeat the Han and become the irst powerful empire of the steppes along the Chinese frontier (Sarkhosh Curtis and Stewart, 2007). he branch of the Yuezhi pushed back by the Xiongnu settled along the Oxus River and in southern Bactria (present day Afghanistan and Tajikistan), but expanded from the 1st to the 4th century AD, especially under Kanishka I (early 2nd century AD), to north-western India, Pakistan, Uzbekistan and Afghanistan (Dani et al., 1996). Between the 1st and the 3rd century AD, the Kushan controlled Gandhara, where the main road connecting India with the Silk Road and the Mediterranean Sea ran (Wood, 2002; horley, 1971). he Kushan Empire was marked by a time of intensive development of the economy (Mukhamedjanov, 1994) and culture, with two distinct periods separated by a ‘dark age’ (Sarianidi, 1980): the downfall of the Greco-Bactrian Kingdom during the middle of the 2nd century BC, and the powerful empire of the Great Kushan in the 1st-2nd century AD (Dani et al., 1996). Chinese, Iranian and Indian traditions were brought together under the Kushan Empire, but the Greek traditions brought by the conquest of Alexander the Great are also evident in the typology of their art. he objects from Tillya Tepe, including the gold Achaemenid type mountain goat, the aureus of Tiberius, the Mithradates II coin, the intaglios with proiles of Athena and inscriptions in Greek, the Bactrian Aphrodite, the jades from China, the garnets from India and other artefacts illustrate all these inluences (Sarianidi, 1980). Little is known about the diferent goldsmithing techniques and choices from the period of the Great Kushan Empire (Drevnosti Tadzhikistana, 1985). he aim of this work is to undertake a irst approach to this question by studying a few items with known provenances, originating from secure archaeological contexts situated in the most important archaeological regions in Tajikistan (Fig. 1): the Beshkent Valley (Medvedskaya et al., 1975; Sedov, 1979) ArcheoSciences, revue d’archéométrie, 33, 2009, p. 177-185 Figure 1: Map of Tajikistan showing the regions of Beshkent and Dangara. Figure 1 : Carte du Tadjikistan avec la localisation des régions de Beshkent et de Dangara. and the Dangara Valley (Mandelshtam, 1968). Five items found in established excavations, representative of the Kushan jewellery and containing gold wires and granulation, were selected from the collections of the National Museum of Antiquities of Tajikistan (Vitobello et al., 2009). hree of them were found in the Beshkent Valley and are dated to a period between the 1st and the 2nd century AD: an amphora-shaped pendant and a globular openwork pendant, found in the mid 1970s during the excavation of the Beshkent cemetery V (Figs. 2a and 2b); and a cloisonné pendant found in 1978 during the excavation of the Beshkent cemetery VII (Fig. 2c). he pendant earring (Fig. 2d) was found in the late 1970s during the excavation of the Ksirov cemetery III, mortuary enclosure 19, in the Dangara Valley and is dated to a period between the 1st century BC and the 2nd century AD. Finally, an amphora-shaped earring dated to a period between the 1st and the 2nd century AD was found in 1973-1974 during the excavation of the “Ittifoq” cemetery in the Parkhar Valley (Fig. 2e). Within the context of Work Package 5 of the European project AUTHENTICO, the jewellery was transported to the C2RMF in Paris for non-destructive examination by optical microscopy, SEM, and X-radiography (for details on the complementary use of these techniques see Guerra, 2008). he composition of the jewellery was determined by PIXE (particle induced X-ray emission) at the AGLAE accelerator of the C2RMF, with a 3 MeV external proton beam of 50 µm diameter and an intensity of 30-40 nA, allowing a depth of analysis which in general is suicient to avoid the copper depletion. Simultaneous PIGE (particle induced 179 Analytical study of the manufacturing techniques of Kushan gold jewellery… a d b e c Figure 2: (See colour plate) he ive gold Kushan items selected for analysis from the collections of the National Museum of Antiquities of Tajikistan: (a) pendant 274 (length 24 mm, width 10 mm, weight 1.870 g); (b) pendant 267 (length 30 mm, width 19 mm, weight 3.780 g); (c) pendant earring 270 (length 55.9 mm, disc length 18.0 mm, width 10.5 mm, weight 2.483 g); (d) earring 271 (length 38.0 mm, width 10.0 mm, weight 3.447 g); (e) earring 274 (length 23 mm, width 14 mm, weight 1.654 g). Figure 2 : (Voir planche couleur) Les cinq objets Kouchanes en or appartenant à la collection du musée National des Antiquités du Tadjikistan, sélectionnés pour analyse : (a) pendentif 272 (longueur 24 mm, largeur 10mm, poids 1,870 g) ; (b) pendentif 267 (longueur 30mm, largeur 19 mm, poids 3,780 g ; (c) pendentif earring 270 (longueur 55,9 mm, largeur du disque 18.0 mm, width 10,5 mm, poids 2,483 g) ; (d) boucle d’oreille 271 (longueur 38,0 mm, largeur 10,0 mm, poids 3,447 g) ; (e) boucle d’oreille 274 (longueur 23 mm, largeur 14 mm, poids 1,654 g). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 177-185 180 Maria Filomena GUERRA et al. gamma-ray emission) was carried out in order to compare the compositions of the deep and surface layers. 2. TECHNOLOGICAL DESCRIPTION OF THE JEWELLERY ITEMS Earring 274 he amphora-shaped earring reference 274 in Catalogue 1983 and reference 8 of Masov et al. (2005), page 120 (Fig. 2a) is composed of gold strips of about 950 µm diameter, bent, folded and joined to produce symmetric handles holding the amphora’s body. Two discs of 130-200 µm thickness separate a bead of certainly iron pyrite (this type of decoration is common for Tillya Tepe jewellery, Collectif, 2006) and pearl elements that constitute the body. he top of a rod going through the body is threaded around the amphora’s suspension hook and is inished with a pyramid of four large granules of 1650 µm diameter. Pendant 272 Globular pendant reference 272 in Catalogue 1983, reference 6 of Masov et al. (2005), page 120 (Fig. 2b) consists of an iron pyrite bead covered with an open net birdcage of 500 µm diameter strip-twisted wires bent to form loops at the bottoms (Fig. 3). he junction points of the wires are ornate with small plain discs of 800 µm diameter. A suspension solid gold ring is joined to a strip of gold folded into a narrow cylindrical shape surrounded by a ringlet of granules of 650 µm diameter. hese elements are joined to a disc. No solder is visible and a scan carried out by PIXE did not detect the presence of an alloy with diferent contents of Cu or/and Ag; however, the morphology of the joins corresponds to the use of hard soldering. A rod with seams running longitudinally along the wire (morphologically similar to the wires from pendant MK01.40.109 of Tillya Tepe, as discussed by Ogden, 2007) goes through the bead, is coiled around the lower half of the loop, and is completed with a pyramid of four granules of 1450 µm diameter. Additional smaller pyramids formed with granules of 650 µm diameter are hard-soldered to the top of each larger granule. Pendant 267 Pendant reference 267 in Catalogue 1983 and reference 4 of Masov et al. (2005), page 120 (Fig. 2c) consists of one disc and one half-disc in cloisonné inlaid with turquoise and glass paste (identiied by visual observation), bordered with ArcheoSciences, revue d’archéométrie, 33, 2009, p. 177-185 Figure 3: he X-radiography shows that all the wires from pendant 272, as well as the sustaining rod that goes through the pendant, are hollow. he top of the rod is coiled around the suspension ring. Under the stereo microscope, the seams on the wires appear typical of strip-twisting. Figure 3 : La radiographie à rayons X montre que tous les ils du pendentif 272 sont creux ainsi que la tige de soutien qui traverse le pendentif. La tige est ensuite enroulée autour de l’anneau de suspension. La morphologie de surface des iligranes sous loupe binoculaire montre les traces typiques des ils creux tordus. Analytical study of the manufacturing techniques of Kushan gold jewellery… 181 granulated pyramids. he cloisons are 300-400 µm thick, and the encircling cloison about 2 mm large (Fig. 4). A 1500 µm rod shaped by hammering is bent as a hook and is joined to the back-plate of both elements, holding them together. An additional hook is joined to the back of the half-disc. One end of the hook was lattened while the other end was sharply cut straight. he outer borders of the setting boxes are surrounded by a row of granules of 1200 µm diameter topped by a tiny pyramid of four granules of 400 µm diameter on average (Fig. 4). he entire front part of the pendant was lattened, partially damaging a large number of granules. Pendant earring 270 Pendant earring reference 270 in Catalogue 1983 and reference 17 of Masov et al. 2005, page 124 (Fig. 2d) is typologically very similar to many objects from Tillya Tepe (Collectif, 2006). A long rod about 1000 µm in diameter, forged as a hook and serving as suspension loop and ear hook, is lattened to be joined to the central body (Fig. 5), consisting of one mango-shaped turquoise setting and one round-shaped garnet setting, both in cabochon, placed on the same back plate. Two gold strips, 2 mm large, form the box settings, which are surrounded by a row of granules about 1 mm in diameter (Fig. 5). A hammered disc is suspended with a 450 µm diameter coiled strip-twisted wire from the main body through a small ring made from a 750 µm diameter strip-twisted wire. Earring 271 Earring reference 271 in Catalogue 1983 and reference 8 of Masov et al. (2005), page 124 (Fig. 2e), is composed of a suspension hook and a complex pendant. he suspension hook consists of strip-twisted wire of uniform 750 µm circular section, each strip bent into a loop, interlocking facing opposite directions, resulting in a Gordian knot that supports a turquoise setting in cabochon surrounded by 700 µm diameter granules. hese granules are half-melted, maybe because of a poor control of the joining process. On the back side of the hook, one set of 500 µm diameter striptwisted wires (Fig. 6a) is joined on the inside of a cylindrical hinge 1000 µm large, while the other set of wires is inserted into the same cylinder, emerging for threading into the earlobe. A ring surrounded by one 850 µm diameter spooled wire is joined to the two dolphin-shaped handles decorated by chiselling and to the golden amphora’s neck. his type of wire is almost absent from Afghani gold work (Ogden, 2007). A strip-twisted wire is wound around the upper part Figure 4: Pendant 267: at the top, detail of the granules under the SEM; below, detail of the cloisonné under the stereomicroscope. Figure 4 : Pendentif 267 : en haut, détail des granules au MEB et en bas, détail du cloisonné sous loupe binoculaire. of the neck from where the amphora hangs. A coiled striptwisted wire joins the amphora to the ring, functioning as a central rod. At its bottom hangs a lower-shaped bell on 100 µm thick gold foil, decorated with 500 µm diameter twisted and spooled wires (Oddy, 1977 deinition), as illustrated in Figure 6b. Two suspension rings are joined to the external side of the dolphins’ tails, hanging a single loop-in-loop chain of 500 µm diameter strip-twisted wire; each chain is obtained by inserting four links into each other. A lowershaped bell decorated with 500 µm diameter strip-twisted wires is suspended at the bottom of a coiled strip-twisted wire (Fig. 6c). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 177-185 182 Maria Filomena GUERRA et al. Figure 5: SEM images showing details of earring 270: on the left, the suspension ring, the coiled wire, and the back plate; on the right, the settings box and the granulation. Figure 5 : Les images MEB montrent diférents détails de la boucle d’oreille 270 : à gauche l’anneau de suspension, les ils torsadés et la plaque de soutien ; à droite la boîte et la granulation. a b b Figure 6: Earring 271: (a) the X-radiography shows that the wires are hollow; (b) and (c) the diferent wires used in the decoration under the SEM. Figure 6 : Pendant d’oreille 271 : (a) la radiographie X montre que les ils sont creux ; (b) et (c) au MEB les diférents ils utilisés dans les motifs de décoration. 3. THE GOLD ALLOYS he composition of the diferent parts of each jewellery item was determined by PIXE at the AGLAE accelerator of the C2RMF, with a 3 MeV external proton beam of 50 µm diameter and an intensity of 30-40 nA. he acquisition times ranged from 3 to 5 minutes. For the bigger parts, a region of 200 x 200 µm was scanned for homogenization. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 177-185 183 Analytical study of the manufacturing techniques of Kushan gold jewellery… Two Si(Li) detectors were used to collect the X-rays emitted by the sample. One of the detectors is dedicated to the measurement of major elements, while the second detector, equipped with a 75 µm Cu ilter to selectively decrease the intensity of the gold L-lines, can be used to identify minor and trace elements when higher acquisition times are used (Guerra, 2004; Guerra and Calligaro, 2004). he compositions of the diferent parts of the items are presented in Table 1. he limits of detection are on average 40 ppm for Cu, 300 ppm for Ag and 400 ppm for Au. he alloys used to fabricate pendant 267 have the following average composition: 82.0±1.2% Au, 14.7±1.0% Ag, 3.1±1.1% Cu. hese alloys are of better quality than those used in the production of the other four objects, which have the following average composition: 75.2±2.5% Au, 22.7±2.6% Ag, 1.9±0.8% Cu. We must however note the slightly higher copper contents of some decorative elements of the disc of pendant 267 as compared to the elements of the half-disc. he compositions of pendant 272 and earring 271 are very similar; the average composition is the following: 76.3±2.1% Au, 21.8±2.3% Ag, 1.7±0.8% Cu. However, the granules of the dome plaque of pendant 272 have higher copper contents than the other elements of this item, and some elements have lower silver contents, such as the granules and the junction disc. he back plaques of earring 271 also present higher copper contents than the other elements of this item. Very few publications consider scientiic analyses carried out on Kushan gold objects. However, the analysis of a few Kushan coins, of a gold pendant with garnets, and of a gold coin struck by Kanishka (Sachs et al., 2003) showed the use of gold alloys with compositions which are rather far from our results. he results obtained for the gold alloys used in the fabrication of the Kushan jewellery pieces analysed in this work are also distinct from the compositions found for several gold objects and decoration gold foils from a Xiongnu necropolis, which are separated in two main groups, with silver and copper contents of respectively 2% and < 1% for one group and ~10% and 1-2% for the other group (Guerra et al., 2005, Guerra and Calligaro, 2003). However, our results are very similar to the composition of the alloys used under Vasudeva II (after circa 261 AD) to strike concave dinars (Fussman, 2003). As previously noted, the pendant earring 270 is typologically very similar to several objects from Tillya Tepe. A very small set of objects from this necropolis was analysed at the C2RMF, and the results were published by Calligaro (2006). In contrast to the Kushan jewellery analysed in this work, all the analysed objects from Tillya Tepe are of very good Cu% Ag % Au % 2.8 2.7 1.8 2.2 2.4 2.1 21.7 20.6 22.6 21.3 21.0 21.7 75.4 76.5 75.5 75.7 76.5 76.2 1.9 1.1 1.1 4.0 4.0 1.2 1.1 1.1 1.6 1.8 1.4 1.4 1.5 2.0 1.8 22.3 23.2 25.0 19.0 18.9 22.3 23.1 22.0 19.2 18.7 23.8 23.5 19.2 18.6 26.2 75.7 75.4 73.6 76.7 76.8 76.4 75.7 76.9 79.1 79.4 74.7 74.9 79.1 79.3 71.9 4.7 3.6 5.9 3.0 2.2 3.2 2.7 3.0 2.5 2.3 2.7 2.5 2.0 14.7 16.1 13.1 16.1 15.8 15.8 14.3 13.9 13.6 15.0 14.2 13.5 15.3 80.6 80.2 80.9 80.6 82.0 81.0 82.6 82.9 83.4 82.5 83.0 83.8 82.4 1.6 1.5 1.1 2.1 2.1 2.8 2.8 2.7 2.3 2.4 27.2 27.7 29.1 24.6 24.7 23.5 20.9 25.2 24.7 25.4 70.8 70.3 69.1 73.1 73.0 73.4 75.3 72.0 72.6 72.1 2.7 1.3 0.9 2.6 1.5 1.0 1.4 1.3 20.2 22.2 26.2 22.8 21.0 21.8 20.2 22.5 76.9 76.4 72.3 74.4 77.3 76.8 78.2 76.1 Earring 274 strip central rod cylinder granule Pendant 272 suspension hoop dome plaque dome granule wire junction disc pyramid big granule pyramid small granule ring Pendant 267 disc front plate disc back plate disc big granule disc small granule disc cloisonné half-disc cup cloisonné half-disc cup external strip half-disc cup back plate Hook Earring 274 disc hoop suspension wire mango box garnet box wire garnet box garnet box granule Earring 271 right pendant plaque right pendant decoration wire right pendant suspension wire left pendant plaque left pendant decoration wire left pendant suspension wire suspension chain Table 1: PIXE results obtained for the composition of the gold alloys used in the Kushan jewellery analysed (in %). Tableau 1 : Résultats obtenus par PIXE pour la composition des alliages utilisés dans la fabrication de l’orfèvrerie kouchane (en %). quality. Among those objects, we draw attention to the pendant MK04.40.109, found in tomb V, and a gold appliqué MK04.40.113, found in tomb II (Collectif, 2006), dated to the 1st century AD, which contain on average 96% Au, 3% Ag and 1% Cu, and 97% Au, 1.8% Ag and 1.1% Cu, respectively. An equivalent composition was obtained for the ingot dated from the 4th-2nd century BC found in the Palace of Al Khaouïn, a town founded by Alexander the Great and situated close to the present day Tajikistan border. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 177-185 184 4. CONCLUSION he analytical study of ive jewellery items containing gold wires and granules from the collections of the National Museum of Antiquities of Tajikistan provides a irst approach to the goldsmiths’ technological choices during the Kushan period. he non-destructive examination of the jewellery revealed the use of a large variety of techniques and decoration elements, sometimes very similar to those observed for the Afghani gold work of Tillya Tepe, such as the strip-twisted wires with seams running longitudinally along the wire, the absence of complex granulation patterns, the stone settings, the hammered elements, and so on. he non-destructive analysis of the objects by PIXE revealed the use of two distinct gold alloys, which contain on average about 15% Ag and 3% Cu for one type of alloy (pendant 267) and 19-26% Ag for the other (the other four objects), with copper contents that are either around 1% or around 2%. Data on particular elements, such as wires and granules, cannot be generalized, but, if we exclude pendant 272, we note that these elements have similar compositions to the other elements of one and the same item. Our elemental results are distinct from the compositions obtained by several authors for a few items and an ingot from Tillya Tepe, for several items produced by the Xiongnu, for one Kushan pendant, and for most of the coins struck by the Kushans. Only the concave dinars struck under Vasudeva II show a composition similar to our objects. We must note that the results obtained for these ive objects cannot be generalized for Kushan jewellery, with technological traditions and skills that were inluenced by Chinese, Iranian, Greek and Indian craftsmen. his combination of diferent traditions certainly also led to the use of many diferent alloys. In addition to this aspect, we must also bear in mind the diversity of the supplies in a region that beneited from contact with the trade routes between Orient and Occident. References BUNKER, E.C., 1993. Gold in the Ancient Chinese World: A Cultural Puzzle. Asiae 53(1/2): 27-50. CALLIGARO, T., 2006. Analyse des matériaux, in Collectif (eds.), Afghanistan, les trésors retrouvés: collections du musée national de Kaboul. Paris, Editions de la Réunion des musées nationaux, 292-293. CATALOGUE, 1983. Shedevry Drevnego Iskusstva i Kul’tury Tadzhikistana. Katalog vystavki (Art Treasures of Ancient Tajikistan. An exhibition ArcheoSciences, revue d’archéométrie, 33, 2009, p. 177-185 Maria Filomena GUERRA et al. Catalogue). Materiali Yuzhno-Tajikistanskoi Arkheologicheskoi Ekspedichii. COLLECTIF (ed.), 2006. Afghanistan, les trésors retrouvés: collections du musée national de Kaboul. Paris, Editions de la Réunion des musées nationaux. DANI, A.H., LITVINSKI, B.A. and ZAMI SAFI, M.H., 1996. Eastern Kushans, Kidarites in Ghandara and Khasmi, and later Hephtalites, in B.A. Litvinski (ed.), History of civilizations of Central Asia Vol. 3, he crossroads of civilization: A.D. 250-750. Paris, Unesco, 163-184. DREVNOSTI TADZHIKISTANA, 1985. Katalog vystavki (he antiquities of Tajikistan. An exhibit catalogue). Dushanbe, Tajikistan, Donish. FUSSMAN, G., 2003. L’empire Kouchan, in Musée de Lattes (ed.), De l’Indus à l’Oxus: Archéologie de l’Asie centrale. Catalogue de l’exposition. Exposition réalisée avec le concours exceptionnel de la bibliothèque nationale de France, département des monnaies, médailles et antiques. Lattes, Association Imago-Musée de Lattes, 171-210. GUERRA, M.F., 2004. Fingerprinting ancient gold with proton beams of diferent energy. Nuclear Instruments and Methods in Physics Research B 226: 185-198. GUERRA, M.F., 2008. An overview on the ancient goldsmith’s skill and the circulation of gold in the past: the role of X-ray based techniques. X-ray Spectrometry 37(4): 317-327. GUERRA, M.F. and CALLIGARO, T., 2003. Analyse des éléments en or du site de Gol Mod, in J.-P. Desroches et al. (eds.), Mongolie le premier empire des steppes. Arles, Actes Sud/Grimaldi Forum/ Mission Archéologique française en Mongolie, 176-181. GUERRA, M.F. and CALLIGARO, T., 2004. Gold traces to trace gold. Journal of Archaeological Sciences 31: 1199-1208. GUERRA, M. F., CALLIGARO, T., DRAN, J.-C., MOULHERAT, C. and SALOMON, J., 2005. Development of a PIXE, PIGE and PIXEXRF combination for the analysis of the gold from the First Empire of the Steppes. Geoarchaeological and Bioarchaeological Studies 3: 343-346. MANDELSHTAM, A.M., 1968. Pamyatniki epokci bronzi v yujnom Tadjikistane. Leningrad, Nauka. MASOV, R., BOBOMULLOEV, S. and BUBNOVA, M. (eds.), 2005. National Museum of Antiquities of Tajikistan, Institute of History, Archaeology and Ethnography of the Academy of Sciences, National Museum of Antiquities. Dushanbe, Tajikistan, ActedBactria Cultural Center. MEDVEDSKAYA, I.N., RASKOPKI, V. and SHAARTUZE, V., 1975. Archaeologicheskie Raboty v Tajikistane: 15, 324-335. MUKHAMEDJANOV, A.R., 1994. Economy and social system in central Asia in the Kushan age. in J. Harmatta, B.N. Puri, G.F. Etemadi (eds.), History of civilizations of Central Asia Vol. 2, he development of sedentary and nomadic civilizations, 700 B.C. to A.D. 250. Paris, Unesco, 265-290. Analytical study of the manufacturing techniques of Kushan gold jewellery… OGDEN, J., 2007. Recovered Treasures from Afghanistan. American Journal of Archaeology 111(3): 559-563. ODDY, A., 1977. he production of gold wire in Antiquity. Handmaking methods before the introduction of the draw-plate. Gold Bulletin 3(10): 79-87. PURI, B.N., 1999. he Kushans, in J. Harmatta, B.N. Puri, G.F. Etemadi (eds.), History of civilizations of Central Asia Vol. 2, he development of sedentary and nomadic civilizations, 700 B.C. to A.D. 250. Paris, Unesco, 247-264 RAPIN, C., 2007. Nomads and the shaping of central Asia: from the Early Iron Age to the Kushan period. Proceedings of the British Academy 133: 29-72. SACHS, C., BLET-LEMARQUAND, M. and GRATUZE, B., 2003. Médaillon avec monnaie Koushane sertie, in De l’Indus à l’Oxus: Archéologie de l’Asie centrale. Catalogue de l’exposition. Exposition réalisée avec le concours exceptionnel de la bibliothèque 185 nationale de France, département des monnaies, médailles et antiques. Lattes, Association Imago-Musée de Lattes, 189-190. SARIANIDI, V., 1980. he Treasure of Golden Hill. American Journal of Archaeology 84(2): 125-131. SARKHOSH CURTIS, V. and STEWART, S., 2007. he age of the Parthians, he Idea of Iran Vol. 2. London, Middle East Institute. SEDOV, A.V., 1979. Raskopki kurganov i poselenii v Beshkentskoi doline. Archeologicheskie Otkrytiya 1978: 584-585. THORLEY, J., 1971. he Silk Trade between China and the Roman Empire at Its Height, ‘Circa’ A.D. 90-130. Greece & Rome II 18(1): 71-80. V ITOBELLO , M.L., GUERRA , M.F., MIRSAIDOV, L.U. and BOBOMULLOEV, S.G., 2009. Scientiic Characterization of Tajikistan Golden Artefacts. Dushanbe, Tajikistan, Delta Print. WOOD, F., 2002. he Silk Road: Two housand Years in the Heart of Asia. Berkeley, University of California Press. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 177-185 Gold technology of the ancient Scythians – gold from the kurgan Arzhan 2, Tuva La technologie de l’or chez les Scythes anciens – l’or du kurgan Arjan 2, Touva Barbara Armbruster* Abstract: In 2000-2003, a ‘royal’ burial site in Tuva Republic, Siberia, dating to early Scythian times, was excavated by a joint German-Russian research program. housands of gold articles were discovered in the wooden chamber of ‘royal’ burial no. 5, dated to the late 7th century BC. he majority of these ornaments are decorated in the Scythian animal style. he study of the technological aspects reveals the variety of manufacturing techniques and designs employed for these extremely rich ornaments, and provides an insight into the complexity of the art of the early Scythian goldsmiths at the end of the 7th century BC. his paper aims to present the large variety of distinct processes used in the manufacture of these gold artefacts. he technical expertise and the equipment at the disposal of the Early Iron Age craftsmen can be inferred from tool marks and analysis of surface structures. he early date of these inds, demonstrated by a combination of radiocarbon and dendrochronology, supports the assumption that this particular, sophisticated style of early Scythian metal work originated in the Tuva region. Résumé : Une sépulture princière, datée du début de la période Scythe, a été fouillée en 2002-2003 dans la République de Touva, Sibérie, dans le cadre d’un programme de recherche germano-Russe. Des milliers d’objets en or ont été découverts dans la chambre funéraire de la sépulture n° 5, datant du VIIe siècle av. J.-C. La majorité de ces parures est décorée dans le style animalier Scythe. L’étude technologique a révélé la variété des techniques de fabrication et des motifs employés pour ces ornements extrêmement riches. Elle a aussi donné un aperçu de la complexité de l’art des orfèvres Scythes à la in du VIIe siècle av. J.-C. Cet article vise à présenter la grande variété des procédés distincts utilisés dans la fabrication de ces objets en or. La compétence technique et l’équipement à la disposition de l’artisan à l’Âge du Fer ancien peuvent être déduits des traces d’outils et de l’analyse des structures de surface. La précocité de ces trouvailles, démontrée par la combinaison de la datation radiocarbone avec la dendrochronologie, soutient l’hypothèse que ce style particulier et sophistiqué du travail du métal au début de l’époque Scythe trouverait son origine dans la région de Touva. Keywords: Scythians, goldwork, Arzhan 2, manufacture, tool marks. Mots-clés : Scythes, orfèvrerie, Arjan 2, fabrication, traces d’outils. 1. INTRODUCTION his paper deals with early Scythian precious metal technology, more precisely with the manufacture of the gold jewellery and ornamented weapons buried in the kurgan (burial mound) of Arzhan 2 in the north of the Republic of Tuva, in Siberia. One kurgan of this important Scythian royal necropolis was fully excavated from 2000 to 2003 as part of a German-Russian research program, headed by Hermann Parzinger and Anatoli Nagler (German Archaeological Institute, Berlin), as well as by Konstantin Chugunov (State Hermitage, Saint Petersburg) (Chugunov * TRACES – UMR 5608 du CNRS – Maison de la Recherche, Université de Toulouse Le Mirail, 31058 Toulouse cedex. (barbara.armbruster@univtlse2.fr) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 187-193 188 et al., 2003; Menghin and Parzinger, 2007). Arzhan, in the district of Turan, is located in a plain where hundreds of kurgans are arranged in alignments. he kurgan Arzhan 2, which is nearly 80 m in diameter and more than 2 m in height, consisted of a platform constructed from stone slabs, as it is characteristic of the early Scythian period. he wooden chamber of the burial no. 5 was dated by radiocarbon and dendrochronology to the 2nd half of the 7th century BC. During the excavations of 2001, more than 5700 gold items were discovered in the undisturbed royal burial no. 5 (Chugunov, 2004; Chugunov et al., 2006). he results of current research on Arzhan 2 are forthcoming (Chugunov et al., 2010). he two individuals in the prestigious burial no. 5, a man and a woman, were literary covered with gold. hey were adorned with personal ornaments, such as torcs, necklaces, ear ornaments, pendants and pins. Gold also embellished the head-dress and boots, while a large number of appliqués covered the dress. he man’s upper garment, clearly a cape, was adorned by about 2500 small cast panther igures, while the woman’s cape was covered by much lighter, but aesthetically identical, panther igures worked in pressed sheet. In addition to gold beads, beads of diferent materials, such as carnelian, garnet, glass paste, malachite, and turquoise, are also present in the ensemble of personal ornaments. Other gold products, such as a miniature cup, a quiver and its carrying belt, as well as a wooden bowl with its handle covered with gold, are also part of this extraordinary discovery. Finally, iron weapons decorated with gold and silver inlay, such as daggers, knives, arrow-heads and a battle-axe form yet another category of these precious grave goods. he majority are decorated in the Scythian animal style. A general problem related to the scientiic study of early Scythian gold lies in the fact that precious metal artefacts are not known from hoards or settlement sites, but exclusively from funerary contexts, or as isolated objects, presumably from looted tombs. Scythian gold work from Siberia is particularly well known due to the Siberian collection of Peter the Great, on display in the State Hermitage Museum, St Petersburg (Schiltz, 1994; Popescu et al., 2001). International exhibitions showed some Scythian gold ornaments from Siberia dating to the 7th and 6th centuries BC, such as the splendid rolled panther from Kélermes, or other animal igures, like the deer from Kostromslaïa (Busch, 1993; Popescu, 2001). However, the undisturbed tomb from Arzhan 2 represents the most luxuriant Siberian grave of the early Scythian period with regards to the number and combination of artefacts, as well as to the variety of diferent styles present, allowing one to distinguish diferent workshop traditions. he grave goods provide evidence of ArcheoSciences, revue d’archéométrie, 33, 2009, p. 187-193 Barbara ARMBRUSTER the high quality of the gold work in both aesthetical and technical terms (Armbruster, 2010). A signiicant part of the gold objects bear traces of wear, proving that they were used. However, there are also items without any traces of wear, which still maintain rough edges that make them unsuitable for use. hey were most probably especially made for the burial and its ceremony, as is known in the case of the gold work from the early Celtic princely grave from EberdingenHochdorf, Ludwigsburg, Germany (Biel, 1985). 2. WORKSHOP AND TECHNICAL KNOWLEDGE For the 7th century BC, there are no written or iconographic sources available to inform us about the technical skill of the Scythian goldsmiths. hat is why our knowledge of techniques is based above all on the study of archaeological artefacts with the help of analogies, experiments and material sciences. Tool marks and surface textures on the gold work from Arzhan 2 testify to the advanced technological knowledge of the goldsmiths and their well equipped workshops. However, no evidence of any early Scythian goldsmith workshop has ever been found. hey must have consisted of a set of specialised tools and furnaces fanned with bellows in order to achieve a temperature of about 1000 °C, necessary for the melting of gold with a low silver or copper content. he furnaces must have been equipped with adequate tuyères and controlled air draft. Analyses of tool marks conirmed the following workshop equipment: A pair of scales and weights are assumed to have been present for measuring. For casting, annealing and soldering, a furnace with bellows, charcoal, crucible, casting mould (clay or stone), and tongs (wood or iron) were needed, and a blow-pipe was probably also used. Plastic shaping was executed with hammers and anvils (stone or iron), and with dies (bronze or hard wood). he bending of wire, particularly of iligree elements, involves some sort of pliers. Decoration techniques required punches, scrapers, engraving tools, and chisels (hard iron or steel). Finishing could have been carried out with grinding stones, sand, ashes and acids. Ethnographic analogies demonstrate that in the nomadic lifestyle these tools can be transported in a sack or chest over long distances (Armbruster, 1995). Before manufacturing an object, the goldsmith irst developed a concept. In the case of the complex Scythian animal style, he irst had to sketch the outline of the desired motif. Gold objects from Arzhan bear tool marks resulting from outlines. Any gold work starts from a cast ingot. Gold must be melted in a crucible and cast in a mould. For the gold from Tuva, two methods are evident. Ingot casting was used Gold technology of the ancient Scythians – gold from the Kurgan Arzhan 2, Tuva for preliminary products; then, the cast ingot was transformed by plastic shaping techniques in order to obtain wire or sheet. A large number of the Arzhan gold items consist of an assemblage of sheet and wire work. Casting However, lost wax casting was used for massive objects decorated with the complex relief motifs. Examples of such cast ornamental objects are the man’s neck-ring and the woman’s pectoral, pins, clasps, and waist buckles (Fig. 1), and various belt and strap elements. In addition to the animal igures that are worked fully in three dimensions (e.g. big cats, ibexes, deer, and eagles), other zoomorphic representations are realised individually in bas-relief, or arranged as animal groups integrated in the metal surface, occupying it entirely (e.g. boars, camels, big cats, deer, horses). Flat igures cut out of thick gold sheet are another type of animal art occurring in Arzhan 2 (e.g. deer, horse, and ibex). he Scythian animal style appears in both a naturalistic version and a more or less stylised one. Sheet work Sheet metal and wire were produced through plastic shaping with a hammer and anvil. Repeated annealing prevented the gold from getting hard and brittle. he animal style sheet ornaments were executed by cutting the silhouette of the desired object using a chisel. Holes were formed with the help of a conical instrument. Chisels and engraving implements were used to remove metal. Sheet metal and wire were joined to create jewellery by soldering, as, for instance, in the case of complex ear-ornaments. 189 Engraving and openwork Several decorative techniques were applied in the early Scythian gold work from Arzhan in addition to the predominant method employed for ornaments cast in relief. Less complicated decorations were executed by engraving and cut in open work (Fig. 3). Engraving is a cutting technique that requires hardened steel tools to remove metal chips. Such hardened tools were needed for engraving certain gold decorations, as well as for preparing the grooves for precious metal inlay in the iron surfaces. Although no steel tools are known from that period, small steel scrapers were most probably used. Inlay was used to embellish the iron artefacts. Pressed sheet technique for serial production Chasing and pressed sheet are plastic deformation techniques also employed for decoration. he tools used for chasing are punches, chisels, and an elastic working surface, such as a chasing cement or thick leather. Only a single object from Arzhan 2, a panther igure, is deinitely chased. In contrast, thousands of small sheet panthers were realised by the pressed sheet technique using dies (Fig. 2b) (Armbruster, 2003). he sheet’s relief is pushed with a pointed tool while the gold metal is positioned over the die. he use of dies serves for the manufacture of a series of pieces with little weight, but much optical efect. he small panther igures of the garment of the princess are examples of the pressed sheet technique. Another technique of pressed relief on sheet decoration is also present, using the carved relief of a wooden itting as a die and producing single objects. he unique ornamental quiver and the handle of a wooden bowl a b Figure 1: (See colour plate) Lost wax casting with ‘kerbschnitt’ relief, waist buckle. Figure 1 : (Voir planche couleur) Fonte à la cire perdue avec relief ‘kerbschnitt’, boucle de ceinture. Figure 2: (See colour plate) Panther igures serial production; a: cast in lost wax; b: pressed sheet technique. Figure 2 : (Voir planche couleur) Figures de panthères fabriquées en série ; a) fonte à la cire perdue ; b) technique de pression sur tôle. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 187-193 190 Barbara ARMBRUSTER Figure 4: (See colour plate) Sheet work, decorated with iligree, granulation and enamel: ear plugs and pendants with chain. Figure 4 : (Voir planche couleur) Travail de la tôle, décorée avec du iligrane, de la granulation et de l’émail : pendants d’oreille et pendentifs avec chaînes. Figure 3: (See colour plate) Flat sheet ornaments, with cut contours and decorated with engraving and open work: horse igure from the woman’s head ornament. Figure 3 : (Voir planche couleur) Ornements en tôle plane, avec des contours découpés, décoration gravée et ajourée : cheval de l’ornement de tête de la femme. Helicoidal tool marks of this technique are clearly present on gold wire decorating the sheet animal igures, boots, decorative bands, and the headgear. Fine iligree wires were ixed on the metal support by soldering with metallic solder, as shown by clearly visible traces of solder. Granulation and iligree are decorative techniques based on the application of granules or ine wires on a metal sur- bear a motif of ish scales. hese scales are in relief, which was transferred to the golden sheet by the pressed sheet technique. he wood was left as a backing, strengthening the gold sheet element. he quiver, composed of the wooden itting, the ornamental gold sheet, a leather recipient and strap, had to withstand the weight of bow and arrows. Filigree and granulation (Fig. 4) Wire of up to a diameter of one millimetre in thickness was produced by manual hammering. Small rings made of wire served for the production of chains made of simple loops or with the loop-in-loop technique (Fig. 5) (Reist Stark and Reist Smith, 2000). Very ine wires for iligree and the outline of the enamel panel were produced by coiling a narrow gold strip, a procedure well known from ancient wire, the so-called ‘strip-twisting’ (Formigli, 1993). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 187-193 Figure 5: Loop-in-loop chain: miniature cauldron with chain. Figure 5 : Chaîne en loop-in-loop: chaudron en miniature avec chaîne. Gold technology of the ancient Scythians – gold from the Kurgan Arzhan 2, Tuva face attached by soldering (Wolters, 1986; 1987). Both techniques of decoration were also applied in combination with enamel. Enamel was already well developed in early Scythian art. Metallic solder was used for hot joining. his solder is a gold alloy whose melting point is lower than that of the base metal. Small particles were placed on the surface to be joined and then melted. Remnants of solder are clearly recognizable on the gold work. In addition to soldering, riveting and folding were also used as joining techniques in the gold work from Arzhan 2. Ear ornaments (Fig. 4), decoration bands from the boots, zoomorphic sheet igures, and beads from Arzhan are outstanding examples of enamelled iligree and granulation work. Enamel is used here as a pale whitish glass on gold sheet, applied on the metal surface as a crushed powder and then melted in the furnace. Metal inlay is a polychrome decorative technique present in Arzhan 2 on iron weapons ornamented with gold and silver (Fig. 6). Small precious metal elements were incrusted in the iron surface. hese small sheet or wire pieces were hammered and rubbed into grooves executed in advance. For this type of work, hardened steel implements were needed to engrave or cut the grooves and hollows. he spectacular iron weapons of the prince of Arzhan were already highly complex, three-dimensional and ornamental works of art created by a blacksmith, before the ine metalworker began to add the gold and silver inlay. 191 3. WORKSHOP GROUPS Diferent technical workshop groups from Arzhan 2 can be distinguished. One group consists of cast objects, massif and hollow, with ‘kerbschnitt’ relief, made by lost wax casting (Figs. 1 and 2a). Another category consists of lat sheet ornaments, with cut contours and decorated with engraving and openwork (Fig. 3). Sheet working was also identiied on three dimensional, hollow sheet ornaments worked with dies or by chasing. Flat sheet work, decorated with iligree, granulation and enamel represents yet another group (Fig. 4). Serial production of identical articles is evident in numerous objects cast in lost wax, and others worked in pressed sheet (Fig. 2). he next group contains wire work, such as simple loop and loop-in-loop chains (Fig. 5). Finally, gold and silver inlay in iron weapons constitutes the last group in the assemblage of ine metal work (Fig. 6). he diferent ine metal working techniques used in the manufacture of the gold and iron artefacts from Arzhan 2 were not new inventions. hey were previously known in other regions, such as the Near East, as well as China, before the 7th century BC, while lost wax casting, soldering, iligree and granulation, and also polychrome metal inlay, were also in use in Anatolia and Mesopotamia, as well as in Iron Age Europe. However, this discovery represents the irst instance of the application of such a wide range of metal working techniques in the production of objects in the Scythian animal style. he gold technology in Arzhan 2 can be summed up according to technical categories as follows: Casting techniques: ◊ ingot casts for sheet and wire production ◊ lost wax casting, massif and hollow single pieces and series Plastic shaping techniques: ◊ lat hammered sheet; chased sheet with repoussé ◊ pressed sheet, serial production with dies ◊ wire production, chains Joining techniques: ◊ soldering, riveting, folding Figure 6: (See colour plate) Gold inlay in iron weapons: detail of the man’s dagger. Figure 6 : (Voir planche couleur) Incrustations d’or dans des armes en fer: détail du poignard de l’homme. Decorative techniques: ◊ cast relief, inished by scraping and chiselling ◊ cut decoration: engraving, chiselling, open work ◊ application of small elements: granulation and iligree (solder) ArcheoSciences, revue d’archéométrie, 33, 2009, p. 187-193 Barbara ARMBRUSTER 192 ◊ enamel (colour spots) ◊ inlay: iron objects with precious metal inlay (polychrome) Finishing techniques: ◊ grinding and polishing ◊ pickling with acids 4. CONCLUDING REMARKS he exceptional discovery of the intact elite burial of Arzhan 2 provides new insights into the complex gold work of the early Scythians in Siberia. he gold ornaments were worked both for the living and for the dead, some exclusively for ceremonial rituals. hey testify to the high technical and artistic level of the complex Scythian goldsmiths’ craft in the late 7th century BC. he gold work provides evidence of the organization and specialization of applied arts and of extended cultural contacts. he early date of the 2nd half of the 7th century BC proves that the gold work from Arzhan 2 is older than parallels from Kazakhstan, and represents the irst evidence of that technology in Scythian Siberia. he particularly rich combination of techniques applied to speciic designs seems to be a local feature. However, the sophisticated gold technology, including lost wax casting, soldering, iligree, granulation, and enamel was known prior to the 7th century BC, both in the East and West. he brief account of the virtuoso and diverse gold work of the early Scythians reveals a sophisticated knowledge of the metal. he tool marks and techniques identiied on the artefacts from Arzhan 2 provide evidence of the equipment of the goldsmith’s workshop. From ethnological analogies, it is known that the tools of a mobile atelier could easily be transported in a sack over long distances. his observation corresponds to the nomadic lifestyle of the Scythians. Four diferent craft traditions were identiied among the objects under consideration: cast products with ‘kerbschnitt’ relief, chased and pressed sheet items, objects with granulation, iligree, and enamel, and gold and silver inlay in iron. More than 2500 years ago, early Scythian goldsmiths equipped with a few archaic tools were able to create gold ornaments of high aesthetic and technical level. Acknowledgements he author should like to express her gratitude to Hermann Parzinger for the invitation to study the gold work from Arzhan 2. Konstantin Chugunov provided every assistance during the research stay at the State Hermitage in St Petersburg, where ArcheoSciences, revue d’archéométrie, 33, 2009, p. 187-193 Svetlana Burshneva kindly supported my investigations in the research laboratories. Anatoli Nagler provided support both in St Petersburg and in Berlin for inishing my task. I am also very grateful to Niamh Whitield for polishing my English. References ARMBRUSTER, B.R., 1995. Traditionelles Goldschmiedehandwerk in Westafrika und bronzezeitliche Metallverarbeitung in Europa. Technologien im ethnoarchäologischen Vergleich. Beiträge zur Allgemeinen und Vergleichenden Archäologie 15: 111-201. ARMBRUSTER, B.R., 2003. Preßblecharbeiten. Technologisches, in H. Beck, D. Geuenich, H. Steuer (eds.), Reallexikon der Germanischen Altertumskunde 23. Berlin and New York, De Gruyter, 409-413. A RMBRUSTER, B.R., 2010. Technologische Aspekte der Goldschmiedekunst aus Arzan 2, in K.V. Chugunov, H. Parzinger, A. Nagler (eds.), Der skythenzeitlichen Fürstenkurgan Arzan 2 in Tuva. Archäologie in Eurasien 26. Berlin, 183-199. BIEL, J., 1985. Der Keltenfürst von Hochdorf. Stuttgart, heiss. BUSCH, R., 1993. Gold der Skythen. Schätze aus der Staatlichen Eremitage St. Petersburg. Veröfentlichungen des Hamburger Museums für Archäologie und die Geschichte Harburgs Helms-Museum 67. Münster, Wachholtz. CHUGUNOV, K.V., 2004. Archaeological discoveries in the Valley of the Kings, Tuva. Artifacts from the Arzhan barrow. Exhibition at the Hermitage, Sankt-Petersburg (in Russian). CHUGUNOV, K.V., PARZINGER, H. and NAGLER A., 2003. Der skythische Fürstengrabhügel von Aržan 2 in Tuva. Vorbericht der russisch-deutschen Ausgrabungen 2000-2002. Eurasia Antiqua 9: 113-162. CHUGUNOV, K.V., PARZINGER, H. and NAGLER A. (eds.), 2006. Der Goldschatz von Arzhan. Ein Fürstengrab der Skythenzeit in der südsibirischen Steppe. München, Schirmer/Mosel. CHUGUNOV, K.V., PARZINGER, H. and NAGLER A. (eds.), 2010. Der skythenzeitlichen Fürstenkurgan Arzan 2 in Tuva. Archäologie in Eurasien 26. Berlin. FORMIGLI, E., 1993. Sulla tecnica di costruzione dei ili d’oro nell’oreiceria etrusca, in C. Eluère (ed.), Symposium “Outils et ateliers d’orfèvre des temps anciens”. Saint-Germain-en-Laye 1991. Antiquités Nationales, mémoire 2. Saint-Germain-enLaye: Société des Amis du Musée des Antiquités Nationales et du château de Saint-Germain-en Laye, 35-38. MENGHIN, W. and PARZINGER, H. (EDS.), 2007. Im Zeichen des goldenen Greifen. München, Berlin, London, New York, Prestel. POPESCU, G.A., 2001. La pantera e il cervo, in G.A. Popescu, A. Alekseev, J. Piotrovskij (eds.), Siberia. Gli uomini dei iumi Gold technology of the ancient Scythians – gold from the Kurgan Arzhan 2, Tuva ghiacciati. Museo Statale dell’ Ermitage, San Pietroburgo. Milano, Electa, 67-77. POPESCU, G.A., ALEKSEEV, A. and PIOTROVSKIJ, J. (eds.), 2001. Siberia. Gli uomini dei iumi ghiacciati. Museo Statale dell’Ermitage, San Pietroburgo. Milano, Electa. REIST STARK, J. and REIST SMITH, J. Classical loop-in-loop chains and their derivatives. London, A. & C. Black. SCHILTZ, V., 1994. Les Scythes et les nomades des steppes, VIIIe siècle avant J.-C.-Ier siècle après J.-C. Paris, Gallimard. 193 WOLTERS, J., 1986. Die Granulation. Geschichte und Technik einer alten Goldschmiedekunst. München: Callwey. WOLTERS, J., 1987. Filigran, in K.-A. Wirth (ed.), Reallexikon zur Deutschen Kunstgeschichte 8. München, Beck, 1062-1184. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 187-193 Funeral golden mask and hand with a ring he necropolis of Trebeniste Le masque funéraire en or et la main baguée – la nécropole de Trebeniste Pavlina Ilieva* and Petia Penkova** Abstract: his study reviews the historical background of the Necropolis of Trebeniste while also highlighting some of the main hypotheses that help date this site and explain its signiicance. Furthermore, it provides a description of the tombs and focuses on two of the golden objects excavated from the graves – a funeral golden mask and a hand with a ring. It lays out the results of an analysis of the crafting techniques used – one that was conducted at a laboratory of the National Archaeological Institute with Museum under the Bulgarian Academy of Sciences in Soia, Bulgaria. his analysis aims to contribute to the further investigation and argumentation pertaining to the dating of these indings. It also hopes to suggest potential new areas of research. Résumé : Cette étude fait un bilan du contexte historique de la Nécropole de Trebeniste et souligne quelques unes des hypothèses qui contribuent à la datation de ce site et qui expliquent sa signiication. En outre, ce travail fourni une description des tombes et se concentre sur deux objets en or fouillés dans les tombeaux – un masque funéraire et une main avec une bageue. Cette étude expose brièvement les résultats d’une analyse des techniques artisanales utilisées – une qui a été menée au laboratoire de l’Institut national d’Archéologie avec Musée, de l’Académie Bulgare de Sciences, à Soia. Cette analyse a pour but de contribuer aux recherches et argumentations se rapportant à la datation de ces découvertes. Ce travail prétend aussi suggérer de nouveaux domaines potentiels de recherche. Keywords: Trebeniste, funeral mask, hand with ring, techniques. Mots-clés : Trebeniste, masque funéraire, main avec bague, techniques. 1. INTRODUCTION Ninety years ago, nearby Lake Ohrid, the seven graves of Trebeniste were excavated. he necropolis was discovered by accident in 1918 by a Bulgarian military unit mending roads in the region. he First World War hindered archaeologists Bogdan Filov and Carel Shkorpil from publishing material about these inds before 1927 (Filov and Schkorpil, 1927). In the past, the Trebeniste region was renowned for its rich silver bed – the silver mines of Damastius mentioned by Strabo were nearby. he most important Balkan roads used to intersect there as well, stretching in all four directions of the world. hree and a half kilometres west of Lake Ohrid lies a plain crossed by the Drin river. he mountains of present day Albania separate this area from the Adriatic Sea. Two rivers, Shkumbin (Genesis) and Semeni (Aspus), connect this land to the shore. * Head of Collections Department, National Archaeological Institute with Museum, Soia, Bulgarian Academy of Sciences – 2, Saborna str., Soia 1000, Bulgaria. (polly_ilieva@mail.bg) ** Head of the Laboratory of Analysis, Conservation and Restoration, National Archaeological Institute with Museum, Bulgarian Academy of Sciences – 2, Saborna str., Soia 1000, Bulgaria. (petiapenkova@yahoo.com) rec. Oct. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 195-199 196 Early Greek inluences had an impact on this area during the period of Greek colonization, when the ist historic contacts started to take place. In the 7th century BC, the Greeks started building a network of apoikoi spreading out across the Adriatic Sea and based on powerful settlements such as Korkyra (present day island of Corfu), Epidamnus (Dirahium, present day Duresu) and Apolonia (present day Pozhan). Korkyra was the irst Greek colony in the Adriatic: it was used by the Greek colonizers as a support base for penetration further north. Sea access was under the control of the colony in Korkyra: its propitious geographic location turned it into an important centre for Greek colonization. A very similar picture can be drawn regarding the shores of the Black Sea and the Aegean Sea. hese developments are usually considered evidence supporting the hypothesis that a signiicant number of items excavated from the Trebeniste necropolis were imported from Greece, while others are considered to be of hracian-barbarian origin with Greek inluences. Since a signiicant chronological gap of eight centuries exists between the end of the Mycenaean civilization and the golden days of Classical hrace, we could also accept interpretations of some items of hracian material culture from the Classic and early Hellenistic period as being ‘reminiscent of the Aegean’. here is, however, strong evidence which documents the relocation of the Aegean civilization to the north – north-east, and the Trebeniste necropolis can thus be viewed as an intermediary link between the hracian and Mycenaean civilizations. Studies of the necropolis show a prolonged period of existence – from the 7th century BC to the end of the 4th and the beginning of the 3rd century BC. he Trebeniste necropolis is dated back to a period between the 7th century BC and 300 BC. A rich funeral inventory is only found in tombs dated towards the end of the 6th century BC. his fact conirms the thesis regarding the presence of lavish ‘imported’ vessels, goblets and silver rhytons – only to be found in the tombs of ‘kings’. Written sources about hrace, Skythia, Macedonia, and Asia Minor interpret these objects as royal insignias (Popovic, 1994). Certain graves of men, women and children, dated back to the late 6th century BC, contain notably rich grave goods, treasured today in three states – Bulgaria (in Soia), Serbia (in Belgrade), and Macedonia (in Ohrid). he graves are pits with loors of trampled down earth and illings of soil mixed with big stones (Vulic, 1933). Warriors have been buried there, in ceremonial attire, with their arms, vessels and other belongings. hin golden lamellae (bratteae) in the shape of embroidery and rosettes beautiied gorgeous mantles made probably of multicoloured fabrics. Elegant silver or silver gilt pins fastened the mantles (Stibbe, 1994). Golden bands also decoArcheoSciences, revue d’archéométrie, 33, 2009, p. 195-199 Pavlina ILIEVA, Petia PENKOVA rated the helmets, cuirasses, shields and swords. A golden band has been fastened above the forehead of a helmet, or runs along the entire front of another piece. he golden lamella shaped as a crescent, and the bronze ones representing horses were once probably fastened upon the breastplate of a cuirass and the leather overall of a shield. A golden rosette also decorated the sword scabbard from grave no. 7. he funeral golden masks, a hand with a ring, and a sandal made of a thin sheet of gold, found in three graves, relect the high social status of the warriors (the graves are dated to a period around the end of the 6th century BC), and their particular role as heralds of the ideology and traditions of the society they belonged to. Beads of amber and glass, and numerous vessels made of silver, bronze, glass or clay are also found among the oferings. 2. ANALYTICAL EQUIPMENT A digital microscope by Keyence has been used for the observation of the production technology. he magniication range is 25x – 175x. The XRF instrument used is model EDX-720 of Shimadzu; Atmosphere: Air; Collimator: 1 mm. Analyte TG kV uA Acq. (keV) Anal. (keV) Ti-U Rh 50 Auto 0-40 0.00-40.00 Na-Sc Rh 15 Auto 0-20 0.00-4.40 he calculation method is Quan-FP (fundamental parameters). he detection limit was 0.01%, but varies for the diferent elements. he analyser automatically detects statistical data concerning the presence or absence of elements around the detection limit. he lack of standard samples and the fact that no preliminary treatment was carried out on the surface renders the interpretation of the results as semi-quantitative. Despite the limitations of the method, the presence or absence of an element is a valid inding. 3. RESULTS AND DISCUSSION Mask: a focus on the technology used in crafting the mask allowed us to trace the sequence and types of techniques that were involved in this process. We can infer that, initially, the craftsperson cut out a thin golden sheet – estimated to be about 0.03 mm in width and approximately the size of a human face. If we assume that the mask was crafted especially for the person buried with it, then its size was Funeral golden mask and hand with a ring. he necropolis of Trebeniste 197 most probably measured accordingly. It is also possible that it was even laid out on this person’s face in order to obtain correct measurements of the exact location of the eyes, nose and lips. he nose is depicted by an aperture – most likely to avoid a potential rupture of the thin golden sheet that a high relief can cause. he nose – crafted separately – and the entire mask were both punctured with eight holes and connected via inserting and folding of lat golden wires (Fig. 1). he following technique has been used before illing the eyes: the relief was formed using repoussé (free hand forging) and then the contours were deined by punching, through chiselling on the side of the face, where this technique left clear marks of the tool. he eyebrows were deined by means of repoussé and chiselling, but these were both crafted from the back side. he lips were crafted with a technique very similar to that used for the eyes – their relief crafted from the back side, while the contours were deined from the front. Particularly aesthetically pleasing and interesting from the standpoint of its crafting is the bee, placed on the forehead right above the nose. he contours on the side of the face are clearly formed (Fig. 2). While it is possible that this part was crafted using a technique similar to the one used for the eyes and lips, we cannot rule out the possibility that a mould or a press was used. It is clearly visible that the chiselled lines along the wings have been crafted from the back. Similarly, we cannot rule out the possibility that moulds were used in the crafting of the decorative ornaments – a meander-shaped and a two linked number-eights-shaped one. A clear pattern of repetition of every ifth element emerges. We can also infer that the mould was placed at the back of the mask. It is clear that at one point the mould was dislocated to the upper left corner, next to the linked number-eights. his is clearly suggested by the surface of the object, which was visibly lattened out and then a relief was formed once again (Fig. 3). he decorative points have been crafted from the back side of the mask. Hand: Similar to the mask, the hand was most probably also initially cut in the respective shape. Subsequently, its anatomic elements were marked of, namely the position of the phalanges and the nails. he main content of the phalange was crafted from the back side and subsequently completed from the front. he nails were crafted by means of a series of dots from the back. In places, we can see the line which served to guide the crafting of the dots by punching. At the base of the hand, Pavlina Ilieva noticed a igure which had not been described by researchers so Figure 1: Fitting of the nose with a lat gold band. Figure 1 : Ajustement du nez avec une bande d’or plate. Figure 3: Distinct decoration outlines on the outside. Figure 3 : Les décorations du contour sur le côté extérieur. Figure 2: A detail of the bee on the forehead. Figure 2 : Détail de l’abeille sur le front. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 195-199 198 Pavlina ILIEVA, Petia PENKOVA Figure 5: (See colour plate) he golden mask Inv. n° 6859. Figure 5 : (Voir planche couleur) Le masque d’or, référence d’inventaire 6859. Figure 4: (See colour plate) he igure on the hand. Figure 4 : (Voir planche couleur) La igure sur la main. far (Fig. 4). It has been crafted mainly from the back side of the hand. he decorative ornament of linked numbereights has been crafted in a way that is identical with that used for the mask. Inv n° Au % Ag % Cu % Fe % 58,9 38,2 0,7 0,8 57,9 40,4 0,7 - 69,7 27,8 - - 25,4 0,8 4,4 6859 Golden mask 6861 Golden hand 6862 Golden ring 6885 Golden appliqué 62,7 Table 1: Results of the EDX analysis of the gold alloys of the four studied objects. Tableau 1 : Résultat de l’analyse EDS des alliages d’or des quatre objets étudiés. his preliminary study is part of a more extensive project which will allow a more comprehensive investigation of the gold objects from the Necropolis of Trebeniste. Keeping in mind that the objects from this necropolis are hosted in the above-mentioned three countries, this study corresponds to ArcheoSciences, revue d’archéométrie, 33, 2009, p. 195-199 Figure 6: (See colour plate) Golden hand with ring Inv. n° 6861 and Inv. n° 6862. Figure 6 : (Voir planche couleur) La main en or avec une bague, référence d’inventaire 6861 et 6862. Funeral golden mask and hand with a ring. he necropolis of Trebeniste the presentation of preliminary results on the objects currently located in Bulgaria. he aim of this work was to describe the crafting techniques and to determine the composition of the gold alloys in order to be able to use these irst results as a basis for further study and for comparison. If a semi-quantitative analysis is not suicient for drawing inal conclusions on the gold objects from the Necropolis of Trebeniste, these irst results nevertheless open up paths for future discussion. 199 References FILOV, B.D. and SCHKORPIL, K., 1927. Die archaische Nekropole von Trebenischte am Ochridasee. Berlin and Leipzig, de Gruyter. POPOVIC, L.B., 1994. Collection of Greek Antiquities. Belgrade, National Museum Belgrade. STIBBE, C.M., 1994. Laconian drinking vessels and other open shapes. Laconian black-glazed pottery, Part II. Amsterdam, Allard Pearson Museum. VULIC, N., 1933. Neue Gräber bei Trebenischte, Archäologischer Anzeiger: 459-482. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 195-199 Les parures en or de la nécropole thrace de Duvanli du e siècle av. J.-C. : le cas du tumulus de Kukova he gold jewellery from the hracian necropolis of Duvanli (5th century B C): the case of the Kukova mound Milena Tonkova* and Petia Penkova** Résumé : L’article présente les premiers résultats du projet de création d’une base de données relative aux caractéristiques typologiques, stylistiques et technologiques des parures de l’époque classique et hellénistique de la collection du Musée National d’Archéologie de Soia. La discussion scientiique porte sur l’origine des bijoux en or de la tombe de Kukova, du tumulus du ve siècle av. J.-C. Ces objets constituent les éléments d’un ensemble original de bijoux de tête et de corps typiques de l’élite thrace également présents dans deux autres riches tombes de la nécropole de Duvanli. Ces parures sont exécutées dans un style qui renvoie à des prototypes grecs, macédoniens et thraces. Les caractéristiques technologiques des bijoux sont examinées au moyen d’analyses semi quantitatives des alliages d’or par la méthode de luorescence à rayons X et des observations sous un microscope optique digital, ain de diversiier les possibilités de recherche concernant leur origine. L’étude a fourni de nouveaux arguments à la description des groupes stylistiques formés auparavant. Abstract: he paper presents the irst results of a project aiming to create a database containing the typological, stylistic and technological characteristics of the Classical and Hellenistic jewellery kept in the collections of the National Archaeological Museum of Soia. he present study discusses the set of gold jewellery originating from a 5th century BC tomb in the Kukova mound. hese adornments are components of an original head and body decoration, a set typical for the hracian elite. his set originates from two other graves of the Duvanli necropolis. he stylistic analysis points toward Greek, Macedonian and hracian prototypes. he technological characteristics of the pieces of jewellery have been studied through a semi-quantitative analysis of the gold alloy by X-ray luorescence. Observations of the artefacts under the microscope have also been carried out in order to diversify the methodology employed for the research of the genesis of Duvanli jewellery. he study added new arguments to the already existing description of the respective stylistic groups. Mots-clés : analyses technologiques, nécropole de Duvanli, Odryses, or, orfèvrerie, hrace. Keywords: Duvanli necropolis, gold, jewellery, Odrysians, technological analysis hrace. 1. INTRODUCTION En 2008 a été lancé un projet de création d’une base de données pour les caractéristiques typologiques, stylistiques et technologiques des pièces d’orfèvrerie de l’époque classique et hellénistique appartenant au Musée National d’Archéologie de Soia. Cet article présente les premiers résultats de ce projet. * Chargé de recherches, Institut et musée national d’Archéologie, Académie des Sciences Bulgare – 2, rue de Saborna, Soia 1000, Bulgarie. (milenatonkova@hotmail.com) ** Responsable du Laboratoire d’analyses, de conservation et de restauration, Institut et musée national d’Archéologie, Académie des Sciences Bulgare – 2, rue de Saborna, Soia 1000, Bulgarie. (petiapenkova@yahoo.com) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 201-209 202 Milena TONKOVA, Petia PENKOVA La discussion scientiique prend comme exemple le plus prometteur des bijoux en or du tumulus de Kukova, une des tombes exceptionnellement riches de la nécropole classique bien connue de Duvanli, région de Plovdiv. Il s’agit d’un ensemble composé de plusieurs bijoux de types diférents, dont le poids total est de 1265,35 grammes. Il est fort probable que lors de leur découverte les bijoux étaient encore plus nombreux, étant donné que la trouvaille n’est pas complète. Le tombeau a été fouillé en 1925 par des paysans et l’inventaire a été conisqué à plusieurs reprises (Filov, 1934: 19-39). Cela a déterminé le sort ultérieur des trouvailles qui ont été partagées entre deux musées archéologiques, ceux de Soia et de Plovdiv. La plupart des parures ont été attribuées au Musée National d’Archéologie de Soia. C’est exactement celles-ci qui font objet de l’actuelle étude. La trouvaille comprend huit boucles d’oreilles en forme d’anneau ouvert et deux pendants d’oreilles en forme d’oméga, dont les bouts se terminent en pyramides ornées de granules (sept sont représentés sur la Fig. 1) ; un collier d’onze perles, composées d’éléments cylindriques et sphériques et reliées par une liaison immobile, et d’une pendeloque en forme de bourgeon ainsi que de deux perles biconiques (Fig. 2) ; une plaque ornée de l’eigie d’un insecte et munie d’un double tube cannelé au dos, que B. Filov a supposé être un élément du fermoir du collier (Fig. 3) ; deux bracelets massifs coulés, se terminant par des têtes de serpent stylisées, ornés de ils torsadés et de grandes globules en or (Fig. 4) ; deux bagues dont les anneaux lourds sont plus épais sur leur partie inférieure. L’une des bagues est ornée d’une eigie de coq incisée, l’autre est sans décor (Fig. 5a, b). La collection du Musée d’Archéologie de Plovdiv possède une autre boucle d’oreille, quatre perles du même collier, un torque de corps constitué de ils torsadés et dont les extrémités sont recourbées vers l’extérieur (Fig. 6), un pectoral, trois plaques en or estampées d’une eigie de poisson, une amulette en forme de boîte cylindrique miniaturisée en or et une pendeloque en or en forme de pyramide (Filov, 1934 : 41-46). À ce jour leur étude technologique n’a pas encore été réalisable. Le tombeau du tumulus de Kukova est parmi les plus riches sépultures de la nécropole de Duvanli, connues par l’abondance d’objets de luxe importés. Les deux autres tumuli sont ceux de Muchovitza et le soi-disant tumulus d’Arabadjiiska. Figure 1 : (Voir planche couleur) Boucles d’oreilles et pendants d’oreilles. Figure 1: (See colour plate) Earrings and ear pendants. Figure 2 : (Voir planche couleur) Collier. Figure 2: (See colour plate) Necklace. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 201-209 Les parures en or de la nécropole thrace de Duvanli du Ve siècle av. J.-C. : le cas du tumulus de Kukova 203 Figure 3 : (Voir planche couleur) Plaque décorative. Figure 3: (See colour plate) Decorated plate. Figure 6 : Torque. Figure 6: Torc. Figure 4 : (Voir planche couleur) Bracelets. Figure 4: (See colour plate) Bracelets. Figure 5a, b : (Voir planche couleur) Bagues. Figure 5a, b: (See colour plate) Finger-rings. Dans le tumulus de Muchovitza, la tombe est demeurée intacte et nous possèdons des informations complètes sur le nombre de bijoux et leur position (Filov, 1934 : 88). La composition des pièces d’orfèvrerie répète celle des bijoux du tumulus de Kukova et leurs types sont aussi les mêmes, malgré les variations dans la forme et le décor : les boucles d’oreille en forme d’anneau ouvert sont au nombre de dix mais plus élaborées, portant des rosettes suspendues, et les deux pendants d’oreilles portent un col orné de granules et de ils perlés. Le collier est identique, présentant une variation des perles du type précédent. La parure inclut aussi un pectoral avec des sphinx estampés, attaché par des ibules en or « de type thrace », avec de longues chainettes et des pendeloques. Comme pour la tombe du tumulus de Kukova, la parure en or est entourée d’ofrandes aussi riches que dans la tombe de Muchovitza, (Filov, 1934 : 89-97). La parure ornant la tête du défunt du tumulus d’Arabadjiiskata est composée des mêmes types de bijoux mais présente une variante encore plus remarquable. Les boucles d’oreilles, en nombre de six, sont en forme de nacelle et les deux pendants d’oreille sont spiralés. Le collier, malgré la diversité des perles, est de type identique aux deux colliers des tumuli de Kukova et de Muchovitza. Tous les bijoux de cette tombe se caractérisent par leur granulation très ine et abondante qui donne à cette parure une allure très remarquable. Une bague en or ornée d’une eigie de cavalier et ArcheoSciences, revue d’archéométrie, 33, 2009, p. 201-209 204 d’une inscription ainsi qu’un pectoral au décor estampé font aussi partie de la parure (Filov, 1934 : 129-133). Une bague d’un style moins rainé présentant une inscription d’un nom propre thrace, accompagnée de deux pectoraux en or au décor estampé, a été découverte dans la tombe d’un guerrier du tumulus de Golyamata de la même nécropole (Filov, 1934 : 105-106). Les objets complexes et exceptionnellement riches furent accumulés pendant longuement. La plupart des vases en argent doré ou en bronze, des vases grecs igurés et en vernis noir, tous de vrais œuvres d’art, se rapportent à la péruode qui va de la in du vie, surtout du deuxième quart, à la moitié du ve siècle av. J.-C. (Reho, 1990 : n° 453, 454, 457 ; Archibald, 1998 : 158-166 ; Bozkova, 2004 : 55-59). Les pièces d’orfèvrerie datent de la même période (DeppertLippitz, 1985 : 122, 126, 130, 149). 2. MÉTHODES L’analyse des parures du tumulus de Kukova est réalisée en trois volets: l’un concerne l’étude typologique des pièces d’orfèvrerie, l’autre portant sur l’analyse de la composition chimique des diférents types de bijoux constituant une parure composite originale et un troisième portant sur l’analyse de leur particularités technologiques. 3. ANALYSE TYPOLOGIQUE ET STYLISTIQUE DES PIÈCES D’ORFÈVRERIE Une partie des bijoux des tombes de Duvanli se rapporte aux types fondamentaux de l’orfèvrerie grecque de l’époque classique. Ces types sont étudiées par R. Higgins (1961 : 119, 122-4, 141, 213), Deppert-Lippitz (1985 : 122, 126, 130) et Boardman (1970 : 296, 301). Z. H. Archibald (1998 : 190194) met l’accent sur leur liaison particulièrement étroite avec des spécimens de la Macédoine et de la Chalcidique. Des analogies peuvent être établies entre les boucles d’oreilles en forme d’anneau ouvert (Fig. 1) du tumulus de Kukova et des spécimens en bronze et en argent d’Olynthe (Robinson, 1941 : Pl. VII, 286, 289, 295). La variante la plus élaborée des boucles d’oreilles en anneau ouvert, ornée de rosettes ‘suspendues’, a été trouvée dans le tumulus de Muchovitza et s’approchent des formes macédoniennes particulières (Archibald, 1998 : 191-192). Les analogies les plus évidentes concernent les boucles d’oreilles en forme de nacelle du tumulus d’Arabadjiiskata. Une ressemblance encourageante ressort des boucles d’oreilles en bronze doré de la in du ve siècle av. J.-C. de la nécropole de l’ancienne ArcheoSciences, revue d’archéométrie, 33, 2009, p. 201-209 Milena TONKOVA, Petia PENKOVA Pydna, en Macédoine (Tsigarida et Ignatiadou, 2000 : 47, ig. 43), ainsi qu’une boucle d’oreille en argent et en bronze d’Olynthe ornée de iligrane et de granulation (Robinson, 1941 : 299, Pl. VII). Les pendants d’oreilles en forme d’oméga (Fig. 1) et spiralés de Duvanli sont tout aussi emblématiques pour l’orfèvrerie grecque, avec une très large dispersion et nombreuses variantes locales (Despini, 1996 : 32). Ici sont à mentionner les boucles d’oreilles spiralées du trésor de Kyme, en Asie Mineure, de la in du ve siècle av. J.-C. (Williams et Ogden, 1994 : 95, n° 47), qui rappellent le décor de plusieurs triangles granulés des boucles d’oreilles du tumulus d’Arabadjiiskata. Quant aux colliers de Duvanli, ils sont aussi des exemples du développement de l’orfèvrerie grecque du ve s. av. J.-C. (Deppert-Lippitz, 1985 : 122, 126, 130, 149). Le pendentif en forme de bourgeon et les perles faites d’une base, d’une tige solide et d’une perle biconique ou sphérique ou en forme de lentille (Fig. 2), avec des rosettes et ornées de iligrane et de granulation, montrent des parallèles généraux des pièces d’orfèvrerie synchrones (Archibald, 1998 : 190-191), mais ils manquent de ressemblances assez proches pouvant ofrir une base de localisation des ateliers initiaux en dehors de la hrace. En tant que composition complète, le modèle des colliers de Duvanli n’est pas retrouvé hors de la hrace, où il est représenté par trois variantes très proches l’une de l’autre. Ceci nous donne des arguments pour avancer l’hypothèse qu’ils étaient faits par des maîtres Grecs, liés eux-mêmes à la cour royale des Odryses. Les bagues du tumulus de Kukova (Fig. 5 a, b) présentent des types grecs de la in du vie ou du début du ve siècle av. J.-C. Le coq est présent sur des gemmes archaïques tardives (Boardman, 1970 : 145, pl. 325, 326), ainsi que sur des bagues métalliques de la même période (Zagdoun, 1984 : 193-195, n° 8). Le style réaliste de l’oiseau nous oriente vers un atelier grec, probablement de la côte sud de la hrace, où le symbole du coq se trouve sur des pièces de monnaie de Selimbria, sur la mer de Marmara, et de la Dycée thrace (Plant, 1979 : n° 1674, 1670), ou encore sur des pièces de monnaies anciennes du Pangée (Svoronos, 1919 : 229). Les bracelets (Fig. 4) du tumulus de Kukova sont pour le moment des spécimens uniques, qui se caractérisent par une tige coulée et lourde, plus large au milieu, se terminant par des têtes de serpents stylisées parsemées de globules d’or et de iligranes. Ces bracelets manifestent une stylisation très spéciique dans l’exécution des têtes de serpent, qui est connue sur des nombreux exemples de la région de la péninsule de Chalcidique (Amandry, 1953 : 50-1, cat.112-9). C’est dans leur style qu’on peut observer (Archibald, 1998 : 192 ; Tonkova, 2000-2001 : 283) le lien le plus proche et étroit entre les parures de Duvanli et de la Macédoine. Les parures en or de la nécropole thrace de Duvanli du Ve siècle av. J.-C. : le cas du tumulus de Kukova Le torque (Fig. 6) du tumulus de Kukova, aux ils torsadés et extrémités recourbées vers l’extérieur, suit des prototypes thraces du Premier Âge du fer faits en bronze ou en fer (Tonkova, 2000-2001 : 283 ; Gergova, 1987 : 65-6, A55-8 ; Archibald, 1998 : 192). Il s’agit sans doute d’une réminiscence du passé qui peut s’expliquer par le rôle d’insigne que joue le torque pour les aristocrates hraces. A l’origine des pectoraux l’on peut saisir l’inspiration des plaques ornementées de Macédoine, mais depuis le début du ve siècle av. J.-C. ils demeureront un phénomène thrace. Les ibules ‘de type thrace’ de Duvanli restent les seules connues exécutées en or. Leur nombre sera abondant dans les siècles suivants et leur dispersion couvrira tout le territoire de la hrace ancienne. L’analyse stylistique des bijoux de Duvanli a démontré une grande diversité de styles. Les boucles d’oreilles et les pendants d’oreilles ainsi que les colliers et les bagues sont des œuvres grecques. Pour certaines parures, des parallèles très concrets avec la Macédoine et les villes grecques de l’Égée peuvent être évoqués. Le style des bracelets nous oriente aussi vers des ateliers de la même région, mais travaillant dans un style local. Les pectoraux, le torque et les ibules portent les traits des œuvres de l’orfèvrerie thrace. 4. ANALYSE DE L’ENSEMBLE DES PIÈCES D’ORFÈVRERIE – UNE COMPOSITION STABLE Dans le groupe homogène de bijoux de style grec de Duvanli ressort une particularité qui le classe dans une catégorie incompatible avec les traditions d’orfèvrerie grecque : le grand nombre de boucles d’oreilles et leur composition avec des pendants d’oreilles dans une parure complexe ornant la tête. La manière de porter plusieurs boucles d’oreilles vient plutôt d’Orient, où il y a des exemples du ve siècle av. J.-C., c’est-à-dire contemporains des parures de Duvanli. Sept boucles d’oreilles en argent de ce type ont été trouvées dans un tumulus de la nécropole de Bayindir en Lycie septentrionale (Özgen et Öztürk, 1996, 164). Un autre exemple du début du ve siècle av. J.-C. vient du sarcophage d’enfant du célèbre tumulus de Kizoldun, de la région de Canakkale, où ont été trouvées huit boucles d’oreilles en or en forme de nacelle, interprétées comme quatre paires qui étaient simplement déposées des deux côtés de la tête et pas portées. Ce n’est qu’en Grèce ancienne que quelques exemples de multiples percements ont pu être trouvés (Özgen et Öztürk, 1996 : 56, ig. 126 ; Sevinc et Rose, 1999 : 500-502). Pourtant, le cas des boucles d’oreilles de Duvanli s’avère diférent, car dans les trois tumuli les paires de 205 boucles d’oreille présentes s’avèrent nombreuses : dix à Mouchovitza, neuf à Kukova (ici ces paires auraient pu être aussi dix, étant donné que la tombe n’était pas intacte), six à Arabadjiiskata. B. Filov (1934 : 88) nous renseigne sur la manière de les porter, en décrivant leur position dans la tombe de Muchovitza. Cette information précieuse nous permet de proposer une reconstruction hypothétique de cette parure originale : les boucles d’oreilles en chaîne ou attachées sur une bande ornaient plutôt la tête jusqu’aux tempes, où étaient suspendus les pendants d’oreilles (Tonkova, 2000-2001 : 277-283). Cet ensemble, se répétant dans trois riches tombes voisines et presque contemporaines de la nécropole de Duvanli, demeure original dans son unité, sans parallèle en Grèce ou ailleurs. Il est logique de considérer l’ensemble comme une parure de prestige, créée intentionnellement comme un riche accessoire d’un costume cérémonial. Combinée avec un collier de perles et un torque, des bagues, des bracelets et un pectoral, la parure devient une insigne, montrant le statut de leur propriétaires, haut et particulier dans la hiérarchie de la société des Odryses, où elle persiste sans changement au cours des deux siècles suivants (Tonkova, 1997 : 20). Étant intentionnelle, la création de ces parures complexes devrait être liée à des ateliers ailiés à la cour des Odryses. Ces ateliers pourraient être localisés dans les centres urbains de la mer Égée, les régions de la Macédoine et de la Chalcidique étant les plus probables, ainsi qu’à l’intérieur de la hrace. La deuxième hypothèse n’est pas infondée : parmi les bijoux il y a des œuvres d’orfèvrerie locale, telles le torque, les ibules et le pectoral ; des bijoux de types semblables sont fabriqués pendant une longue période et sont organisés dans des compositions presque identiques au cours de deux siècles durant. Le savoir-faire grec dans l’exécution des bijoux en or de Duvanli est bien visible et l’on peut admettre qu’auprès de la cour des Odryses eurent travaillé des orfèvres qualiiés dont le pays d’origine fut dans les régions mentionnées. Une présence grecque est signalée dans des sites de la in du vie et au ve siècle av. J.-C., très proches de la nécropole de Duvanli. Dans la région de la montagne Sredna gora, où se trouvent tous les sites mentionnés, sont concentrés des restes de gisements d’or exploités dans l’antiquité, qui pourraient être une des explications de la richesse de parures en or de la nécropole de Duvanli. Dans ce contexte, l’étude des bijoux en or de Duvanli pourrait trouver une perspective nouvelle. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 201-209 206 Milena TONKOVA, Petia PENKOVA 5. ANALYSES ÉLÉMENTAIRES DE L’OR 7. RÉSULTATS ET OBSERVATIONS TECHNOLOGIQUES L’analyse chimique des pièces d’orfèvrerie du tumulus de Kukova a été faite au Laboratoire du Musée national d’Archéologie, équipé d’un système d’ED-FX et d’un microscope optique digital. Comme les bijoux en question font partie de l’exposition permanente du musée, le temps nécessaire pour leur investigation était très limité et leur déplacement impossible. Ces circonstances ont limité le choix des méthodes d’analyse à l’utilisation des deux appareils disponibles. L’observation sous microscope optique avait pour but l’identiication et la documentation des indices des techniques utilisées dans l’exécution et la décoration des bijoux. Les analyses EDX avaient pour but, malgré le manque de standards de référence mais utilisant les mêmes conditions expérimentales, de nous donner la possibilité de former des groupes de bijoux à partir des alliages d’or utilisés. 6. MÉTHODE D’ANALYSE L’équipement de luorescence à rayons X utilisé est le modèle EDX-720 de Shimadzu. Les analyses sont réalisés à l’air, avec un tube à rayons X à anode de rhodium et fonctionnant à 50 kV et 100 μA, et avec un collimateur de 1 mm. Le tableau suivant présente les caractéristiques de l’équipement. Analyte anode kV µA Acq. (keV) Anal. (keV) Ti-U Rh 50 Auto 0-40 0,00-40.00 Na-Sc Rh 15 Auto 0-20 0.00-4.40 Le calcul des teneurs est réalisé au moyen du logiciel Quan-FP (paramètres fondamentaux). Une seule analyse est faite dans chaque région d’un objet. Les limites de détection sont de 0,01% pour les diférents éléments déterminés. La présence d’éléments dans les limites de détection est déterminée automatiquement par le logiciel. En tenant compte du fait qu’aucun standard de référence n’ait été utilisé ainsi qu’un quelconque traitement préliminaire de la surface des objets, les résultats obtenus doivent être considérés comme semi quantitatifs. Malgré les restrictions de la méthode utilisée et le fait que les données obtenues ne concernent que la surface des objets, les résultats sur la présence ou l’absence d’un élément sont valables ainsi que les compositions relatives. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 201-209 Les techniques décoratives (méthode d’exécution des ils iligranés des boucles d’oreilles en forme d’anneau ouvert (Fig. 1 ; Tableau 1) et du collier (Fig. 2 ; Tableau 4) sont proches. Les boucles d’oreilles possèdent un corps creux, et consistent en deux parties en repoussé jointes et deux bouts massifs (Fig. 1). Ces objets sont décorés de ils creux tordus tréilés (strip-twisted wire) (diam. 0,54 mm), de il perlé (beaded wire) (diam. 0,83 mm) et de granules seuls ou triples de dimensions diférentes (diam. 0,82 mm ; 0,92 mm). Les perles du collier et le pendentif (Fig. 2) sont faits en repoussé, et décorés de ils creux tordus tréilés (diam. 0,39 ; 0,47 mm), de il bobine (spooled wire) (diam. 0,68 mm) et de granules de diférentes dimensions. Ces bijoux possèdent des alliages d’or de composition identique. Pour les boucles d’oreilles et les perles du collier, les analyses FX ont montré la présence de cuivre uniquement dans les régions de jonction des granules et du iligrane (Tableaux 1, 4). L’interprétation de ce fait demande des analyses futures plus approfondies. Quant à la diférence des résultats 12 et 13 du Tableau 4, il s’agit de deux perles biconiques qui difèrent des autres perles et pourraient ainsi avoir été rajoutées plus tard à la composition originale du collier, ou lors de la reconstruction hypothétique faite au musée (étant donné que les bijoux du 1 Au% Ag% Cu% Zn% 76,2 23,3 – 0,2 2 77,6 23,6 0,6 – 2-1 76,4 21,5 – – 3 76,4 22,8 0,6 0,3 3-1 82,9 15,8 – 0,3 4 73,9 22,3 – – 4-1 77,3 21,9 – – 5 76,6 22,2 0,3 0,3 5-1 75,0 23,1 – 0,3 6 76,8 23,2 0,5 0,1 6-1 75,2 22,6 – – 7 75,7 23,0 0,4 0,3 7-1 76,7 20,9 1,0 0,5 Tableau 1 : Sept boucles d’oreilles, No d’inv. 6130, 6131, 6191. n : n° d’analyse dans la partie décorée ; n-1 : n° d’analyse sur la surface lisse ; n 3-1 : n° d’analyse de la plaque de réparation de la boucle d’oreille n° 3. Table 1: Seven earrings, Archive n. 6130, 6131, 6191. n:analyse reference on the decorated part; n-1: analyse reference on the lat surface; n 3-1: analyse reference on the restoration plaque of earring n°3. 207 Les parures en or de la nécropole thrace de Duvanli du Ve siècle av. J.-C. : le cas du tumulus de Kukova tumulus de Kukova n’ont pas été découverts au cours de fouilles régulières). La présence de Cr ne peut pas être expliquée, sauf par le manque de documentation du traitement de la surface depuis 1929 quand les bijoux ont été rachetés par le musée. Les pendants d’oreille en forme d’oméga (Fig. 1) sont décorés de pyramides recouvertes de granules, dont une plus grande se trouve au sommet ; la jonction avec la tige étant recouverte d’une plaque cannelée et de il perlé. L’alliage d’or (Tableau 2) est diférent des alliages utilisés pour fabriquer les bijoux commentés précédemment. Nous pouvons supposer qu’ils étaient exécutés simultanément aux autres mais à partir d’un lingot diférent, ou qu’ils ont été ajoutés plus tard pour compléter l’ensemble des boucles d’oreille. La plaque décorative (Fig. 3, Tableau 6) est ornée d’une eigie d’insecte en repoussé, contournée d’une ligne de granules (plutôt que de il granulé), de ils creux tordus tréilés et de granules. La composition de l’alliage d’or est proche de celle des bagues. Les bagues (Fig. 5a, b ; Tableau 3) sont exécutées en deux parties séparées (technique visible surtout dans la deuxième bague (Fig. 5b) : le chaton et l’anneau (coulé ou martelé) Au% Ag% Cu% Cr% 1 74,0 24.9 – 0,2 2 73,2 23,1 – 0,1 2-1 73,7 24,8 0,5 0,3 3 71,3 26,7 – – 3-1 72,4 25,8 0,6 – 4 75,8 21,9 – – 4-1 73,7 23,0 0,5 – 5 73,0 25,2 – – 5-1 72,9 23,8 0,6 – 6 72,2 25,4 – – 6-1 73,0 25,0 – – 7 67,8 21,0 – – 7-1 76,1 21,5 – – 8 72,04 25,9 – – 8-1 70,2 26,4 0,4 – 9 73,8 24,2 – – 9-1 71,8 25,2 0,7 – 10 71,3 26,6 – – 10-1 60,3 36,2 2,2 – 11 71,5 26,2 – – 73,0 25,5 0,3 – – Au% Ag% Cu% 11-1 1 84,4 15,2 - 12 76,0 19,6 1,1 1-1 82,8 15,6 0,7 12-1 77,7 19,3 0,9 – 76,0 18,5 1,7 0,1 76,5 16,6 1,6 – 1-2 82,7 16,0 0,8 13 2 84,4 15,2 0,3 13-1 2-1 85,0 13,7 1,2 Tableau 2 : Deux pendants d’oreilles, n° d’inv. 6134, 6190. n : n° d’analyse sur la surface lisse ; n-1 : n° d’analyse de la pyramide granulée. Table 2: Two ear pendants, Archive n. 6134, 6190. n: analyse reference on the lat surface ; n-1: analyse reference of the granulated pyramid. Au% Ag% Cu% 1 76,8 21,5 0,8 1-1 77,5 18,9 1,2 2 76,2 21,8 - 2-1 75,4 22,6 0,5 Tableau 3 : Deux bagues, n° d’inv. 6129, 6132. n 1, n-1 : n° d’analyse de la bague à l’eigie de coq ; n-2, n-1 : n° d’analyse de la bague à chaton lisse. Table 3: Two inger-rings, Archive n. 6129, 6132. n 1, n-1: analyse reference of the cock’s inger-ring; n-2, n-1: analyse reference of the lat inger-ring. 14 73,0 24,8 – – 14-1 73,1 25,4 0,2 – Tableau 4 : Collier, n° d’inv. 6136, 6192, 6193. n : n° d’analyse de chaque perle sur la surface lisse ; n-1 : n° d’analyse de chaque perle de la partie décorée ; n-12, 13 : n° d’analyse des deux perles biconiques cannelées ; n-14 : n° d’analyse du pendentif. Table 4: Necklace, Archive n. 6136, 6192, 6193. n: analyse reference for each bead on the lat surface; n-1: analyse reference for each bead on the decorated part; n-12, 13: analyse reference of both biconical luted beads; n-14: analyse reference of the pendant. d’une section ronde avec au milieu le gonlement caractéristique de l’époque. Le chaton de la bague à l’eigie de coq est ciselé (Fig. 5a) d’après l’observation sous microscope. Le chaton de la bague en forme de nid (Fig. 5b) est décoré d’un il lisse qui l’entoure à l’intérieur, et d’un il torsadé (très usé) qui accompagne les méandres de la bordure. Les bijoux de style grec du tumulus de Kukova sont exécutés avec une grande inesse, la qualité des granules et des ils, ArcheoSciences, revue d’archéométrie, 33, 2009, p. 201-209 208 Milena TONKOVA, Petia PENKOVA 1 Au% Ag% Zn% Cr% 82,2 15,7 – 0,2 1-1 81,4 17,3 0,4 – 2 83,9 14,1 – – 2-1 80,9 17,5 – – Tableau 5 : Deux bracelets, n° d’inv. 6128, 6189. n : n° d’analyse sur la surface lisse ; n-1 : n° d’analyse de la partie décorée. Table 5: Two bracelets, Archive n. 6128, 6189. n: analyse reference on the lat surface; n-1: analyse reference on the decorated part. Au% Ag% Cu% 1 77,1 20,2 0,3 2 76,7 21,4 0,5 Tableau 6 : Plaque décorative, n° d’inv. 6135. 1 : n° d’analyse sur la plaque en repoussé ; 2 : n° d’analyse du côté inverse. Table 6: Decorated plaque, Archive n. 6135. 1: analyse reference on the in repousse plaque; 2: analyse reference on the obverse. leur diversité, les jonctions précises, bien inies et cachées, sont dans les meilleurs traditions de l’orfèvrerie grecque. Toutes les techniques décoratives utilisées sont typiques de l’orfèvrerie de l’époque (Nicolini, 1990 ; De Cuyper, 1983). Les bracelets (Fig. 4 ; Tableau 5) du tumulus de Kukova qui difèrent nettement par leur style inspiré des modèles macédoniens, se diférencient aussi par leurs paramètres technologiques et par la composition de leurs alliages d’or. Les bracelets sont coulés et la jonction des deux parties du moule est bien visible. Les têtes de serpent, très caractéristiques, sont ornées de gros granules (de diam. max. de 3,78 mm) et légèrement aplaties ; une décoration qui demeure pour nous sans parallèle. Le il est massif, à section rectangulaire et torsadé. Visiblement, ces objets sont l’œuvre d’un atelier diférent. 8. CONCLUSION Bien que les observations des caractéristiques technologiques ne portent que sur une partie des parures du tumulus de Kukova, elles ouvrent une bonne perspective de recherche sur ces bijoux. Quoique préliminaires, les analyses de composition des alliages d’or utilisés et les particularités technologiques de l’exécution des bijoux de ce complexe funéraire complètent les connaissances sur les groupes formels et styArcheoSciences, revue d’archéométrie, 33, 2009, p. 201-209 listiques déjà établis. Ces données dessinent aussi de nouvelles pistes d’étude de la collection unique de bijoux en or du ve siècle av. J.-C. de la nécropole de Duvanli. Les variétés des caractéristiques technologiques des bijoux, dont les types sont eux-mêmes divers, viennent conirmer les résultats des analyses visuelles, indiquant la présence d’objets provenant de multiples ateliers d’orfèvres. Le style des bijoux aide à supposer des milieux culturels diférents pour leur provenance. Remerciements Le projet de recherche a pu être réalisé grâce à la subvention du Ministère de l’éducation et de la science de la République de Bulgarie (Contrat 01-1167/15.12.2007 entre l’Institut et Musée National d’Archéologie, Académie des Sciences Bulgare et le Ministère). Références AMANDRY, P., 1953. Collection Hélène Stathatos. Les bijoux antiques, Strasbourg. ARCHIBALD, Z. H., 1998. he Odrysian Kingdom of hrace. Oxford, Clarendon Press. BOARDMAN, J., 1970. Greek Gems and Finger Rings. Early Bronze Age to Late Classical, London, hames and Hudson. BOZKOVA, A., 2004. Attic black-glazed kylikes with inset lip from Bulgaria. – Archeologia, Soia, XLV, (1-2): 52-60 (en bulgare, un résume en anglais). DE CUYPER, F., 1983. Filigree, in T. Hackens and R. Winkes (eds.) Gold Jewelry. Craft, Style and Meaning from Mycenae to Constantinopolis (Aurifex 5), Louvain-La-Neuve, 196-204. DESPINI, A., 1996. Ancient Gold Jewellery. Athens, G.A. Christopoulos ed. DEPPERT-LIPPITZ, B., 1985. Griechischer Goldschmuck, Mainz am Rhein, Kulturgeschichte der Antiken Welt, 27. GERGOVA, D., 1987. Früh- und ältereisenzeitliche Fibeln in Bulgarien, PBF XIV, 7. Band, München. FILOV, B., 1934. Nadgrobnote mogili pri Duvanli, Plovdivsko (Les tumuli près de Duvanli, région de Plovdiv). Soia. HIGGINS, R., 1961. Greek and Roman Jewellery. London, Methven and co. NICOLINI, G. 1990. Techniques des ors antiques. La bijouterie ibérique du VIIe au IVe siecle. Paris, Picard. ÖZGEN, I. et ÖZTÜRK, J., 1996. he Lydian Treasure. Istanbul, Ministry of culture. PLANT, R., 1979. Greek Coin Types. London, Numismatic Fine Arts. Les parures en or de la nécropole thrace de Duvanli du Ve siècle av. J.-C. : le cas du tumulus de Kukova REHO, M., 1990. La ceramica attica a igure nere e rosse nella Tracia Bulgara. Roma, Bretschneider. ROBINSON, D., 1941. Excavation at Olynthus. Olynthus, Part X, London. SEVINC, N. et ROSE, Ch. B., 1999. A child’s sarcophagus from the salvage excavations at Gumuscay, Studia Troica, band 9, 489-509. SVORONOS, J.-N., 1919. L’hellénisme primitif de la Macédoine. Paris. TONKOVA, M. 1997. Traditions and Aegean inluences on the Jewellery of hracia in early hellenistic times, Archaeologia Bulgarica I, (2): 18-31. 209 TONKOVA, M. 2000-2001. Classical Jewellery in hrace: origins and development, archaeological contexts, Talanta, XXXIIXXXIII: 277-288. TSIGARIDA, B. and IGNATIADOU, D., 2000. he Gold of Macedon. Archaeological Museum of hessaloniki. Athens. WILLIAMS, D. and OGDEN, J., 1994. Greek Gold. Jewelry of the Classical World. New York, H.N. Abrams ed. ZAGDOUN, M.-A., 1984. L’antre Corycien II. Bagues et anneaux. BCH, supplément IX, chapitre IV, 183-260. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 201-209 he technique of gold inlaid decoration in the 5th-4th centuries BC: silver and iron inds from the early Sarmatian barrows of Filippovka, Southern Urals La technique de décoration par inclusion d’or aux Ve-IVe siècles av. J.-C. : les trouvailles en argent et fer des tumulus de la période Sarmate ancienne de Filippovka au sud de l’Oural Marina Shemakhanskaya*, Mikhail Treister** and Leonid Yablonsky*** Abstract: he present paper addresses silver and iron objects (iron sword awl, silver quiver hooks, and an iron knife with a silver handle) decorated with inlaid gold strips and plates. he artefacts were excavated recently in the barrows near the village of Filippovka, in the Southern Urals, and were dated to the 5th-4th centuries BC. Typological and stylistic analyses are complemented with the study of technological features of the objects. Analysis under the Analytical Scanning Electron Microscope (SEM), carried out during the restoration of the objects, revealed the speciic features of Urals gold. Most of the objects analysed represent local forms of the early Sarmatian culture. his technique of gold inlaid decoration was spread by the nomads of the Eurasian steppes from the Altai Mountains in the east to Central Kazakhstan and the Southern Urals in the west during a period between the 7th and the 4th centuries BC; furthermore, the technique is also attested in Achaemenid metalwork. A unique silver black patinated handle of a knife, shaped in the form of a stag, was inspired by the ‘Achaemenid international style’, but the peculiarities of style and the inlaid technique do not exclude the possibility that it could have been manufactured in a provincial Achaemenid workshop for a Sarmatian customer. Résumé : L’article traite les objets en fer et argent (épée et alêne en fer, agrafes de carquois en argent et couteau avec poignée en argent) décorés de bandes et de feuilles d’or, trouvés récemment dans les tumulus proches de la ville de Filippovka, dans le sud de l’Oural, et datant des Ve-IVe siècles av. J.-C. Les analyses stylistiques et typologiques sont complétées par l’étude des détails techniques des objets. Les observations réalisées sous microscope électronique à balayage (MEB) pendant leur restauration ont révélé les caractéristiques de l’or de l’Oural. La majorité des objets analysés appartiennent aux formes locales de la culture Sarmate ancienne. La technique de décoration par inclusion d’or, observée dans le travail du métal des Achéménides, a été difusée pendant le VIIe-IVe siècle av. J.-C. par les nomades des steppes Eurasiennes, à partir des monts Altaï vers l’est jusqu’au Kazakhstan central et vers l’ouest jusqu’au sud de l’Oural. Une poignée de couteau, pièce unique en forme de cerf patiné à l’argent noir, a été inspirée par le style Achéménide international, mais les particularités du style et de la technique d’inclusion n’excluent pas la possibilité d’une fabrication dans un atelier Achéménide provincial destinée à un Sarmate. Keywords: Early Sarmatian culture, gold inlaid decoration, iron, silver, Southern Urals. Mots-clés : Culture Sarmate, inclusion d’or, décoration du fer, argent, Oural du sud. * State Institute for Conservation – Gastello St., 44, 107014 Moscow, Russia. (71aniram@gmail.com) ** Institute of Near Eastern Archaeology, Free University of Berlin – Weißenburgstr. 59, 53175 Bonn, Germany. (mikhailtreister@yahoo.de) *** Institute of Archaeology, Russian Academy of Sciences – Dmitriya Ul’yanova St., 19, 117036 Moscow, Russia. (leonid.yablonsky@mtu-net.ru) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 211-220 212 1. IntroductIon he recent excavations (2004-2008) of the barrows near the village of Filippovka, in the Southern Urals, carried out by the Institute of Archaeology, Russian Academy of Sciences, brought to light remarkable burial and ritual complexes belonging to the early Sarmatian culture (5th4th centuries BC) (Yablonsky and Meshcheryakov, 2007; Catalogue Orenburg 2008). his paper focuses on a series of silver and iron objects with gold inlays, originating from the Filippovka barrows nos. 4 and 15, currently on display in the Orenburg Museum of local lore. 2. Methods Typological and stylistic analyses of archaeological objects were performed. he gold inlays of the iron and silver objects from Filippovka were studied under the microscope during restoration works. he gold inlaid decoration of the sword was studied with the SEM coupled with an energy-dispersive X-ray spectrometer. 3. results (M. Shemakhanskaya, M. Treister and L. Yablonsky) Barrow no. 4, one of the two largest, centrally located, and most probably princely barrows, included numerous burials of horses, a sacriice place, and burial pits, four of which are dated to the early Sarmatian period (Yablonsky and Meshcheryakov, 2007). he objects under discussion were found in burials nos. 2, 3 and 5. An iron sword from burial no. 2 has a butterly-shaped cross-piece decorated with engraved compositions, including a scene of hunting and immolation of deer. Its base is formed with two iron bars, with cast silver gold inlaid reliefs ixed on them. On the blade, there are inlays showing animals and the sacriice of a stag by the warriors (Catalogue Orenburg 2008: no. 39) (Fig. 1). Various types of gold wires and narrow plates were hammered in the hollows carved on the iron blade (Fig. 1[2]) and on the pommel (Fig. 1[3]). he proiles of the hollows may be seen, mirror-relected, on the rear side of the inlays. he hollows on the blade and on the pommel are made with diferent instruments. Two other objects originate from the same burial: a small iron awl with a pendant attached to its inial and a blade with gold inlays forming a spiral pattern, executed in the same way as the inlays on the blade of the sword (Catalogue Orenburg 2008: no. 40) (Fig. 3, [1-2]), and a cast silver quiver hook, ArcheoSciences, revue d’archéométrie, 33, 2009, p. 211-220 Marina SHEMAKHANSKAYA, Mikhail TREISTER, Leonid YABLONSKY with thin gold inlays of various shapes (Fig. 4), showing a predator devouring a griin. A quiver hook of cast silver from burial 3 shows a composition with a griin and a panther. he details are accentuated with gold inlays of thin sheet gold, hammered in the hollows (Catalogue Orenburg 2008: no. 9) (Fig. 5, [1-3]). Numerous burials were investigated in barrow no. 15, located in the north-eastern part of the cemetery. Burial 1 yielded remains of a battle-axe, with its head in the form of a bear, inlaid with gold wires (Fig. 3, [3]) (Balakhvantsev and Yablonsky 2007: 147, Fig. 1, 7). In burial 3, a silver quiver hook showing a feline predator and a head of a bird of prey, inlaid with gold (Catalogue Orenburg, 2008: no. 84) (Fig. 5, [4-5]), was found. Stylistically and technologically, this artefact it is similar to the ind from burial 3 of the barrow no. 4 (Fig. 5, [1-3]), although it varies in details and the workmanship is more crude. An iron knife with a cast silver handle in the form of a stag with antlers, decorated with inlays of gold (Col. pl., a-c) (Catalogue Orenburg, 2008: no. 42), found in the central burial 5 of the barrow no. 4, attracted our special attention. he handle was cast after the wax model of silver, with admixture of copper (2-5%) over the blade of the knife. he material used for the inlays shows ca. 84% gold, 14% silver, and small admixtures of copper. he surface is covered by silver sulphide (silver chlorides, characteristic for archaeological silver, are completely absent) of almost black colour. his aspect is unusual for archaeological silver objects, and indicates that it was probably done intentionally, to contrast with the gold inlays. Gold inlays in the form of circles, curls, strips and triangles decorate the head (Col. pl., b) and the body (Col. pl., a) of the stag, which is rendered in a low relief. Nine parallel inlays, slightly waved and narrowing towards the inial, deviated originally from the bottom of the handle towards the head of the stag – three of them are lost, leaving open the hollows of the inlays (Col. pl., c). he objects from Filippovka under discussion here have various shapes and functions, and the inlaid gold strips and plates were used to stress elements of animal bodies, or to create ornamental or igural compositions. On the majority of objects, these strips and plates were inserted in the hollows cut in their bodies (as on the sword or quiver hooks). In iron objects, they were hammered inside the hollows. In the silver hook shown in Figure 4, the gold inlays have even edges, and only minor signs of hammering are present. In some cases, the inlays are somewhat smaller than the hollows, and they were most probably glued in – in antiquity, natural resins were used for gluing gold foils (Hockey et al., 1992: 281-291; Oddy, 1990: 29-33). he bottoms of the hollows on the basis of the silver knife handle are lat and he technique of gold inlaid decoration in the 5th-4th centuries BC… 213 Figure 1: Iron sword with silver cross-guard and gold inlays. Filippovka, barrow no. 4/2006, burial 2. (1): general view; (2-4): undersides of the gold inlays showing marks of the tools used to create the groove in which it was set – (2), (3): on the blade; (4): on the pommel; (5): detail of the inlay decoration on the blade. Figure 1 : Épée en fer hachurée à l’argent et avec des inclusions d’or. Filippovka, tumulus no. 4/2006, sépulture 2. 1 – vue générale; 2-4 –dessous des inclusions d’or montrant les marques des outils utilisés pour la production de la rainure (2-3 – sur la lame ; 4 – sur le pommeau) ; 5 – détail de la décoration en inclusion de la lame. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 211-220 214 Marina SHEMAKHANSKAYA, Mikhail TREISTER, Leonid YABLONSKY even, the walls are vertical, and there are no marks of cutting instruments. his leads us to the assumption that they were made by the insertion of thin gold plates in the wax model, which were later inlaid in the hollows, and ixed with an amalgam of mercury – the presence of the latter is conirmed by the microprobe analysis of the edge of the inlay, showing gold, silver and mercury. he use of mercury for ixing gold foils to the surface of silver objects is attested in ancient metalwork (Vittori, 1979: 35-39). 4. the analytIcal study of gold (M. S. Shemakhanskaya) During the restoration of the inlaid decoration of the sword, in all the parts analysed by various techniques applied to the gold surface, small hard bright inclusions of indeinite shapes and various dimensions could be observed’? But what does ‘light’ mean in this context. he SEM study of the gold inlays, carried out at the Institute of Geology of Ore Deposits, Petrography, Mineralogy and Geochemistry, Russian Academy of Sciences (IGEM RAS), revealed that these inclusions contain platinum group elements: osmium, iridium, ruthenium and platinum (Fig. 2, [1-2]). he correlation of these elements in diferent inclusions of the same object varies. Such inclusions, in various quantities, are attested in the gold inlays of all examined objects, except for the two silver hooks from burials 3 in the barrows nos. 4 and 15 (Fig. 5). Similar inclusions were revealed during the study of other gold objects found in contemporary nomad burials in the Southern Urals; one particular example is that of a gold animal style bracelet found in the barrow no. 3 at Kichigino (Zaikov et al., 2008b: 47). he presence of osmium, iridium, ruthenium and platinum in the gold inlays (Fig. 2) is generally comparable with the composition of granules of alluvial gold from the Miass and Kochkar mining regions, located 500-600 km north-east of Filippovka (Yushkin 1990: 46-48; Kobiashev and Nikandrov, 2007: 223; Zaikov et al., 2008a: 404; Zaikov et al., 2008b: 47, 49). Although methods of reining gold from inclusions were known in antiquity (Craddock, 2000: 200-211), it is evident that the gold used for the inlays of the sword from Filippovka was not reined. 5. dIscussIon (M. Treister) Most of the objects under discussion represent local forms, characteristic of the early Sarmatian culture of the ArcheoSciences, revue d’archéométrie, 33, 2009, p. 211-220 Figure 2: Results of the analytical study of the gold inlays of the sword (see Fig. 1). (1): inclusion of a platinum group metal in the gold inlay of the sword under SEM; (2): the Roentgen spectrum of the platinum group metals obtained with SEM; the parameters of the spectrum are given at the bottom. Figure 2 : Résultats de l’étude analytique au MEB des inclusions d’or de l’épée (voir Fig. 1). 1. inclusion de platinoïdes sur la feuille de l’épée ; 2. spectre à rayons X de l’inclusion de platinoïdes, en bas les paramètres d’acquisition. Southern Urals. hus, the quiver hooks analysed here, representing compositions of griins and predators, ind parallels among the bronze inds from the barrows of the Southern Urals, Mountainous Altai, Tuva, and Eastern Kazakhstan (Korol’kova, 2006: pls. 60, 1-6; 61, 13-14; 63, 1. 3). At he technique of gold inlaid decoration in the 5th-4th centuries BC… Figure 3: (a): Iron awl with gold inlays. Filippovka, barrow no. 4/2006, burial 2; (b): detail of the gold inlay of the awl; (c): detail of an iron battle-axe. Filippovka, barrow no. 15/2004, burial no. 1; gold inlays shown as lighter grey/white. Figure 3 : 1. Alêne en fer avec des inclusions d’or. Filippovka, tumulus no. 4/2006, sépulture 2 ; 2. détail de l’inclusion d’or ; 3. détail de la hache de guerre en fer. Filippovka, tumulus no. 15/2004, sépulture no. 1 ; inclusions d’or en gris clair/blanc. the same time, the style of execution of a panther’s head, especially on the hook from burial 3 in the barrow no. 4 (Fig. 5, [2]) reminds us of that of the gold torc from burial 2 of the same barrow (Catalogue Orenburg, 2008: no. 1). In contrast, the ‘running wave’ pattern decorating the beak of the griin is a decorative element unusual for Sarmatian objects, but widespread in the Greek world. he decoration of the sword may be compared with those present on the swords from the Filippovka barrow no. 1 (Aruz et al., 2000: nos. 5-6; Catalogue Orenburg 2008: 98, no. 39), and with those from the princely Saka burial in the Issyk barrow in 215 Kazakhstan (Akishev, 1978: 105, pl. 24, right; Popescu et al., 1998: 187, no. 331), while the general category to which these swords belong may be traced back to the 7th century BC sword and dagger (Čugunov et al., 2006: 118, no. 8, pls. 19-20; 133, no. 28, pls. 63, 65) from the Arzhan-2 barrow in the Altai Mountains. Also, the decoration style of the small iron awl, in the form of elaborate S-shaped spirals (Fig. 3) is characteristic for numerous gold objects from Filippovka, including golden overlays of wooden vessels (Aruz et al., 2000: nos. 25, 27-28, 69). he knife handle (Col. pl., a-c) inds a close parallel in a hippopotamus ivory handle of a knife (Sevinç et al., 1998: 312, Figs. 8-9; 320, no. 17) found in a tumulus at Dedetepe in north-western Turkey, dating to ca. 480-460 BC. Its Achaemenid inspiration is further conirmed by the outlined beard leading to the ears, and especially by a typical representation of the animal’s hindquarters in relief, as seen on some of the Achaemenid rhyta, or on the majority of silver amphora-rhyta (Amandry, 1959: 38-56; Pfrommer, 1990: 191-209; Calmeyer, 1993: 152-153, pls. 43-44; Boardman, 2000: 188-189, 246, note 129), including the recent ind from burial 4 of the Filippovka barrow no. 4 (Yablonsky and Meshcheryakov, 2007: 57, Fig. 3, Col. pl. 1; Yablonsky, 2007: 88, 93, Fig. 7; Catalogue Orenburg 2008: no. 5; Balakhvantsev and Yablonsky, 2008: 30, 34, Fig. 5, 1; Treister, 2008: 158-160, Fig. 6). We can also ind similar treatments of animal bodies in gold Achaemenid jewellery, in particular on the bracelets (Yablonsky and Meshcheryakov, 2007: 57-58, Fig. 4, Col. pl. 2; Yablonsky, 2007: 89, 95, Fig. 9; Catalogue Orenburg, 2008: no. 14) and on the torc from the same barrow in Filippovka (Yablonsky and Meshcheryakov, 2007: 58, Fig. 5, Col. pl. 1; Catalogue Orenburg, 2008: no. 2). At the same time, the treatment of separate elements of the representation (the eyes, wings, joints, shoulder blades, etc.) on the knife handle from Filippovka diverges from the canons characteristic for art of the Achaemenid style. While on the Achaemenid rhyta/amphora-rhyta and jewellery, as well as on the ivory knife handle from the Dedetepe tumulus, we come across the canonical depiction of the eye, framed with relief rims and with an accentuated lacrimal sac (Rehm, 1992: 257), on the knife handle from Filippovka, the stag’s eye is round, slightly convex, and framed with a cut and a wide gold strip. he wings of the animals on the Achaemenid silver vessels always show carefully treated detailed feathering (Rehm, 1992: 256-257, 263-264); in contrast, in the case of the object under consideration here, the modelling of the wing is very coarse, and the feathering is shown by strips of the same width as those making up the outline of the wing. In Achaemenid jewellery and toreutics, the joints are usually ArcheoSciences, revue d’archéométrie, 33, 2009, p. 211-220 216 Marina SHEMAKHANSKAYA, Mikhail TREISTER, Leonid YABLONSKY Figure 4: Silver quiver hook with gold inlays. Filippovka, barrow no. 4/2006, burial 2. (a), (b): side views; (c): view from above; (d): detail – the hollow for the gold inlay; (e): side view with gold inlays fallen out; (f ): gold inlays. Figure 4 : Agrafe de carquois en argent avec inclusions d’or. Filippovka, tumulus no. 4/2006, sépulture 2. 1 ; 2. vues de côté ; 3. vues d’en haut ; 4. détail de l’oriice pour l’inclusion de l’or ; 5. vue de côté avec des inclusions d’or tombées ; 6. inclusions d’or. worked in detail, and there are canonical ways of rendering the shoulder blades, consisting of two oval, or one circular and one drop-shaped, elements (Rehm, 1992: 254-255, 262-263). Once again, this is not the case for the knife handle from Filippovka. he technique of decorating silver objects with gold inlays is attested in Achaemenid metalwork already since the 6th century BC (Moorey, 1988: 231-246), in particular for the silver ArcheoSciences, revue d’archéométrie, 33, 2009, p. 211-220 bowls and a rhyton, the outer surface of which was decorated with the friezes of gold matrix-hammered appliqués showing a Persian king, warriors, birds, or stepped pyramids. A silver vessel, decorated with concentric incised lines, and inlaid with gold wire, originates from cache 1 of the barrow no. 1 in Filippovka (Aruz et al., 2000: no. 19). A horse-harness strapdivider in the form of a griin head, of unknown provenance, currently on display in the British Museum, requires special he technique of gold inlaid decoration in the 5th-4th centuries BC… 217 Figure 5: Silver quiver hooks with gold inlays. (1)-(3): Filippovka, barrow no. 4/2006, burial 3 (1, 3: general view; 2: detail – head of a panther); (4), (5): Filippovka, barrow no. 15/2004, burial 3 (4: side view; 5: detail – head of a panther); gold inlays shown as lighter grey. Figure 5 : Agrafe de carquois en argent avec inclusions d’or. 1-3. Filippovka, tumulus no. 4/2006, sépulture 3 (1, 3. vue générale ; 2. détail : tête de panthère) ; 4-5. Filippovka, tumulus no. 15/2004, sépulture 3 (4. vue de côté ; 5. détail : tête de panthère) ; inclusions d’or en gris clair. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 211-220 218 Marina SHEMAKHANSKAYA, Mikhail TREISTER, Leonid YABLONSKY Col. pl. 1: (1)-(3): (See colour plate) Silver knife handle with gold inlays, Filippovka, barrow no. 4/2006, burial 5; (4): gold work on the blade of the sword (see Fig. 1); (5): gold inlays in the silver quiver hook (see Fig. 5, 1-3). Col. pl. 1 : (Voir planche couleur) 1-3: Poignée de couteau en argent avec inclusions d’or, Filippovka, tumulus no. 4/2006, sépulture 5 ; 4 : travail de l’or sur la lame de l’épée (voir Fig. 1) ; 5: inclusions d’or de l’agrafe du carquois en argent (voir Fig. 5, 1-3). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 211-220 219 he technique of gold inlaid decoration in the 5th-4th centuries BC… attention, as its details are rendered with gold inlays. he object is a cast piece of copper-gold alloy, and it has been deliberately given a dark patina in antiquity in order to enhance the surface contrast with the inlay (Curtis and Tallis, 2005: no. 389). his, we suggest, was also the case with the knife handle from Filippovka. No technical investigation of the Achaemenid silver vessels with gold inlays has been published to date. In the bowl from the British Museum, ANE 134740 (Moorey, 1988: 233-234, no. 1, pl. Ia; Seipel, 1996: no. 60; Curtis and Tallis, 2005: no. 111), the hollows were cut for the inlays, and they were hammered into the undercut edges; no mercury was found on the gold inlays by XRF. Although the silver igures of animals forming a circle, which were used primarily as Achaemenid vessel handles, are often additionally inlaid with gold, the decoration of the knife handle from Filippovka difers from these in the shape of the gold inlays. he treatment of the details of the stag body on the handle of the silver knife (Col. pl., a) also difers from the images of stags in the inlaid decorations on the sword (Fig. 1[5]; Col. pl., d), from the gold overlays of wooden vessels (Aruz et al., 2000: nos. 25-28, 30, 34), and from wooden sculptures of stags overlaid with gold (Aruz et al., 2000: nos. 1-4, 20-22, 84-92) originating from Filippovka, with their highly stylized decorations of elaborate spirals. he style is much more similar, although not identical, to the more realistic images which we encounter in the art of the Pazyryk culture of the Altai, dating to the 4th-3rd centuries BC, for example, that of the wooden carved plates of horse-harness from the barrow no. 11 of the Berel cemetery in Eastern Kazakhstan (Samašev et al., 2002: 269, Fig. 26), and of the silver belt plaques from the Pazyryk barrow no. 2 (Rudenko, 1953: 124, pl. XXVII, 1-2; Polos’mak and Barkova, 2005: 49, Fig. 2.22, d). 6. conclusIon (M. S. Shemakhanskaya and M. Treister) he use of the technique of decorating iron and bronze objects with gold inlays by the nomads of Eurasia already in the 7th century BC is attested by the decoration of a battle axe (Čugunov et al., 2006: 121, no. 13, pl. 26) and arrowheads (Čugunov et al., 2006: 123-124, no. 16, pls. 31-32) from the princely Arzhan-2 barrow in Southern Siberia (Armbruster, 2007: 99). his technique remained in use in the period between the 6th and the 4th century BC, as indicated by the inds of: a gold-inlaid iron knife handle in the Shibe barrow in Southern Siberia (Popescu et al., 2001: no. 201), a gold-inlaid iron pin in the barrow no. 22 of the cemetery Doge-Bary in Tuva (Čugunov, 1998: 302, Fig. 16, 3), and the plaques in the barrow no. 3 of the Tasmola-V necropolis in Central Kazakhstan (Popescu et al., 1998: nos. 145-150). he locations of the inds span the vast belt stretching from the Altai Mountains in the east to the Southern Urals in the west. he handle of the knife from Filippovka (Col. pl., a-c) cannot be discussed as an example of ‘international Achaemenid style’, although it was inspired by such an item, most probably dating to the irst half of the 5th century BC. We do not know who created this knife handle. It is worth noting that the technique of making hollows for the inlays already in the wax model is diferent from the technique of cutting hollows, used for the decorations of the hooks and the crossguard of the sword. Also remarkable is the method used for ixing the inlays in the hollows. he peculiarities related to its style and to the inlaid technique itself do not exclude the possibility that the knife was manufactured in a provincial Achaemenid workshop and that the respective craftsmen considered the tastes of a Sarmatian customer. he characteristic features of Urals gold were irst attested in the course of our study in numerous objects from nomad burials. his allows a correlation of the metal of these items with the sources of gold. he geographical and chronological distribution of objects with such features requires further studies. Acknowledgements he present study was carried out with the inancial support of the German Science Foundation (Grant no. BO 1599/61) and the Russian Foundation for Humanities (Grant no. 09-01-00070a/d). he authors are grateful to Dr. Andrew V. Mokhov and Dr. Pavel M. Kartashov (IGEM RAS, Moscow) for the analyses of the objects discussed, and to Dr. Susan La Niece for the information on the bowl ANE 134740 from the British Museum. References akIshev, a.k., 1978. Issyk Mound. he Art of Saka in Kazakstan. Moscow: Isskustvo Publications. aMandry, P., 1959. Toreutique achéménide. Antike Kunst 2: 38-56. arMbruster, b., 2007. Die Goldschmiedetechnik von Aržan 2, in H. Parzinger (ed.), Im Zeichen des Goldenen Greifen. Königsgräber der Skythen. München, Berlin, London, New York, Prestel, 95-99. aruz, J., farkas, a., alekseev, a. and korol’kova, e.f. (eds.), 2000. he Golden Deer of Eurasia. Scythian and Sarmatian Treasures from the Russian Steppes. New York, he Metropolitan Museum of Art and Yale University Press. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 211-220 220 balakhvantsev, a.s. and yablonsky, l.t., 2007. Akhemenidskaya emal’ iz Filippovki (problema khronologii pamyatnika). Rossiiskaya archaeologiya 1: 143-149. balakhvantsev, a.s. and yablonsky, l.t., 2008. Serebrannaa amfora iz Filippovki, in L.T. Yablonsky (ed.), Rannie kochevniki Volgo-Ural’skogo regiona. Materialy mezhdunarodno nauchno konferentsii “Rannie kochevniki uzhnogo Priural’ya v svete noveishikh arkheologicheskikh otkryti”, 21-25 aprelya 2008. Orenburg, Orenburg State Pedagogical University, 29-38. boardMan, J., 2000. Persia and the West. London, hames and Hudson. calMeyer, P., 1993. Die Gefäße auf den Gabenbringer-Reliefs in Persepolis. Archäologische Mitteilungen aus Iran 26: 147-160. catalogue orenburg, 2008. Sokrovishcha sarmatskikh vozhdej (Materialy raskopok Filippovskikh kurganov). Orenburg, Dimur. craddock, P.t., 2000. Reconstruction of the Salt Cementation Process at the Sardis Reinery, in A. Ramage, P.T. Craddock (eds.), King Croesus’, Gold. London, he British Museum Press, 200-211. curtIs, J. and tallIs, n. (eds.), 2005. Forgotten Empire. he World of ancient Persia. London, he British Museum Press. Čugunov, k.v., 1998. Der skythenzeitliche Kulturwandel in Tuva. Eurasia Antiqua 4: 273-308. Čugunov, k.v., ParzInger, h. and nagler, a., 2006. Der Goldschatz von Aržan. Ein Fürstengrab der Skythenzeit in der südsibirischen Steppe. Munich, Schirmer/Mosel. hockey, M., Johnston, a., la nIece, s., MIddleton, a., swaddlIng, J., 1992. An Illyrian Helmet in the British Museum. he Annual of the British School at Athens 87: 281-291. kobIashev, yu.s. and nIkandrov, s.n., 2007. Mineraly Urala (mineral’nye vidy i raznovidnosti). Ekaterinburg: Quadrat. korol’kova, E.F., 2006. Zverinyi stil’ Evrazii. Iskusstvo plemen nizhnego Povolzh’ya Yuzhnogo Priural’ya v skifskuyu epokhu (VII-IV vv. do n.e.). St. Petersburg, Peterburgskoe vostokovedenie. Moorey, P.r.s., 1988. he Technique of Gold-igure Decoration on Achaemenid Silver Vessels and its Antecedents. Iranica Antiqua 23: 231-246. oddy, w.a., 1990. Gilding – an Outline of the Technological History of the Plating of Gold on to Silver or Copper in the Old World. Endeavour 15(1): 29-33. PfroMMer, M., 1990. Ein achämenidisches Amphorenrhyton mit ägyptischem Dekor. Archäologische Mitteilungen aus Iran 23: 191-209. Polos’Mak, n.v. and barkova, l.l., 2005. Kostyum i tekstil’ azyryktsev Altaya (IV-III vv. do n.e.). Novosibirsk, Infolio. PoPescu, g.a., antonInI, c.s. and baIPakov, k. (eds.), 1998. L’uomo d’oro. La cultura delle steppe del Kazakhstan dall’età del bronzo alle grandi migrazioni. Venice, Electa. PoPescu, g.a., alekseev, a. and PIotrovskIJ, J. (eds.), 2001. Siberia. Gli uomini dei iumi ghiacciati. Milan, Electa. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 211-220 Marina SHEMAKHANSKAYA, Mikhail TREISTER, Leonid YABLONSKY rehM, e., 1992. Der Schmuck der Achämeniden. Münster: UgaritVerlag. rudenko, s.I., 1953. Kul’tura naseleniya gornogo Altaya v skifskoe vremya. Moscow and Leningrad, Publishing House of the Academy of Science of the USSR. saMašev, z.s., bazarbaeva, g.a and ŽuMabekova, g.s., 2002. Die goldhütenden Greife des Herodot und die archäologische Kultur der frühen Nomaden im kazachischen Altai. Eurasia Antiqua 8: 237-276. seIPel, w. (ed.), 1996. Weihrauch und Seide. Alte Kulturen an der Seidenstrasse. Kunsthistorisches Museum Wien. Vienna, Skira. sevInç, n., rose, c.b., strahan, d. and tekkök-bIçken, b., 1998. he Dedetepe Tumulus. Studia Troica 8: 305-327. treIster, M.yu., 2008. Proizvedenija torevtiki akhemenidskogo stilya i na “akhmenidskuyu temu” v sarmatskikh pogrebeniyakh Prokhorovskogo Filippovskogo kurgannykh mogil’nikov v Yuzhnom Priural’e, in L.T. Yablonsky (ed.), Rannie kochevniki Volgo-Ural’skogo regiona. Materialy mezhdunarodno nauchno konferentsii “Rannie kochevniki uzhnogo Priural’ya v svete noveishikh arkheologicheskikh otkryti”, 21-25 aprelya 2008. Orenburg, Orenburg State Pedagogical University, 145-170. vIttorI, o., 1979. Pliny the Elder on Gilding. Gold Bulletin 12(1): 35-39. yablonsky, l.t., 2007. Excavations of a “Royal” Barrow-Grave in the Southern Ural. Folia Archaeologica LIII: 85-100. yablonsky, l.t. (ed.), 2008. Rannie kochevniki Volgo-Ural’skogo regiona. Materialy mezhdunarodno nauchno konferentsii “Rannie kochevniki uzhnogo Priural’ya v svete noveishikh arkheologicheskikh otkryti”, 21-25 aprelya 2008. Orenburg, Orenburg State Pedagogical University. yablonsky, l.t. and Meshcheryakov, d.v., 2007. Raskopki „tsarskogo“ kurgana v Filippovke (predvaritel’noe soobchenie). Rossiiskaya Archaeologiya 2: 55-62. yushkIn, I.P. (ed.), 1990. Mineralogiya Urala. Elementy. Sverdlovsk, Ural Department of the Academy of Sciences of the USSR. zaIkov, v.v., yuMInov, a.M., kotlyarov, v.a., taIrov, a.d., ePIMakhov, a.v. and zdanovIch, d.g., 2008a. Mikrovklyucheniya mineralov v metallakh i shlakakh kak indikatory mineral’no-syr’evoi bazy drevnikh obshchestv, in A.B. Derevyanko, N.A. Makarov (eds.), Trudy II(VXIII) Vserossiiskogo arkheologicheskogo s’ezda v Suzdale, V. I. Russian Academy of Sciences, Institute of Archaeology, 403-405. zaIkov, v.v., taIrov, a.d., yuMInov, a.M., churIn, e.I. and kotlyarov, v.a., 2008b. Sostav zolotykh izdelii iz kurganov Yuzhnogo Urala, in L.T. Yablonsky (ed.), Rannie kochevniki Volgo-Ural’skogo regiona. Materialy mezhdunarodno nauchno konferentsii “Rannie kochevniki uzhnogo Priural’ya v svete noveishikh arkheologicheskikh otkryti”, 21-25 aprelya 2008. Orenburg, Orenburg State Pedagogical University, 46-49. Dacian bracelets and Transylvanian gold: ancient history and modern analyses Bracelets de Dacie et or de Transylvanie : histoire ancienne et analyses modernes Bogdan Constantinescu*, Roxana Bugoi*, Viorel Cojocaru*, Martin Radtke**, homas Calligaro***, Joseph Salomon***, Laurent Pichon***, Stefan Röhrs***, Daniele Ceccato****, Ernest Oberländer-Târnoveanu*****, Corine Ionescu****** and Dana Pop******* Abstract: he recovery of the Dacian gold bracelets was the most thrilling archaeological event of the last years in Romania. he artefacts are exhibited in the Historical Treasure section of the National History Museum of Romania, Bucharest, being recovered from the international antique markets through a concerted efort of the Romanian, French, and German authorities. his paper reviews the X-Ray Fluorescence (XRF) analyses carried out on these exquisite artefacts. hese measurements followed the visual examination of qualiied experts, which concluded that the gold spiralled bracelets showed strong stylistic similarities with the Dacian silver bracelets discovered until that date in archaeological contexts. Since the most likely source – for geographical and historical reasons – for the metal used to manufacture these artefacts was the Transylvanian Apuseni mountains, additional measurements on several fragments of natural Transylvanian gold, both from placers and primary deposits, were performed using Synchrotron Radiation X-ray Fluorescence (SR-XRF) at the BESSY synchrotron, Berlin, and micro Particle Induced X-ray Emission (micro-PIXE) at the AGLAE accelerator, Paris, and at the AN2000 accelerator of LNL, Padova. he conclusion of this study was that the recovered spirals are genuine Dacian artefacts, made by cold hammering, chiselling and punching Transylvanian unreined gold – gold panned from riverbeds mixed with primary gold, most likely extracted from surface veins. Résumé : La saisie des bracelets en or de Dacie a été un des événements archéologiques les plus remarquables de ces dernières années en Roumanie. Les objets sont exposés dans le Trésor Historique du Musée National d’Histoire de Roumanie, Bucarest, et ont été récupérés du marché international d’antiquités grâce aux eforts concertés des autorités Roumaines, Françaises et Allemandes. Cet article fait le bilan de l’analyse de ces objets exquis, réalisée par luorescence à rayons X (FX). Ces mesures suivent l’examen visuel par des experts qualiiés qui ont conclu que les bracelets en or à spirales montrent des similitudes stylistiques fortes avec les bracelets en argent de Dacie découverts en contexte archéologique. Comme pour des raisons historiques et géographiques la source la plus probable du métal utilisé pour la fabrication de ces objets se trouve dans les Carpates transylvaniens, des mesures complémentaires ont été réalisées sur diférents fragments d’or natif de Transylvanie, provenant tant de placers comme de * Horia Hulubei National Institute for Nuclear Physics and Engineering – Bucharest 077125, Romania. ** BAM Federal Institute for Materials Research and Testing – Berlin D-12489, Germany. *** Laboratoire de Recherche et de Restauration des Musées de France, CNRS UMR 171 – Palais du Louvre, Paris cedex 01, France. **** Laboratori Nazionali di Legnaro, Istituto Nazionale di Fisica Nucleare – Padova 35020, Italy. ***** National History Museum of Romania – Bucharest 030026, Romania. ****** Department of Mineralogy, Babeş-Bolyai University – Cluj-Napoca 400084, Romania. ******* Museum of Mineralogy, Babeş-Bolyai University – Cluj-Napoca 400084, Romania. rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 221-225 222 Bogdan CONSTANTINESCU et al. gisements primaires, par luorescence de rayons X avec radiation de synchrotron (SR-XRF) au synchrotron BESSY, à Berlin, et par émission de rayons X induite par particules chargées (micro-PIXE) à l’accélérateur AGLAE, à Paris, et à l’accélérateur AN2000 du LNL à Padoue. L’étude à permis de conclure que les bracelets en spirale récupérés sont des objets Daciens authentiques, produits par martelage à froid, ciselure et poinçonnage d’or de Transylvanie non ainé – de l’or de rivière obtenu par batée et mélangé avec de l’or primaire, certainement extrait de ilons de surface. Keywords: gold, authentication, provenance, XRF, micro-PIXE, micro SR-XRF Mots-clés : or, authentiication, provenance, FX, micro-PIXE, micro SR-XRF. 1. IntroductIon he eleven Dacian gold bracelets recovered until the summer of 2009 are multi-spiralled (5-7 spirals), and weigh from 682 g up to 1196 g. he spirals feature at both ends zoomorphic protomes, most likely representations of snakeheads, and stylized palmettes (Fig. 1). he artefacts were found by looting an UNESCO World Heritage archaeological site – Sarmizegetusa Regia, located in the Carpathian Mountains (nowadays Romania), the capital of the Dacian Kingdom during the period between the 1st century BC and the 1st century AD (Daicoviciu and Daicoviciu, 1963). he poachers resorted to illegal metal detectors, the bracelets being found buried in several places (most of them being intentionally specially built hideouts) near the sacred precincts of Sarmizegetusa. he artefacts were illegally exported and sold on the antiquity market, being recovered through the concerted eforts of the Romanian, French and German authorities, and with the involvement of the INTERPOL (Lazăr et al., 2008). To asses their authenticity, the bracelets were visually examined by a group of experts, who decided on stylistic grounds that the artefacts present strong similarities with the Dacian silver bracelets discovered until then in archaeological contexts, and which are exhibited in (or stored in the deposits of ) several Romanian and foreign museums. By studying the marks left on the surfaces of the bracelets, it was concluded that the manufacturing techniques used for the spirals were cold hammering, punching and chiselling (Lazăr et al., 2008). Due to their special shape, decoration and heavy weight, and, not least, to the material they were made of – the fascinating and powerful gold – it was suggested that these artefacts had a tremendous signiicance for Dacian society, most likely being ritual objects and/or power insignia. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 221-225 Figure 1: (See colour plate) Bracelet no. 3. Figure 1 : (Voir planche couleur) Bracelet référence 3. Since the Romanian judicial bodies that investigated the case of poaching an UNESCO World Heritage site required further arguments supporting the authenticity of the bracelets, a compositional analysis of the gold alloy of the spirals was requested. Because only local, non-destructive and noninvasive analyses were allowed, XRF was chosen to provide the details regarding the elemental composition of the artefacts. Taking into account the wealth of ancient gold artefacts unearthed on the Romanian territory, an extensive study of Transylvanian native gold fragments was performed as well, since this geographical area is known to have been a source for gold exploitation since prehistory (Hauptmann et al., 1995). hese investigations started with the gold sources – primary and secondary – located closest to the places where the artefacts were found (results reported in Bugoi et al., 2008), and continued with other Transylvanian gold sources (see Neacşu et al., 2009). 223 Dacian bracelets and Transylvanian gold: ancient history and modern analyses 2. Methods he XRF compositional analysis of the bracelets was performed using a setup based on a 241Am (30 mCi) annular radioactive source and a Si(Li) horizontal detector – see the XRF spectrum shown in Figure 2. The micro-PIXE measurements carried out on Transylvanian native gold samples were performed both in air with a 3 MeV proton micro-beam (50 × 50 µm2) at the AGLAE accelerator in Paris, and in vacuum with a 2 MeV proton beam (4 × 4 µm2) at the AN2000 accelerator of the LNL, INFN, Padova. he micro SR-XRF analyses of another set of Transylvanian geological gold samples were performed at the BESSY synchrotron, Berlin, in air, the characteristic X-rays being excited with a 34 keV monochromatic X-ray beam (100 × 200 µm2). he detailed description of the employed experimental setups, as well as the previously reported results, can be found in Bugoi et al. (2008) and Constantinescu et al. (2008). 3. results and dIscussIon he results of the XRF analyses obtained for each bracelet are provided in Table 1. It can be easily observed that the spirals contain large amounts of silver (11% on average) and very small amounts of copper (0.9% on average). his compositional pattern is consistent with the one of natural gold, which contains up to 40% silver and up to 1% copper (Guerra, 2000; Tylecote, 1987). Obviously, there was no intention to reine the gold employed to manufacture these artefacts. Presumably, different gold ingots were used to manufacture the bracelets, an aspect suggested by the slight diferences in the compositions of the eleven artefacts. Traces of Sn were observed in the XRF spectra – see Figure 2 and Table 1. Most likely, the Sn originates from the irst melting in reducing atmosphere of alluvial gold, in which cassiterite (SnO2) grains were present (Dube, 2006; Guerra and Calligaro, 2003). he composition of bracelet no. 8, with the highest Sn and Cu content of all the investigated armbands, points towards a hypothesised tin-bronze alloying of the gold ingot used to manufacture this object (Dube, 2006). Ca, Ti, Mn, Fe, and Ba traces, most likely from soil contamination (the artefacts were supposedly buried for two thousand years), were detected on almost all bracelet surfaces. Sb traces were also detected in some bracelets, pointing towards the use of a primary deposit gold source. Figure 2: XRF spectrum of bracelet no. 11. Figure 2 : Spectre FX du bracelet référence 11. Bracelet no. 1 2 3 4 5 6 7 8 9 10 11 Au (wt%) 89.8 78.2 82.4 91.5 92.8 92.0 92.9 85.0 87.1 87.1 86.1 Ag (wt%) 9.5 20.3 16.2 8.1 6.9 7.1 6.3 12.8 12.2 11.8 12.6 Cu (wt%) 0.6 1.5 1.4 0.4 0.3 0.9 0.7 2.1 0.6 0.7 0.7 Sn (mg/kg) 200 60 360 125 < MDL 230 < MDL 1500 120 425 400 Table 1: XRF concentrations of the Dacian gold bracelets (the overall relative uncertainty is estimated to be ~1% for Au concentrations, ~10% for Ag and Cu concentrations and ~20% for Sn concentrations). Tableau 1 : Compositions obtenues par XRF des bracelets en or Daciens (incertitudes relatives moyennes sont estimées à ~1% pour les teneurs en Au, ~10% pour les teneurs en Ag et en Cu et ~20% pour les teneurs en Sn). Both micro-PIXE and micro SR-XRF led to the following conclusions: Transylvanian gold, both placer and primary, contains high and variable amounts of Ag (from 8% up to 35%), low amounts of Cu (hundreds of mg/kg on average, but up to 1.5%), traces of Sn for placer gold (500 mg/kg on average), and traces of Sb (up to 900 mg/kg) and Te (up to 3000 mg/kg) for primary gold. hese results are in agreement with the geochemical data for Transylvanian gold from Berbeleac (1985) and Tămaş et al. (2006). hey also conirm the previously published data, showing that Transylvanian primary gold is characterized by relatively large Ag amounts (more than 10% on average, but more often 20% or even 30%) (Tylecote, 1987; Hauptmann ArcheoSciences, revue d’archéométrie, 33, 2009, p. 221-225 224 et al., 1995; Bugoi et al., 2008 and references therein). he amount of Ag is generally lower for the Transylvanian gold samples of alluvial origin, an aspect explained through the weathering of the gold particles, Ag and Cu being more susceptible to dissolution or leaching, depending on the pH of the surrounding environment (Dube, 2006). he native gold samples turned out to be very inhomogeneous, an aspect highlighted by employing the scanning option of these micro-analytical techniques. From a methodological point of view, it is worth mentioning that the sensitivity of micro SR-XRF is signiicantly superior to the one of micro-PIXE, especially for Sn detection. he Cu content of the bracelets was slightly higher than the one of the natural gold samples. his was most likely due to the melting of the accompanying gold minerals, which leads to an ‘enrichment’ in terms of Cu content in the inite objects (Hauptmann et al., 1995). By comparing the XRF results of the bracelets with the micro-PIXE and micro SR-XRF results of Transylvanian gold samples, the determined trace elements and the overall compositional pattern, it was concluded that, most likely, the Dacian gold bracelets were made from unreined Transylvanian native gold – natural panned gold (gold that has not been deliberately alloyed or puriied) mixed with primary gold, using a primitive metallurgy. hus, in addition to the stylistic and historical arguments, the XRF analyses further supported the authenticity of the Dacian bracelets. 4. conclusIons he recovery and investigation of the Dacian gold bracelets is a good example of the close cooperation between the judicial and police bodies, archaeologists, cultural heritage experts, physicists and geologists. Despite their turbulent recent history, these gorgeous artefacts inally found their peace, allowing themselves to be admired in their shining splendour in the showcases of the Historical Treasury of the National History Museum of Romania in Bucharest. he elemental analyses carried out by means of X-ray methods played an important role in the authentication of these artefacts. Taking into account the location where the bracelets were found, the historical arguments about the Dacian civilization, and, not least, the results of the elemental analyses undertaken both on natural gold samples and on the recovered artefacts, it was concluded that, most likely, natural panned gold (gold that has not been deliberately alloyed or puriied) mixed with primary gold from Transylvania was used to manufacture the Dacian spiralled bracelets. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 221-225 Bogdan CONSTANTINESCU et al. Acknowledgements he inancial support of the EU-ARTECH project (allowing access to the AGLAE accelerator, Paris), of the EU FP6 contract EURONS no. RII3-CT-2004-506065 (granting access to the AN2000 LNL accelerator, Padova) and of EU FP6 contract no. RII3-CT-2004-506008 (funding the experiment at the BESSY synchrotron, Berlin) are gratefully acknowledged. References BerBeleac, I., 1985. Zăcăminte de aur. Bucureşti, Editura Tehnică (in Romanian). BugoI, r., cojocaru, V., constantInescu, B., callIgaro, t., PIchon, l., röhrs, s. and saloMon, J., 2008. Compositional studies on Transylvanian gold nuggets: Advantages and limitations of PIXE-PIGE analysis. Nuclear Instruments and Methods in Physics Research B 266(10): 2316 -2319. constantInescu, B., BugoI, r., cojocaru, V., radtke, M., callIgaro, t., saloMon, j., PIchon, l., röhrs, s., ceccato, d. and oBerländer-târnoVeanu, E., 2008. Micro SR-XRF and micro-PIXE studies for archaeological gold identiication – the case of Carpathian (Transylvanian) gold and of Dacian bracelets. Nuclear Instruments and Methods in Physics Research B 266(10): 2325-2328. daIcoVIcIu, c. and daIcoVIcIu, H., 1963. Sarmizegethusa: les citadelles et les agglomerations daciques des Monts d’Oraştie. Bucureşti, Ed. Meridiane (in French). DuBe, R.K., 2006. Interrelation between gold and tin: A historical perspective. Gold Bulletin 39(3): 103-113. Guerra, M.F., 2000. he study of the characterisation and provenance of coins and other metal work using XRF, PIXE and Activation Analysis, in D.C. Creagh, D.A. Bradley (eds), Radiation in Art and Archeometry. Amsterdam: Elsevier, 378-416. guerra, M.F. and callIgaro, T., 2003. Gold cultural heritage objects: a review of studies of provenance and manufacturing technologies. Measurement Science and Technology 14: 15271537. hauPtMann, a., rehren, th. and PernIcka, E., 1995. he composition of gold from the ancient mining district of Verespatak/ Roşia Montană, România, in G. Morteani, J.P. Northover (eds.), Prehistoric Gold in Europe – Mines, Metallurgy and Manufacture. Berlin, Springer Verlag, 369-381. lazăr, a., dePPert-lIPPItz, B. and FerrI, P.G. (eds.), 2008. Combating the criminality against the European archaeological heritage. Bucureşti, Lumina Lex Publishing House. neacşu, a., PoPescu, g.c., constantInescu, B., VasIlescu, a. and ceccato, d., 2009. he geochemical signature of native gold from Roşia Montană and Musariu ore deposits, Metaliferi Dacian bracelets and Transylvanian gold: ancient history and modern analyses Mts. (Romania); Preliminary Data. Carpathian Journal of Earth and Environmental Sciences 4(1): 49-59. tăMaş, c., BaIlly, l., ghergarI, l., o’connor, g. and MInuţ, a., 2006. New occurrences of tellurides and argyrodite in Roşia Montană, Apuseni Mts., Romania, and their metallogenetic signiicance. he Canadian Mineralogist 44(2): 367383. 225 tylecote, r.h., 1987. he early history of metallurgy in Europe. London and New York, Longman. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 221-225 he gold belt buckle from Apahida III (Romania), 5th century AD La boucle de ceinturon en or d’Apahida III (Roumanie), Ve siècle apr. J.-C. Rodica Oanţă-Marghitu *, Gheorghe Niculescu **, Doina Şeclăman *, Roxana Bugoi *** and Migdonia Georgescu ** Abstract: his study focuses on the understanding of the elaborate construction of a cloisonné gold belt buckle, found in a princely grave from Apahida, dated to the 5th century AD. Starting from the careful analysis of the object’s proportions and dimensions, a geometrical pattern of the belt plate construction and decoration design is proposed. he intricate composition of this complicated cloisonné adorned object is also illustrated with the exploded view of its structure. X-Ray Fluorescence (XRF) analyses of the composition, performed on diferent elements forming the buckle, led to conclusions regarding the relations between the variation of gold purity and the role played by diferent elements in the structure of the object. Some sequences of actions from the entire chaîne opératoire are also emphasized. Most likely, there was a certain degree of standardization in the workshops producing these cloisonné objects, and probably the craftsmen had at their disposal large batches of individual constructive elements – cell walls and gems. Résumé : Cette étude se propose de décrire la construction élaborée de la boucle de ceinture en or et cloisonné, trouvée dans la tombe princière d’Apahida et datée du Ve siècle de notre ère. L’analyse minutieuse des proportions et dimensions de l’objet, a permis de proposer un motif géométrique, base de la construction de la plaque-boucle et du dessin de décoration. La composition très élaborée de cet objet de parure au cloisonné complexe est aussi illustrée par une vue éclatée de sa structure. Les analyses par luorescence à rayons X (FX) de la composition des diférents éléments de la boucle, ont apporté des hypothèses sur les rapports entre la variation de la pureté de l’or et le rôle joué par les diférents éléments dans la structure de l’objet. Quelques séquences de la chaîne opératoire ont aussi été mises en évidence. Il est probable qu’un certain degré de standardisation avait été mis en place dans les ateliers de production de ces objets en cloisonné et que l’artisan avait à sa disposition de grands lots d’éléments individuels de construction – gemmes à sertir et cloisons. Keywords: gold, cloisonné, Apahida, 5th century AD, XRF. Mots-clés : or, cloisonné, Apahida, ve siècle AD, FX 1. IntroductIon A large number of cloisonné adornments dated to the 5th-6th century AD were discovered by archaeological research and/or by chance on the former territory of the Roman Empire and also beyond its boundaries. Among all these artefacts, a group of items crafted in gold, inlaid with garnets on all visible sides, with S and Ω-shaped cells, undulated (‘stepped’) cell walls, and bordered with a string of small cabochon garnets can be distinguished (Werner, * National History Museum of Romania, Bucharest 030026, Romania – 1-97. (rodicamarghitu@yahoo.com), (tezaur@mnir.ro) ** National Institute for Research, Conservation and Restoration of the Cultural Heritage, Bucharest 030026, Romania. *** “Horia Hulubei” National Institute for Nuclear Physics and Engineering, Bucharest 077125, Romania. rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 227-233 228 1958; Carnap-Bornheim, 1995; Schmauder, 2002: inds lists nos. 2, 3, 4). It was suggested that these spectacular pieces of jewellery, manufactured with special attention to every single detail, were most likely produced in the most prestigious workshops of the Roman Empire (Arrhenius, 1985: 100-124). he occasional presence of such exquisite ornaments deep in barbarian territory was explained as a result of diplomatic relations between the imperial court and the barbarian elite (Harhoiu, 1998: 155). To date, due to their special value and intricate structure, none of these objects was thoroughly investigated. he present paper tries to cover this gap by studying a gold belt buckle inlaid with garnets found at Apahida. hree princely graves, all dating from the second half of the 5th century AD, were discovered at Apahida, near Cluj, the ancient Napoca (Finály, 1889; Horedt and Protase, 1972; Matei, 1982; Harhoiu, 1998: 157-158; Périn et al., 2000, 172-191). hese inds made Apahida one of the most important sites on the map of the Migration Period in Europe. he impressive number of garnet inlaid objects that were preserved and the close resemblance with the adornments found in the Childeric grave from Tournai (Böhner 1981; Périn and Kazanski, 1996) emphasize its signiicance even more. he existence of a third grave was deduced from the discovery of a gold belt buckle (Fig. 1) in 1979 (Matei, 1982)1. After its discovery and before being inally taken in custody by the National Bank of Romania, the buckle was dismantled, all garnets being extracted from the cells and the illing material completely removed. It is the unfortunate discovery conditions and the history of the artefact immediately after its recovery that favoured our investigation; this approach would have been impossible on a better preserved object. In 2001, the National Bank of Romania donated the gold belt buckle from Apahida III to the National History Museum of Romania. At that moment the artefact was composed of 130 distinct elements: the buckle with a belt plate and a detachable tongue, all these ornamented with a total of 120 empty cells (Fig. 1, a-h); 34 gold fragments – thin patterned foils and cell walls (Fig. 1i); 56 garnets – among them, we can ind both whole and broken stones, lat cut gems and oblong, slightly rounded garnets (Fig. 1j); 38 semispherical garnets – of which 26 are larger, initially mounted on the belt plate border, and 12 are smaller, initially mounted on the border of the tongue plate (kidney shaped) (Fig. 1k). 1. According to Prof. Dumitru Protase and Dr. Ioan Stanciu (personal communications), it is quite likely that this buckle is also part of the second Apahida grave inventory. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 227-233 Rodica Oanţă-Marghitu et al. 2. PremIses and methods As already mentioned above, this study focuses on the understanding of the elaborate construction of the cloisonné gold buckle found in the Apahida III grave. he poor conservation state of this artefact represents an advantage for the investigation, ofering the opportunity for a better understanding of its construction. By carefully examining the buckle, we noticed its extremely elaborate structure. Consequently, we concluded that the only possible way to assemble it was starting from an initial geometrically constructed project. By analyzing the ratios between diferent dimensions of the buckle, we found a solution for the construction of the buckle (Fig. 2a). In order to obtain additional information on the manufacturing technologies, it was decided to perform X-Ray Fluorescence (XRF) measurements on as many as possible constructive elements of the belt buckle. Prior to the analysis, all gold items were cleaned in an ultrasonic bath. he preliminary XRF analyses were performed using an Innov-X Systems Alpha Series X-Ray Fluorescence portable spectrometer with an X-ray tube based on a W anode, operated at 35 kV and 40 µA. Due to its technical characteristics, we used this equipment only for the investigation of the metallic elements larger than the sampling window (17 mm diameter) of the measuring head, or for the analysis of the smaller and detached elements, such as the patterned foils. hese XRF measurements provided a rough estimate of the elemental composition for a number of gold components and stand-alone fragments. he results of these irst XRF measurements on Apahida items were reported elsewhere (Niculescu et al., in press). However, due to the complicated structure of the object, it was necessary to reine the XRF information, especially for the areas that were diicult to reach using the portable spectrometer. herefore, it was decided to perform a new set of analyses on diferent points from each metallic element composing the buckle – including the thin cell walls. hese measurements were performed using an ARTAX 400 microXRF spectrometer, with an X-ray tube with a W anode, operated at 50 kV and 700 µA, with a 650 µm diameter collimator. he quantiication of the XRF measurements was performed through direct comparison of the count rates method and by using standards with known elemental concentrations, measured with exactly the same spectrometer settings as the investigated archaeological items. he standards were ternary Au-Ag-Cu alloys from the former 229 he gold belt buckle from Apahida III (Romania), 5th century AD Figure 1: (See colour plate) Photographs showing the Apahida III belt buckle, as well as the related (and dismantled) parts. he black dots on the photographs indicate the points in which the XRF analyses were performed; the quantitative results are summarized in Figure 3. Figure 1 : (Voir planche couleur) Photos illustrant la boucle de ceinture d’Apahida III ainsi que les parties en relation (et démontées). Les points noirs sur les photos pointent les régions où les analyses par FX ont été efectuées ; les résultats quantitatifs sont résumés sur la Figure 3. Assay Oice of the National Bank of Romania2 that are routinely used for touchstone testing of gold jewelleries, and in which the Au content was determined by the ire assay method. For the quantiication of Ag, the K X-ray lines of this element were employed. In this study, there was no special interest for the determination of trace elements that might have been contained in the gold alloy of the investigated artefacts; consequently, at this stage of the research, the interpretation of the results obtained was based only on the content of the major elements (Au, Ag and Cu). he relative standard deviations of the determined concentrations were less than 0.2% for Au and less than 2% for Ag and Cu. 2. Functioning nowadays as the Precious Metals and Precious Stones Department of the National Authority for Consumer Protection, Romania. he ternary diagram (Au-Ag-Cu) in Figure 3 summarizes all concentrations (expressed in weight percent) obtained with the ARTAX spectrometer on diferent metallic parts of the buckle. 3. results and dIscussIon Although the object we are examining today is only one buckle, we should not neglect the fact that the inite object was a belt, not a buckle. he obverse of the belt plate extends to a rectangular frame which connects the three elements of the buckle: the frame is bent around the link and welded on the back to a plain golden sheet having a shape identical to that of the belt plate. he hook of the tongue passes through the hole of the frame and is fastened to the ring. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 227-233 230 Rodica Oanţă-Marghitu et al. Figure 2: a. Geometrical construction of the decorative pattern; b. Reconstructed, ideal aspect of the buckle plate; c-d. Exploded view of the buckle plate. Figure 2 : a. Construction géométrique du motif de décoration ; b. Reconstruction, aspect idéal de la plaque-boucle ; c-d. Vue éclatée de la plaque-boucle. he distance between the reverse of the belt plate and the golden plain sheet was about 1-2 mm, corresponding to the belt’s thickness. All these three ‘layers’ – the belt plate, the leather (or a type of textile material), and the golden foil on the reverse – were connected and fastened with four rivets – nowadays lost – whose hemispherical heads were integrated into the decoration and visible from the obverse. he same ArcheoSciences, revue d’archéométrie, 33, 2009, p. 227-233 system could be observed on similar buckles preserved in the inventories of the irst and second Apahida graves (Périn et al., 2000: 176, 188). he belt plate was assembled from more than 100 gold fragments and 74 garnets; the cells were initially illed with a kind of putty. By analogy with other buckles of the same category, we can assume that the quatrefoil and the round 231 he gold belt buckle from Apahida III (Romania), 5th century AD group. A separate group corresponds to the S-shaped cell walls of the tongue. he obverse and the reverse golden thick foils of the buckle plate are clustered in the fourth group, while all the interior and edge cell walls of the buckle plate are found in the ifth group. hree isolated points correspond to the measurements of the link and four others to the tongue’s hook. he same diagram suggests that perhaps for diferent sets of elements, ‘prefabricated’ components cut from diferent gold ingots were used. It is worth emphasizing that the purity of the gold used to manufacture the buckle was never less than 22.5 karats. he gold concentration derived from the XRF measurements varies between a minimum of 94.0% for the link and a maximum of 99.4% for one of the patterned foils. he very high ineness of the patterned foils suggests that the goldsmith wished to take advantage of the remarkable malleability of pure gold. For the cell walls, an alloy with a lower gold content was chosen, a material which could at the same time ensure a high malleability to bend the gold straps in the appropriate shapes, as well as a certain hardness (the upper edges were always slightly bent over the stones). A possible explanation for the relatively high amount of silver and copper found in the link was that an alloy with improved mechanical properties was intentionally chosen to manufacture this component of the buckle. In particular, a constructive material with increased hardness makes this mobile element more resistant to stress and wearing. Taking into account that the metallic composition of the cell walls depends on their shape, but does not show considerable diferences, we could infer that the goldsmiths determined an optimal gold title for these elements. hey prepared a certain number of similar cell walls and used Uncia (1/12 pes ≈ 24.7 mm) Digitus (1/16 pes ≈ 18.5 mm) Diameter of the quatrefoil cell 12 1/24 1/12 2/3 Actual length (buckle + belt plate) 72 ≤1/4 ≤3 ≤4 Large diameter of the buckle 57 1/5 Length of the belt plate 54 ≈ 2/11=1/5.5 ≈ 11/5 <3 Width of the belt plate 37 1/8 3/2 2 Diameter of the large hemispherical garnet 3 1/96 1/8 1/6 Dimensions (Fig. 1) Table 1: he dimensions of the Apahida III belt buckle. Tableau 1 : Les dimensions de la boucle de ceinture d’Apahida III. Roman system Pes (296 mm) Metric system (mm) central cells were not inlaid with garnets. he graphical reconstruction in Figure 2, c-d, may give an idea about the complicated structure of this object. By examining the shape of the belt plate and the geometrical composition of the cloisonné panel, it can be noticed how the peculiar kidney contour was created (Fig. 2a). hree corners of a square inscribed in a circle represent the centres of three other circles, joined to compose the kidney contour. he round prominences on the edge are evenly distributed at 18° intervals. he careful harmonious design of the buckle is also suggested by its proportions, which it with a remarkable precision into the Roman measurement system (see Table 1) (Wikander, 2008: 766-767). Illustrative of this aspect, the cloisonné decoration of the belt plate is built around a quatrefoil cell with a length of 12.3 mm, which corresponds to ½ uncia. In order to craft the cloisonné adorned jewellery, the goldsmiths had to assemble a lot of similar entities, like cell walls, gold patterned foils or garnets of diferent shapes. hus, it is quite likely that before connecting all these elements in one single object, the craftsmen had to prepare all the necessary components, i.e. a complete set of S-shaped, Ω-shaped, undulated or simple rectangular cell walls. Signiicant data about the crafting process of the buckle, especially regarding the preparation of all the elements to be assembled, were inferred from the XRF analysis of its components. In the diagram in Figure 3, one can notice several clusters, most likely corresponding to elements cut from the same gold ingot. Gold patterned foils can be found in one group, while the three longitudinal walls of the tongue, the Ω-shaped cell walls (used on the edge of the tongue kidney plate), as well as a patterned foil, are clustered in another Actual thickness of the belt plate 7,7 Probable total thickness of the mounted belt plate 8,2 1/36 1/3 4/9 9,3 1/32 3/8 1/2 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 227-233 232 Rodica Oanţă-Marghitu et al. procedure with a precise sequence of operations, based on the exact knowledge of the required amount of raw materials and also of the characteristics (shapes and numbers) of the components (gold and garnets) to be assembled. 4. conclusIons Figure 3: Ternary diagram (Au-Ag-Cu) of the concentrations (expressed in weight percent) determined by XRF on the Apahida III belt buckle. Figure 3 : Diagramme ternaire (Au-Ag-Cu) des concentrations (exprimées en pourcent) déterminées par FX de la boucle de ceinturon d’Apahida III. them to assemble diferent objects, depending on the requirements of the decoration design. hey could also have used old, recuperated metal. Most likely, the available gold pieces were separated in diferent groups, according to their title, and used afterwards when an alloy of a certain type/purity was needed. In any case, we can assume the existence of an important degree of standardization in the workshop/ workshops where adornments similar to this Apahida III buckle were produced. hus, it is possible that large stocks of metallic elements, such as cell walls, cut into diferent shapes and sizes, were readily available at the actual moment of manufacturing the objects. Recently, some test measurements on the garnets belonging to the Apahida belt buckle were also performed with the ARTAX spectrometer, this time in a helium-lushed atmosphere. Taking into account the results of these additional XRF analyses, as well as the previously published data on Merovingian gems (Calligaro et al., 2002), the irst conclusion was that the garnets used to decorate the Apahida III buckle were of diferent types, and, most likely, originated from diferent sources. he interpretation of the XRF results obtained on the garnets is still in progress, and further measurements are to be carried out. Our investigation points out that the efective construction of the buckle was only the last phase of a much longer endeavour. he assembling process implied a well deined ArcheoSciences, revue d’archéométrie, 33, 2009, p. 227-233 his paper summarizes a thorough study of the Apahida III belt buckle, which was recovered from a 5th century AD princely grave. Starting from the object’s proportions, a geometrical pattern of the belt plate construction and decoration design was proposed. he buckle’s appearance was irst ideally conceptualized, and then a reference model was geometrically designed. By analyzing sets of several similar constructive elements, we gained a better understanding of the assembling process, underlining the stages of transforming an ideal project into a concrete object. It is highly possible that the workshop where the gold belt buckle from Apahida III was crafted employed a certain standardization of the objects, assembling components with similar shapes to obtain diferent adornments and diferent decorative patterns. It is quite likely that, in order to produce new objects of this kind, recuperated metal was often used. he hypothesis of the use of ‘prefabricated’ elements – such as cell walls and garnets, cut into diferent shapes and sizes – to assemble cloisonné jewelleries of this kind appears reasonable. he main goal of the craftsmen was the inal aspect of the object, which could include elements of variable gold purity, as it was demonstrated by the XRF analyses. In the structure of such elaborated objects as the garnet inlaid adornments of Apahida, the manufacturing process required not only a high technical background, but also a particular specialization of the craftsman and appropriate skills. Finally, the value of the object largely exceeded that of the gold and gems built in it, the intrinsic material value blending with the much more important symbolic one. Acknowledgements he inancial support of the ROMARCHAEOMET project in performing this study is kindly acknowledged. We are grateful to Prof. Dumitru Protase and Dr. Ioan Stanciu for the information on the discovery of the Apahida III belt buckle. We also thank Mr. Marius Amarie for taking the photos of the Apahida III gold belt buckle and related parts. he gold belt buckle from Apahida III (Romania), 5th century AD References arrhenIus, B., 1985. Merovingian Garnet Jewellery. Emergence and Social Implications. Stockholm, Almqvist & Wiksell International. Böhner, K., 1981. Childeric von Tournai. Reallexikon der germanischen Altertumskunde, Band D, Lief. 3/4: 441-460. CallIgaro, t., colInart, s., PoIrot, J.-P. and sundres, C., 2002.Combined external-beam PIXE and µ-Raman characterization of garnets used in Merovingian jewellery. Nuclear Instruments and Methods in Physics Research B 189: 320-327. carnaP-BornheIm, c. von, 1995. Eine cloisonnierte Schnalle mit wabenförmigem Zellenwerk und Almandinrundeln aus Olbia. Germania 73(1): 151-155. FInály, H., 1889. Az Apahidai lelet. Archeologiai Értesitö 9: 305320. harhoIu, r., 1998. Die frühe Völkerwanderungszeit in Rumänien. Archaeologia Romanica I. Bucureşti, Editura Enciclopedică. Horedt, K. and Protase, d., 1972. Das Zweite Fürstengrab von Apahida (Siebenbürgen). Germania 50(1-2): 174-220. mateI, Ş., 1982. Al treilea mormânt princiar de la Apahida. Acta Musei Napocensis 19: 387-392. nIculescu, g., oanţă-marghItu, r. and georgescu, M., in press. On the Gold Adornments from Apahida – 5th 233 Century AD, Transylvania, Romania, in Proceedings of the 37th International Symposium of Archaeometry (ISA 2008), Siena, Italy, 12-16 May 2008. PérIn, P. and KazansKI, m., 1996. Das Grab Childerichs I, in A. Wieczorek, P. Périn, K. v. Welck, W. Menghin (eds.), Die Franken Wegbereiter Europas. Vor 1500 Jahren: König Chlodwig und seine Erben. Mainz: Verlag Philipp von Zabern; Mannheim: Reiss-Museum, 173-182. PérIn, P., Vallet, F. and KazansKI, m. (eds.), 2000. L’or des princes barbares. Du Caucase à la Gaule ve siècle après J.-C. Paris, Éditions de la Réunion des Musées Nationaux. schmauder, m., 2002. Oberschichtgräber und Verwahrfunde in Südosteuropa im 4. und 5. Jahrhudert n. Chr. Zum Verhältnis zwischen spätantiken Reich und barbarischer Oberschicht aufgrund der archäologischen Quellen. Archaeologia Romanica 3. Bucureşti: Editura Academiei Române. Werner, J., 1958. Eine ostgotische Prunkschnalle von KölnSeverinstor. Kölner Jahrbuch für Vor- und Frühgeschichte 3: 55-61. WIKander, c., 2008. Weights and Measures, in J.P. Oleson (ed.), he Oxford Handbook of Engineering and Technology in the Classical World. New York, Oxford University Press, 759-769. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 227-233 ‘More like the work of fairies than of human beings’: the filigree on the ‘Tara’ brooch, a masterpiece of late Celtic metalwork « Œuvre des fées plus qu’œuvre des hommes » : le filigrane de la fibule de « Tara », un chef d’oeuvre de l’orfèvrerie celtique tardive Niamh Whitfield* Abstract: In the early Middle Ages, Irish goldsmiths achieved a rare mastery of gold working techniques. he ‘Tara’ brooch, dated to c. 700 AD, shows many characteristics of this late Celtic style of metalwork. his paper discusses the manufacture of its iligree, drawing on the writer’s microscopic study and hitherto unpublished scientiic investigations carried out in the British Museum in 1962-1963. Résumé : Au Haut Moyen Âge les orfèvres irlandais ont réussi un niveau technique rare dans les arts du métal. La ibule de ‘Tara’, fabriquée vers 700 de notre ère, présente un grand nombre de traits distinctifs de ce style celtique tardif. Cet article décrit la fabrication des iligranes de cette ibule. L’étude est basée sur des examens au microscope optique par l’auteur, et aussi sur des études scientiiques inédites faites au British Museum en 1962-1963. Keywords: Beaded wire, Celtic, iligree, granules, Ireland, ‘jewellers’ stitches’. Mots-clés : Fil perlé, Celtique, iligrane, granules, Irlande, millegrifes. 1. INTRODUCTION Although the art of goldsmithing, including the manufacture of gold iligree, lourished in prehistoric Ireland, the working of gold seems to have been completely forgotten there in the early centuries of our era. However, in the 6th and 7th centuries AD, the Irish, through their missionary activities, forged strong contacts with rich and powerful Anglo-Saxon, Frankish and Lombardic leaders. One result of these contacts was the reintroduction of gold and simultaneously, it seems, of gold iligree. he inluence of Germanic, and in particular Anglo-Saxon, traditions is clearly seen in the new so-called ‘Hiberno-Saxon’ (or ‘late Celtic’) iligree style they developed (Whitield, 1987; 2007). However, the Irish goldsmiths also introduced innovations, and developed the craft to an exceptionally high degree of sophistication and virtuosity. he ‘Tara’ brooch (Fig. 1), which dates to the late 7th or very early 8th century AD, is one of the earliest pieces of Irish metalwork to display this new style of iligree. he brooch was discovered in 1850 at Bettystown, Co Meath, on the east coast of Ireland, rather than at the nearby, ancient necropolis of Tara. Its irst owners, * 47 Faroe Road, London W14 0EL, UK. (mail@niamhwhitield.com) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 235-241 236 Niamh WHITFIELD Figure 1: (See colour plate) ‘Tara’ brooch iligree: a. Pin-head; b. Part of the brooch; c. Diagram showing attachment of iligree disc to amber stud; d. Amber stud on right side. (Photography: a, b, d: reproduced by the kind permission of the National Museum of Ireland; Drawing: M. Bimson). Figure 1 : (Voir planche couleur) Filigranes de la ibule de « Tara » : a) Tête de l’ardillon ; b) Une partie de la ibule ; c) Schéma de la monture du disque orné de iligranes sur une perle d’ambre ; d) Perle d’ambre à droite. (Photographies : a, b, d : reproduction avec la permission du Musée national d’Irlande ; dessin : M. Bimson). Waterhouse and Co., jewellers of Dame Street, Dublin, were in the habit of attaching evocative names to brooches of which they sold replicas: they are responsible for the misleading reference to Tara in the ‘name’ given to the brooch. he object was almost completely intact when discovered, but some iligree was subsequently lost. It measures 8.6 cm in diameter and weighs 224.36 g. Although it is cast in silver, the underlying silver is barely visible, because it is elaborately decorated on all faces with a profusion of designs and materials, either inset into the brooch or directly applied on its surface. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 235-241 his paper is concerned only with the manufacture of its iligree, all of which is made of gold. he writer, an archaeologist, not a scientist, has not been able to carry out any scientiic analyses. his paper draws primarily on her microscopic examination of the object in the National Museum of Ireland. However, when the ‘Tara’ brooch was cleaned in the British Museum Laboratory in the early 1960s, it was closely examined using the scientiic methods then available. Robert Organ has kindly made available an unpublished lecture in which he presented the new discoveries made at that time (Organ, 1964), and access has also been given to other iles ‘More like the work of fairies than of human beings’: the iligree on the ‘Tara’ brooch… 237 in the British Museum. In addition, some excellent black and white photographs of the ‘Tara’ brooch taken in the British Museum in the 1960s, not hitherto published in the specialist literature, are reproduced here. With the exception of some lattened gold beaded wire as rings around the three rivets on the back, gold iligree is found only on the front of the ‘Tara’ brooch. here, it was never applied directly. Instead, it was soldered to small backplates made of thin gold sheet to form individual panels, each edged by a iligree border and decorated with an allover pattern ‘drawn’ with wires and granules (Figs. 1 and 2). he thickness of the gold sheet was not measured, but a back-plate measured on the similar Hunterston brooch, from the west coast of Scotland, was found to be 0.1 mm thick (Stevenson, 1974: 23), while another, on the closely related Westness brooch, from a 9th-century Viking grave in Orkney, was 0.1-0.15 mm thick (Stevenson, 1989: 248). Nor were analyses of the solder joining the wires and granules to the back-plates carried out. However, clear traces of metallic solder were observed at some points by both the writer and by the British Museum team. 2. MOUNTING THE FILIGREE PANELS Filigree panels were ixed on the front of the brooch: – In sunken compartments about 3 mm deep, covering the front of the brooch and pin-head – twenty-eight compartments in all, separated from each other by studs of various types. – On the tops and edges of the ive dome-shaped amber studs. – On the tops of three of the glass studs, one lozengeshaped and two circular. Diferent methods were used to attach the gold panels in each of the three diferent locations, and these will now be described. In the sunken compartments After the brooch had been cast, the walls of the sunken compartments were gilded, and wires and granules made separately and soldered to small gold back-plates, which had been cut exactly to shape. hese were then itted snugly in the base of each compartment (Fig. 1a, b). How were they held in place there? When the brooch arrived in the British Museum for conservation in the 1960s, it was discovered that the gold panels were secured by three types of mastic: red wax, beeswax, and (apparently, the earliest of the three) a mixture Figure 2: ‘Tara’ brooch: four iligree panels removed from their compartments (Photography: British Museum Conservation Laboratory). Figure 2 : Fibule de « Tara » : Quatre plaques de iligranes enlevées de leur boîtes. (Photographie : Laboratoire de Conservation du British Museum). of rosin and beeswax (Organ, 1964). However, the mastic was so thick that the panels were raised above the level of features in the cell walls, which seems to have been the primary method of holding them in place. hese were ‘jewellers’ stitches’, i.e., tiny tongues of silver, gouged from the compartment wall, which projected over the edge of each back-plate and held it in place (Fig. 3). Some intact stitches survive on the ‘Tara’ brooch, but many have corroded and broken of, which partly explains the loss of the minute panels in the 19th century. Traces of this are still visible, because they pierced the gilding on the compartment walls revealing the silver beneath (Fig. 1d). he stitches were at a height from the bottom of the cells that corresponded exactly with the thickness of the iligree panels (Werner, unpublished). It was therefore concluded that originally jewellers’ stitches alone secured the iligree panels, and that the mastic was modern, applied to preArcheoSciences, revue d’archéométrie, 33, 2009, p. 235-241 Niamh WHITFIELD 238 On the three glass studs Figure 3: ‘Tara’ brooch: three ‘jewellers’ stitches’ (Photography: British Museum Conservation Laboratory). Figure 3 : Fibule de « Tara » : trois «millegrifes». (Photographie : Laboratoire de Conservation du British Museum). vent further losses; the remaining panels are now secured by Duraix (Organ 1964). Jewellers’ stitches (also known as ‘bead settings’ or ‘milligraining’) are used in modern stone setting (Untracht, 1982: 619-621), but in the early Middle Ages they seem to occur only in a late Celtic context. On the dome-shaped amber studs Filigree originally decorated each of the ive dome-shaped amber studs, in two positions: (i) on lat discs, about 3.25 mm in diameter, ixed to the apex of the studs, only two of which survive today (Fig. 1b), and (ii) as a iligree collar circling the base of the studs, the largest of which is approximately 7 mm in diameter (Fig. 1d). Scrutiny in the British Museum revealed that these were secured in the following ways: (i) In the case of the iligree on the tops of the amber studs, the latter were drilled through their centres, and each gold disc soldered to a hollow gold tube to make a mushroom-like unit. his was then passed through the drilled hole, and the tube splayed out at the lower end, anchoring the iligree-decorated disc to the top of each stud (Fig. 1c). he gold tube can still be seen, embedded in a stud that has lost its iligree disc (Fig. 1d). Moreover, when another stud was removed from its socket, scrutiny of the back revealed that only just over half the tube is intact; the rest was carefully cut away by the goldsmith, who must have had to use gold sparingly (Organ, 1964). (ii) In the case of the iligree collars, the circumference of the stud was enclosed in a ring of thin gold sheet: a decorative band of iligree wire is soldered around it (Fig. 1d). When the studs were removed from their sockets, the gold sheet could be seen to extend underneath each one, where it had been nicked and bent back under the stud to hold the sheet as a whole fast. he amber studs in their entirety appear to have been secured to the brooch itself by a rivet passing through a hole in the loor of each stud socket. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 235-241 A third method was used to attach the little iligree panels to the tops of the glass studs (Fig. 1b). Mavis Bimson of the British Museum observed that on the lozenge-shaped stud (maximum width: 12 mm), the glass slightly overlaps the iligree panel, particularly on the upper part of the lozenge. his showed that the glass must have lowed into this position. hus, the gold lozenge-shaped iligree panel was placed on the stud when the glass was molten, and stuck to it when it solidiied. his seems to have been the most secure method of all of attaching the iligree panels to the ‘Tara’ brooch, because only on the tops of the glass studs has all the original iligree survived. 3. ORNAMENTAL WIRES AND GRANULES As Figures 4 and 5 show, many ornamental forms of iligree adorn the ‘Tara’ brooch. Individual plain round wires of gold are not used decoratively. However, as indicated below, individual strands are twisted together to form wire ropes. hese strands are solid and helical marks occasionally appear, suggesting that the round wire was made by block-twisting, i.e. by twisting a gold strip of squarish or rectangular section and then rolling it between two lat surfaces to produce a smooth round-sectioned wire (Whitield, 1990: 14). he following simple iligree forms are found: (a) Beaded wire (Fig. 4a): All the beading is located at right-angles to the long axis of the strand. Although there are some laws, the beads normally follow each other in an orderly succession and gaps between them are rare. hey are generally regular in form, and a medial seam (äquatorschnitt) is often present. All wires appear to have been hand-rolled at right-angles under a rolling swage or ‘beading ile’. his tool probably had a single groove lanked by teeth, like the tool described by heophilus (Whitield, 1998: 62), but the possibility of more grooves cannot be ruled out. A number of beaded wires were measured across their broadest point. hree, and possibly four, gauges seem to have been made deliberately. In descending order, they have maximum diameters of approximately 0.35 mm, 0.25 mm, 0.2 mm, and 0.15 mm. A few measurements were made of the number of beads per ive divisions on the graticule, which equalled c. 0.9 mm. On the beaded wire measuring c. 0.25 mm in diameter there were approximately four beads per 0.9 mm, while on the beaded wire of c. 0.15 mm in diameter there were approximately six beads per 0.9 mm. his suggests that a iner tool was used to bead the latter. ‘More like the work of fairies than of human beings’: the iligree on the ‘Tara’ brooch… Figure 4: ‘Tara’ brooch: simple iligree forms (Drawing: Nick Griiths). Figure 4 : Fibule de « Tara » : Formes simples de iligrane. (Dessin: Nick Griiths). (b) Flattened beaded wire on edge: he scalloped ribbon illustrated in Figure 4b is, in fact, beaded wire which was hammered lat, and set on the edge in the manner described by heophilus (Whitield, 1998: 79). his is evident because, when these wires are examined under the microscope, it is visible that the undulations on each side not only match, but are also often linked by tiny transverse lines representing the breaks between each bead. his produces a very narrow upper surface, in this case of the order of 0.10.15 mm wide. (c) Twisted ribbon (Fig. 4c): his consists of a lat strip of gold that has been twisted by securing at one end and turning the other. he torsion is often uneven. he ribbons themselves were not measured, but they are all similarly narrow. Where measured, the thickness of the ribbon was found to be just under 0.1 mm, and the width of the twisted strand as a whole c. 2.5mm. (d) 2-ply twisted wire rope (Fig. 4d) (sometimes known as ‘rope-twist’ or ‘twined wires’), made by twisting two round wires together, either to the right or to the left. he torsion is fairly even. On some panels the individual round wires were found to be c. 0.17-0.2 mm in diameter; on others c. 0. 12 mm. (e) 2-ply twisted beaded wire rope (Fig. 4e), made by twisting two beaded wires together in the manner described above. hese wires were not measured, but the inest beaded wires seem to have been used, probably with a diameter of the order of 0.15 mm. 239 (f ) 3-ply twisted wire ropes (Fig. 4f ), a variation of (d) made by twisting three, rather than two, plain round wires to produce a thicker cord. No measurements were recorded. (g) 3-ply twisted beaded wire rope (Fig. 4g), a variant of (e) made by twisting three, rather than two, beaded wires together. he individual beaded wires appear to be of very ine gauge, but were not measured. (h) Granules (Fig. 4h), i.e. minute spheres formed when small chips of gold are melted, and upon cooling coalesce naturally into minuscule spheres which retain their shape. Some variations in granule size can be seen with the naked eye. Diameters of c. 0.15, 0.25, 0.35 and 0.45 mm were he following forms occur. Not only are there a large number of simple iligree forms on the ‘Tara’ brooch, but a range of stock combinations of various types also occur, as Figure 5 shows. Sometimes wires and/or a granule are juxtaposed; alternatively, one wire is placed on top of another; at other times yet, wires are coiled. as follows: he imitation plait (Fig. 5a) (sometimes also known as a ‘herring-bone’), formed by neatly paired 2-ply wire ropes, one twisted to the right, the other to the left, so as to form a succession of ‘V’ patterns along its length and create the illusion of plaiting. he illusion depends on extraordinary precision, particularly as the individual round wires which form the components of the ‘ropes’ may measure only c. 0.15 mm in diameter. he 3-strand band has a thicker central and iner lanking strands of beaded wire, in this case measuring c. 0.25 and 0.15 mm in diameter, respectively. Figure 6b shows a similar arrangement but with a iner central wire. he individual Figure 5: ‘Tara’ brooch: combined iligree forms (Drawing: Nick Griiths). Figure 5 : Fibule de « Tara » : Formes combinées de iligrane. (Dessin: Nick Griiths). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 235-241 240 wires are sometimes of diferent lengths and were not cut as a band. hey may have been soldered separately to the back-plate. he herring-bone band (Fig. 5c), which builds on the imitation plait by adding a beaded wire to either edge. he herring-bone band with spine (Fig. 5d), a variant on the above formed by placing a wire between the individual wire ropes in the middle of the band. One beaded wire on top of another (Fig. 5e). It seems that the two wires were soldered together before being mounted on the back-plate, because they appear to have been cut in one piece. he two wires appear to be of about the same gauge; the upper one is c. 0.25 mm in diameter. A beaded wire on a ribbon on edge (Fig. 5f ), a variant on the theme described above at (e): the uppermost beaded wire is soldered onto a lat ribbon of gold set on edge, the two components being again probably soldered together before being mounted on the back-plate. he beaded wire has a diameter of the order of 0.15-0.2 mm. he height of the ribbon was not measured, but it is relatively broad from edge to edge, so that this particular combination creates an efect comparable to empty cloisonné cells. However, study in the British Museum established that the ‘cells’ between the wires were not inilled. A conical spiral (Fig. 5g) formed of a beehive-shaped cone of beaded wire capped by a granule. It is diicult to see how many wires are involved, but similar cones on the Ardagh chalice were made by coiling up a pair of beaded wires (Organ, 1973: 258, pl. 41). A collared granule (Fig. 5h), which consists of a single granule enclosed in a ring of beaded wire, to form a dot and ring. None of the individual forms and combinations listed above is unique to the ‘Tara’ brooch: all are part of the stockin-trade of the late Celtic goldsmiths. hree compound forms (one beaded wire on top of another; a beaded wire on a ribbon on edge; and the conical spiral) seem to be Celtic innovations, but all the other forms have longer pedigrees, some originating in the Classical world (Whitield, 1987; 2007). Nevertheless, the ‘Tara’ brooch iligree has a very distinctive character. As shown above, it is remarkable for the sheer number and diversity of its components, and for the extreme delicacy of the wire. Not only is the wire extremely ine, but also the panels themselves are very small. his aspect can be illustrated by making reference to Figure 2, which shows four panels photographed together. he two long narrow panels at the top are just 4 mm wide. he panel beneath, which is the largest on the brooch, has a maximum width of 12 mm, and a chord of 20 mm along its curved side. he ArcheoSciences, revue d’archéométrie, 33, 2009, p. 235-241 Niamh WHITFIELD lowest panel is 14 mm high and has a maximum width of 8 mm. Yet, these minute panels are ornamented with a wide range of types of iligree. here is order among all these features. he front of the ‘Tara’ brooch is sub-divided into diferent zones of ornament (Fig. 1), and, technically, whatever the type of pattern displayed, the panels in each zone are treated as a unit. However, within these constraints, the goldsmith managed to avoid repeating himself to an extraordinary degree, and to create many variations on simple themes. Beaded wire is always used to ‘draw’ the motif in the panel centres, but diferent variants of it appear in the diferent ornamental zones. To take Figure 2 as an example, the top two panels are from the side-compartments on the brooch (Fig. 1b). he motifs there difer, but each is outlined by a lattened beaded wire on edge, as on all other panels in this zone. Here, the choice of wire imposes order, even if the patterns display diversity. In contrast, the outline of the interlaced animal in the sub-trapezoidal panel below (from the body of the brooch, Fig. 1b) is ‘drawn’ by one beaded wire on top of another. On the other hand, the other interlaced animal at the bottom of this photograph (which is from the centre of the pin-head, Fig. 1a) is outlined by a beaded wire on a ribbon on edge. Yet another variant is used to ‘draw’ the interlace on the hoop: a 3-strand band of beaded wire (not illustrated). It is remarkable that nowhere on the ‘Tara’ brooch is simple beaded wire alone used to draw a pattern. Simple beaded wire is reserved for panel borders, where it may be juxtaposed with twisted ribbon and twisted wire ropes of various types. hese borders show a similar virtuoso spirit. A particular ploy was to juxtapose twisted ribbon with twisted wire ropes, plain or beaded, 2-ply or 3-ply, to form a patterning reminiscent of the more traditional ‘imitation plait’. Given the diferent nature of the strands, however, the resultant ‘plait’ is slightly ‘of-key’, as the juxtaposed twists fail to match up exactly. he length to which the goldsmith was sometimes prepared to go in order to vary the efect is evident from a comparison between the two panels with animal ornament in Figure 2. On the body of the brooch (Figs. 1b, 2), the border is composed of an outer 3-ply beaded twisted wire rope, inside which is a, far iner, twisted ribbon (Fig. 2). he animal’s outline, ‘drawn’ by one beaded wire on top of another, is inilled with opposed 2-ply twisted plain wire ropes, separated from each other by a beaded wire spine, but where the body expands there are also granules and short strands of twisted ribbon. he shoulder and hip are each emphasised by a beaded wire cone, while other focal points, including the eye, are marked by granules. It is diicult to identify the particular forms of iligree on this panel without ‘More like the work of fairies than of human beings’: the iligree on the ‘Tara’ brooch… magniication, but the diverse elements have the efect of creating considerable glitter. On the panel with an animal ornament on the pin-head (Figs. 1a, 2), on the other hand, the iligree is relatively sparse: there is a single-wire border of twisted beaded wire rope; the animal is outlined by a beaded wire on a ribbon on edge, and there are granules on focal points. Scrutiny of the border, however, reveals a very subtle detail: 2-ply twisted beaded wire rope is used for most of its length. At this point, the smith drew attention to the animal’s head by replacing the 2-ply rope with a short length of thicker, 3-ply beaded rope. hese examples must suice for showing how ingenious and delicate is the iligree on the ‘Tara’ brooch. Not only is the work carried out on a minute scale, it is also so exact that it can be greatly magniied without revealing laws in manufacture. his is no doubt one of the reasons why when the ‘Tara’ brooch was displayed in London in 1863, one of the Castellani brothers remarked that it had been “worth a journey from Italy to see it” (Waterhouse et al., 1897), and why a contemporary commentator in the London Times remarked that it was “more like the work of fairies than of human beings” (O’Neill, 1863: 54). Acknowledgements I am grateful to Éamonn Kelly and other staf of the National Museum of Ireland for facilitating the examination of the ‘Tara’ brooch; to Robert Organ for giving me the text of his unpublished lecture; to Susan La Niece for locating photographs in the British Museum, and to Claudio Mari and Ivor Kerslake for digitizing them; and to Barbara Armbruster for other help. 241 References O’NEILL, H., 1863. he ine arts and civilization of ancient Ireland. Dublin, George Herbert. ORGAN, R., 1964. Text of a lecture on the ‘Tara’ brooch given to the UK Group of IIC on 28 May 1964. Department of Conservation and Scientiic Research, he British Museum. Unpublished. ORGAN, R., 1973. Examination of the Ardagh Chalice – A Case History, in W.J. Young (ed.), Application of Science in Examination of Works of Art. Proceedings of the Seminar: June 15-19, 1970. Boston, Massachusetts, Research Laboratory, Museum of Fine Arts, 238-271. STEVENSON, R.B.K., 1974. he Hunterston brooch and its signiicance. Medieval Archaeology 18: 16-42. STEVENSON, R.B.K., 1989. he Celtic brooch from Westness, Orkney, and hinged pins. Proceedings of the Prehistoric Society of Scotland 119: 469-477. UNTRACHT, O., 1982. Jewelry concepts and technology. London, Robert Hale. WATERHOUSE and Co, 1852. Ornamental Irish Antiquities. Irish Antique Brooches. Dublin, Waterhouse & Co. WATERHOUSE and Co, 1897. Letter dated 3 December 1897 to Daily Irish Independent. WERNER, A., Unpublished. Text of an undated lecture on the ‘Tara’ brooch and Ardagh chalice. Department of Conservation and Scientiic Research, he British Museum. WHITFIELD, N., 1987. Motifs and Techniques of Celtic Filigree: Are they Original? in M. Ryan (ed.), Ireland and Insular Art AD 500-1200. Dublin, Royal Irish Academy, 75-84. WHITFIELD, N., 1990. Round wire in the early Middle Ages. Jewellery Studies 4: 13-28. WHITFIELD, N., 1998. he manufacture of ancient beaded wire: experiments and observations. Jewellery Studies 8: 57-86. WHITFIELD, N., 2007. Motifs and Techniques of Early Medieval Celtic iligree: their Ultimate Origins, in R. Moss (ed.), Making and Meaning. Dublin, he Four Courts Press, 18-39. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 235-241 From the goldsmith’s point of view: gilding on metals during the first millennium AD – techniques and their development in the Germanic area Du point de vue de l’orfèvre : dorure sur métal au premier millénaire apr. J.-C. – les techniques et leur évolution dans l’aire germanique Iris Aufderhaar * Abstract: During the irst millennium AD, gilding was of great importance for the decoration of ornaments in the Germanic area. With respect to archaeological inds, the diferent techniques used to produce them are hard to distinguish without scientiic investigations. Nevertheless, it is possible to put forth some hypotheses regarding these methods on the basis of the analyses carried out so far, and the observation of typical attributes, like workmanship and choice of material. he gilding techniques used for some ibulae and brooches are analysed exemplarily, and discussed together with the advantages and disadvantages related to their speciic attributes, and changes in their technique and design during the irst millennium AD. In addition, the supply of raw materials and the correlations between gold imports into the Germanic area and the amount of gilded material are discussed. Résumé : La dorure était une technique majeure de décoration d’ornements dans l’aire germanique au premier millénaire AD. Les diverses techniques employées dans la production des trouvailles archéologiques sont diiciles à diférencier sans une étude scientiique. Néanmoins, il est possible d’avancer quelques hypothèses sur ces techniques, en se basant sur des analyses réalisées jusqu’à présent et l’observation de caractéristiques précises comme l’exécution et le choix du matériel. Les techniques de dorure utilisées dans la fabrication de ibules et broches pendant le premier millénaire AD sont ici analysées et discutées, en ce qui concerne les avantages et désavantages correspondant à leurs caractéristiques spéciiques et aux changements techniques et stylistiques. En plus, les approvisionnements en matière première et les corrélations entre les importations d’or dans l’aire germanique, ainsi que la quantité de matériel doré, sont discutés. Keywords: Gilding, gold work, brooch, Roman Iron Age, Migration Period, Merovingian Period. Mots-clés : Dorure, travail de l’or, broche, Âge du Bronze romain, Migrations, Mérovingien. * Niedersächsisches Institut für historische Küstenforschung, Lower Saxony Institute for Historical Coastal Research – Viktoriastraße 26-28, 26382 Wilhelmshaven, Germany. (aufderhaar@nihk.de) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 243-253 244 1. INTRODUCTION During the irst millennium AD, gilding was of great importance for the decoration of ornaments in the Germanic area. he difusion bonding method was veriied on several items from the Late Roman Iron Age in Germany and Norway (Becker, 2002a; b; Becker et al., 1996a; b; 1998; von Carnap-Bornheim, 2006: 91; Füting, 2001; Plather et al., 1995; Plather and Siemensen, 2002) around the middle of the 1990s, and represented an addition to the three gilding techniques that were known for the Germanic area until that date: plating, leaf gilding, and ire gilding. In most of the goldsmith workshops in existence today, those ancient techniques were replaced by electro-gilding; however, some of them, like ire gilding, are carried out on rare occasions. Leaf gilding is mainly used for the decoration of non-metallic items. With respect to archaeological inds, the diferent techniques used to produce them are diicult to distinguish without scientiic investigations. Nevertheless, it is possible to put forth some hypotheses regarding these methods on the basis of the analysis carried out so far, and the observation of typical attributes, like workmanship, as well as the choice and thickness of the material. 2. GILDING TECHNIQUES he term ‘plating’ is used for a variety of techniques having in common a feature in which the gold foil is ixed mechanically to a substructure by langing or punch marks, as well as by gluing it to the base (Fig. 1). Some disadvantages of these techniques are the weak connection between the components, which cannot prevent a corrosion of the base, and the high quantity of gold required by the comparably thick foils (von Carnap-Bornheim, 2006: 89f; Ebert, 1929: 131; Hammer, 1998: 190; Hammer and Voß, 1998: 325; Oddy, 1993: 172f ). In contrast, the advantage of leaf gilding is the relatively small and easily measurable amount of gold required. On archaeological objects, the gold foils that were attached to a substructure coated with glue (Fig. 1) and considered as gold leafs were only up to 6 μm thick (von Carnap-Bornheim, 2006: 88f; Hammer, 1998: 190f; Hammer and Voß, 1998: 315). On objects like ornaments, whose surface is frequently touched, a thin coating and a bond based on glue are less suitable because they can be easily removed (Aufderhaar, 2005: 8). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 243-253 Iris AUFDERHAAR he difusion bonding method (Fig. 1) is based on the natural difusion process of atoms, which can be accelerated by heating at a temperature of about 300 °C. A thin difusion layer is formed between two metal surfaces that are in close contact, causing a bond that provides a long lasting and even gilding of high quality and with good resistance to corrosion. A detailed description of the gilding process is included in the book De Diversis Artibus, written in the 12th century AD by the monk and goldsmith heophilius Presbyter (Brepohl, 1987: 236): after gilding had already been executed on thicker material, the gilded silver plate was hammered to foil and embossed (von CarnapBornheim, 2006: 91; Hammer, 1998: 189f; Hammer and Voß, 1998: 316f; Oddy et al., 1981: 240; Riederer, 1987: 90). According to the description, this technique is mainly suitable for the gilding of foils, but hardly for objects that are already produced in a certain shape, like cast items (Becker et al., 1996b: 58f ), or for objects with deeply proiled surfaces, where the gold foils can hardly be attached neatly enough, as well as for gilding copper alloys, because the quick oxidation of copper exposed to heat prevents the development of a difusion layer (Anheuser, 1999: 8). With coatings between 8 and 20 μm as detected on archaeological objects, this method requires a slightly higher quantity of gold than leaf gilding (Becker, 2002a: 29; Becker, 2002b; Hammer, 1998: 191f ). In addition to this repeated heating, other procedures, such as annealing processes or soldering, can lead to the inishing of the gilding on silver. he process of ire gilding is also based on the difusion of metal atoms. In this case, gold and mercury are mixed into an amalgam which is distributed evenly on the surface of the object (Fig. 1). he mercury in the amalgam supports and accelerates the difusion between substructure and coating; it evaporates from the coating at temperatures between 250 and 350 °C, leaving only a very small but characteristic amount of mercury of 5 to 25% in the gold coating (Anheuser, 1999: 34; Auleger, 1996: 628; Brepohl, 1987: 111; 2003: 401f; Bühler, 1998/1999: 443f; von CarnapBornheim, 2006: 90f; Hammer, 1998: 192f; Hammer and Voß, 1998: 325; Lins and Oddy, 1975: 370; Roth, 1986: 54f). On the basis of the amalgam paste, ire gilding is not only suitable for copper alloys, but also for items with deeply proiled and uneven surfaces, as well as cast objects (Becker et al., 1998: 207f ). In addition to the comparably high quantity of gold required, due to coatings with a thickness of up to 100 μm (Hammer, 1998: 191, 193), the other main disadvantage of this technique results from the use of mercury: to date, there is no archaeological or historical evidence that the cinnabar deposits in the Germanic area were exploited for the distillation of mercury before the Middle From the goldsmith’s point of view: gilding on metals during the irst millennium AD… 245 Figure 1: Schematic illustration of gilding techniques: plating, leaf gilding, difusion bonding and ire gilding (von Carnap-Bornheim, 2006, Fig. 1). Figure 1 : Illustration schématique des techniques de dorure: placage, dorure à la feuille, mise en couleur et dorure à l’amalgame (von CarnapBornheim, 2006, Fig. 1). Ages (Anheuser, 1999: 17; Weisgerber, 2003: 10; 2007). It can also be assumed that another possibility to obtain mercury was trade, for example with the Roman Empire (Becker et al., 1998: 208f ), and later on with the Byzantine Realm. As mercury was not an easily obtainable raw material, the idea that gilders were searching for an adequate replacement seems reasonable. Several arguments can be mentioned arguing against the theory developed by B. Arrhenius (1964: 38f; Arrhenius et al., 1968: 236; von Carnap-Bornheim, 1994: 101), that during the Vendel and Viking Ages, resin instead of mercury was used for ire gilding: resin chars during heating and would rather prevent a difusion of the ArcheoSciences, revue d’archéométrie, 33, 2009, p. 243-253 246 gold into the substructure than support it; additionally, the advantage of mercury in lowering the melting point would be lost (Anheuser, 1999: 16). Resin would be more likely it for gluing gold onto an object. At irst sight, the method entailing the replacement of mercury with lead in ire gilding seems more practicable, as proposed by A. Oldeberg (1966: 187). A similar description is preserved in the Leyden Papyrus X, a collection of receipes from the 3rd or 4th century AD (Raub, 1993: 104): gold and lead are ground together, mixed with gum serving as an adhesive, and distributed on the object to be gilded (Caley, 1926: 1156, here referred to as receipe 38; Halleux, 2002: 94, here referred to as receipe 37). According to the translation by E.R. Caley (1926: 1156), the lead was supposed to be consumed by heat, and thus evaporate, like mercury. As the boiling point of lead at 1750 °C is considerably higher than the melting point of gold (1057 °C) or those of materials like silver (950 °C) or coppee (1084 °C), a vaporization of the lead is not feasible (Aufderhaar, 2009: 33f; discussion with S. Greif, RGZM). Since some receipes of the Papyrus Leyden X contain inaccurate descriptions, K. Anheuser (1999: 20f) put forth the hypothesis that the papyrus relies on the knowledge of craftsmen, but that the writer himself was most likely not one of them. Based on Caley’s translation, C. Raub pointed out that it was not obvious if metallic lead was meant, and proposed that probably yellow lead oxide (PbO) may have been used to dilute the gold particles and stick them to the object. As a variation, he considered that by heating the mixture of metallic lead and gold in an oxidising manner up to temperatures of more than 800 °C, the lead would not alloy with the copper but oxidise to PbO, lux the copper oxides and accordingly enable a difusion of the gold (Raub, 1993: 104; see also von Lippmann 1919: 7). R. Halleux (2002: 94) provided a new translation and another interpretation of the Leyden Papyrus X: according to it, the lead was not consumed by the heat, but liqueied. As the melting point of lead at 327 °C is considerably lower than the one of gold, the lead was most likely meant to serve as a soft solder ixing a supericial gold layer to the copper object (Halleux, 2002: 174-175, remark 7). Gold and lead do not alloy but build up diferent intermetallic compounds inluenced by the amount of lead in the mixture (Brepohl, 1987: 99f; 2003: 29, 75; Wolters, 1981: 49). 3. THE DEVELOPMENT OF GILDING ANALYSED ON THE EXAMPLE OF BROOCHES In addition to their practical function, fibulae and brooches also represented decorative elements of the garb, ArcheoSciences, revue d’archéométrie, 33, 2009, p. 243-253 Iris AUFDERHAAR and therefore frequently became subject to decorations. his aspect makes them especially suitable for an exemplary analysis of changes in the technique and design of gilding. Furthermore, ibulae and brooches are types of ornaments presenting the greatest continuity, and are preserved in large numbers and numerous variations throughout the irst millennium AD. Indeed, an analysis on the development of gilding can only be carried out exemplarily due to the huge amount of gilded material from this period; however, the number of ibulae that can be presented in this article is limited. he brooches discussed in this article are considered to be representative for groups of ornaments featuring the same characteristics in terms of their manufacturing process, type of decoration and choice of material. In period B2 (Eggers, 1974) of the early Roman Iron Age, the predominant method of gilding was plating with gold sheets that were mechanically ixed to the object. One example of this is gold plating covering silver iligree: the embossed marks of the iligree work are reminiscent of decorations with gold wire known from some ibulae from period B1 (Eggers, 1974). Plated iligree is therefore a comprehensive decoration necessitating a lesser amount of gold (von Carnap-Bornheim, 1998: 467f; 2006: 90). Some specimens belonging to ibulae of the Almgren types 27 to 30 (Almgren, 1973), found mainly in Jutland, Denmark, bear small scale stamped gold foil inlays (Cosack, 1979: 49f, 98f; Nørling-Christensen, 1942). he evidence that a centre point was used for drilling the sockets for those inlays on a ibula from Quern-Scheersberg, Schleswig-Holstein, Germany (Fig. 2), and the decoration of the triangular inlays with ir branch-shaped punch marks indicate some technical inluences from the Roman provinces (Bölckow, 2006: 66; Cosack, 1979: 49f ). During the late Roman Iron Age, the type of decoration with stamped foils was continued. he important innovation was the use of gilded silver sheets that were ixed to the substructure of the given ornament. he advantage of this technique was that it required a smaller amount of gold. A polychrome design was achieved by various combinations of methods, and decorations range from simple collars of gilded foil attached to the bow of a ibula, like on an example from the 3rd century burial in Gommern, Saxony-Anhalt, Germany (Becker, 2001: 131f ), to rather complex combinations of gilded foils, silver elements such as small rivets, and rims and glass inlays, like on the Scandinavian rosette brooches (Almgren, 1973: group VII), or on brooches of type Mackeprang IX (Eggers, 1964: 37f; Ethelberg, 2000: 51f; Lund Hansen, 1971; Lund Hansen et al., 1995: 212f ). Difusion bonding was scientiically identiied as a gilding technique on the foil collar of the ibula from the Gommern From the goldsmith’s point of view: gilding on metals during the irst millennium AD… 247 Figure 2: (See colour plate) Silver ibula with gold inlays and plated iligree work from Quern-Scheersberg, Schleswig-Holstein, Germany (Bölckow, 2006: 67). Figure 2 : (Voir planche couleur) Fibule en argent avec des inclusions d’or et du iligrané plaqué, de Quern-Scheersberg, Schleswig-Holstein, Allemagne (Bölckow, 2006: 67). Figure 3: (See colour plate) Disc brooch from Tangendorf, Lower Saxony, Germany (Brandt, 2006: 71). Figure 3 : (Voir planche couleur) Broche discoïde de Tangendorf, Basse Saxonie, Allemagne (Brandt, 2006: 71). burial (Becker, 2002a; b; Becker et al., 1996a; b; 1998), and on the fragments of two rosette brooches from Norway: the bonding was achieved by hammering the gold foil to the silver plate and subsequently heating both components. On two other Norwegian rosette brooches, evidence that the bonding was supported by a copper-bearing solder was found (Plather et al., 1995: 12f; Plather and Simensen, 2002: 547f). he design of the disc brooches follows the same principles, using gilded and stamped silver foils, silver elements, and sometimes glass inlays. Since difusion bonding had already been carried out before hammering the metal, the foils must have been stamped after the gilding, but some of the disc brooches show traces of a ire gilded surface. he production of chased foil, like on the disc brooch from Tangendorf, Lower Saxony, Germany (Fig. 3), with its elaborate representation of a deer (Brandt, 2006: 70; Drescher, 1955: 25f), or other intensely curved foils, like on tutulus shaped brooches (Adomat, 2000: 411f; Eichhorn, 1908: 903f; Ethelberg, 2000: 312f; homas, 1966: 103) requires repeated annealing processes, as the distortion of precious metals through techniques such as hammering or embossing has to take place in a cold state. During such procedures, the crystal structure is forced into an unnatural condition; the metal hardens and can break under further distortion. In order to prevent this, precious metals are annealed after certain stages of distortion to achieve a re-crystallization of the structure and to regain their lexibility. he annealing process requires a minimal temperature of 200 °C, but can also be carried out at higher temperatures (Brepohl, 2003: 184f; Wolters, 1981: 26f ). Silver that is already gilded would easily fade if heated several times. It is reasonable to assume therefore that the respective foils were gilded after curving by the technique of ire gilding. Scientiic evidence that ire gilding was carried out during the Late Roman Iron Age was detected on the top of a silver shield boss from the aforementioned Gommern burial (Anheuser, 1999: 16; Becker et al., 1998: 207, 209; Bühler, 1998/1999: 444). he same argumentation is valid for gilded foils soldered to the substructure of the brooches. From the times at the end of the Roman Iron Age and the Migration Period, several types of brooches are preserved that display gilding applied to more massive material, and in combination with several new decorative techniques, as for example on the silver bow brooches of the Wiesbaden ArcheoSciences, revue d’archéométrie, 33, 2009, p. 243-253 248 Iris AUFDERHAAR type from the period between the end of the 4th and the early 5th century AD (Fig. 4), which are mainly distributed throughout the Middle Rhine Area. he decorations of those ibulae, carried out in niello and with punch marks as well as motifs like beaded string, are reminiscent of objects manufactured in the nearby Roman provinces during the late Roman Iron Age. hose objects can be considered as models for the decorations of the ibulae in terms of technique and ornamentation, as was already pointed out by Joachim Werner (1981). Mercury contents detected by spectral analysis on a closely related ibula from GroßKöris, Brandenburg, Germany (Fig. 5) support this thesis: ire gilding was carried out both on the thicker material of the foot, itself decorated by chip-carving, but also on silver foils mechanically attached to the bow of the ibula (Franke, 1987: 237f; Gustavs, 1987: 215f; Voß et al., 1998). Mercury contents on the surface of a gilded silver ibula from a grave ind on the farm of Roligheten, Norway, represent yet another evidence of ire gilding during this period (Plather and Simensen, 2002: 548, 555). Massive cast items gilded directly on their copper based alloys are found among the Saxon and Anglo-Saxon equal arm brooches and disc brooches, which are also decorated with elements deriving from the Roman provinces, like the Figure 5: (See colour plate) Fibula decorated with chip-carving technique from Gross-Köris, Brandenburg, Germany (Voß et al., 1998: pl. 65b). Figure 5 : (Voir planche couleur) Fibule décorée au moyen de la technique de chip-carving, de Gross-Köris, Brandebourg, Allemagne (Voß et al., 1998: pl. 65b). Figure 4: Fibulae of the Wiesbaden-type. Fibula from Wiesbaden, Hesse, Germany, decorated with niello inlays and punch marks, and ibula from Wulfen, Saxony-Anhalt, Germany, decorated with beaded string imitation (Werner, 1981: attachment 2, Figs. 15 and 1). Figure 4 : Fibules de type Wiesbaden. Fibule de Wiesbaden, Hesse, Allemagne, décorée avec des inclusions de niello et marques de poinçonnage, Fibule de Wulfen, Saxonie-Anhalt, Allemagne, décorée avec des imitations de cordons perlés (Werner, 1981: document joint 2, Fig. 15 et 1). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 243-253 egg and dart motif (Böhme, 1974: 14f, 220; Bruns, 2003: 32, 57f; Haselhof, 1979: 153f; Kühn, 1981: 64f ). he only possible method for obtaining a durable gilding of those alloys is ire gilding. Fire gilding was clearly established as the main gilding technique when the Germanic goldsmiths started to produce ornaments of more massive materials, like silver plate, or by casting of copper based alloys. hese profound changes occurred together with the adoption of several techniques and decorative styles from metal objects manufactured in the Roman provinces during the end of the Roman Iron Age. During the Migration and Merovingian Periods, the combination of decorative techniques (niello, chip-carving and ire gilding) on brooches cast of silver or copper alloy became a regular pattern for the decoration of the bow ibulae (Göldner, 1987; Koch, 1998). In contrast, a unique phenomenon in the gilding of brooches during the irst millennium AD is veriied on some From the goldsmith’s point of view: gilding on metals during the irst millennium AD… samples of small garnet brooches from the period between the end of the 5th and the beginning of the 6th century AD: while made of iron, the brooches were gilded by mechanically ixed gold or gilded silver sheets (Vielitz, 2003: 20f, 162, 165, 207). Clearly, the technically challenging operation of ire gilding on iron, which requires a preliminary copper plating of the iron surface (Anheuser, 1999: 22; Hammer, 1998: 198), was not known or at least not carried out before then. Gilded embossed foils were not abandoned as decorative elements, but were frequently used on disc brooches throughout the Migration and Merovingian Periods. While disc brooches bearing sockets or iligree decorations (Rademacher, 1940; hieme, 1978) soldered to the foil, or brooches bearing copper alloy foils, must have been to be ire gilded, some of the disc brooches bearing stamped silver foils (Klein-Pfeufer, 1993) could also have been gilded by difusion bonding. To date, no analysis was carried out on those brooches, but a hint that the difusion bonding method was not at all forgotten by the end of the Roman Iron Age is the detailed description provided in the 12th century AD by heophilius Presbyter (Brepohl, 1987: 236). he number of gilded pieces among brooches increased considerably until the 6th century AD, and gilding of brooches was common during the Merovingian Period (Göldner, 1987; Koch, 1998; Kühn, 1981). From the end of the Merovingian Period onward, a clear decline in the amount of gilded material can be observed, which continued, at least on the continent, throughout the Carolingian Period. his development can be exempliied by the equal arm brooches manufactured between the beginning of the 7th century AD and the Carolingian Period: only approximately 2% of the material was gilded (hörle, 1998: 106f; 2001: 1, 5, 13, 309). Only very few gilded brooches are preserved from the Carolingian Period, and most of them are quite small objects, like bird shaped brooches or brooches with enamel inlays (Frick, 1992/1993; Haselhof, 1990). In Viking Age Scandinavia, the situation was quite different. A signiicantly higher number of brooches, like disc brooches (Jansson, 1984a; b) and oval brooches (Jansson, 1985), mainly made of copper based alloys, were gilded. After carrying out microscopic analysis on oval brooches, A. Oldeberg (1943: 270f; 1966: 186f ) postulated that both leaf gilding and ire gilding were employed in their making. he diferences in the use and frequency of gilding during this period may have several reasons. In the Carolingian Empire, a signiicant number of brooches were made of tin or lead, metals not appropriate for ire gilding (Frick, 1992/1993). Furthermore, Viking Age Scandinavia also had the advantage of being able to purchase mercury because of its trade connections with the Byzantine Realm and Arabian countries. 249 4. THE SUPPLY OF GOLD To date, there is only limited evidence regarding gold extraction in the Germanic area during the irst millennium AD (Andersson, 1995: 11; Siegmund, 1998: 308). he only indications concerning placer gold mining in the Upper Rhine Area are the distinctive platinum and silver contents detected on gold objects from the Merovingian Period (Hartmann and Wolf, 1975: 23ff ). Therefore, imports from the Roman Empire, and later from the East Roman and Byzantine Realm, are commonly considered to represent the source of the gold used in the Germanic area (Andersson, 1995: 11; Siegmund, 1998: 308). Nevertheless, evidence of gold imports is not readily available. In spite of this, there are some indications that the gold imports correspond quite well with the amount of gilded material. he irst few gilded objects were from period B2 (Eggers, 1974) of the early Roman Iron Age, corresponding to a irst increase of the gold imports into the Germanic area, which, after a brief caesura, continued during the Late Roman Iron Age (Andersson, 1995: 10; Roggenbuck, 1988: 55, 61). It is not evident in which form the gold was imported. F. Siegmund (1998: 308) suggests that imported gold coins served as raw material. His argument is supported by the inding of gold coins at the workshop site of Helgö, Sweden, bearing marks which may originate from testing the gold contents (Kyhlberg, 1986: 30). In Denmark and Sweden, the amount of imported gold coins in deposits indicates an increasing import, which was interrupted after reaching its peak during the 5th and 6th centuries AD (Jørgensen and Vang Petersen, 1998: 279; Lund Hansen, 1987: 231). A shortage of precious metals from the middle of the 6th century AD onwards becomes apparent in the Frankish Realm as well, as golden ornaments became scarce in Middle and Northern Europe (Siegmund, 1998: 212). In addition, examinations carried out by J.P.C. Kent (1972: 70f) on Merovingian gold coins indicate a reduction of the gold contents from 90-100% down to 30% during the period between the end of the 6th and the beginning of the 7th century AD. hey were inally replaced by silver coins at the end of the 7th century AD. he considerable decrease of gilded objects from the 7th century AD onwards follows this tendency, showing however a certain delay. For Viking Age Scandinavia, an increase in gold imports can be observed from the amount of gold ornaments and deposits (Jørgensen and Vang Petersen, 1998: 286), which once again corresponds to the amount of gilded material. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 243-253 250 5. CONCLUSIONS Addressing the reasons for the development of gilding techniques, it can be assumed that the shift from plating to difusion bonding was most likely caused by an efort to save gold. Nevertheless, this argument is not appropriate for explaining the success of ire gilding, because in that case the gold layer is considerably thicker than those produced by difusion bonding. In addition to this, ire gilding required not only gold, but also mercury, as raw material. he success of the ire gilding method is therefore most likely related to its more comprehensive applicability, associated with a high quality of the gilding. It represented the best technique for gilding massive or deeply proiled objects, and pieces made of copper alloys. Nevertheless, it is premature to set a limit for the difusion bonding method to the Roman Iron Age. he development of gilding was inluenced by several factors. he coherence of gilded material with the amount of gold available in the Germanic territories is noticeable. One other inluence was related to developments in the manufacturing of jewellery. his meant that the technique that was most suitable for a given type of ornament was used more frequently. During the early Roman Iron Age, the use of gilded and stamped silver foils enabled a production of a larger amount of similar decorative elements without a considerably higher use of gold (von Carnap-Bornheim, 2006: 92.). he ire gilding method opened up new possibilities for the mass production of gilded ornaments, because it was suitable for gilding cast objects. Inluences from the Roman Empire were most likely another factor having an impact on further developments: evidently, changes in gilding techniques occurred contemporarily with new decorative techniques and styles deriving from the Roman Provinces (Aufderhaar, 2005: 122f). References ADOMAT, A., 2000. Die drei Prachtibeln von Skovgårde, in P. Ethelberg (ed.), Skovgårde. Ein Bestattungsplatz mit reichen Frauengräbern des 3. Jhs. n. Chr. auf Seeland. Nordiske Fortidsminder Serie B 19. Kopenhagen, Det Kongelige Nordiske Oldskriftselskab, 409-413. ALMGREN, O., 1973. Studien über Nordeuropäische Fibelformen der ersten christlichen Jahrhunderte mit Berücksichtigung der provinzialrömischen und südrussischen Formen. MannusBibliothek 32. Bonn, Kabitz. ANDERSSON, K., 1995. Romatida guldsmide i Norden III. Övriga smycken, teknisk analys och verkstadgrupper. AUN 21. Uppsala, Societas Archaeologica Upsaliensis. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 243-253 Iris AUFDERHAAR ANHEUSER, K., 1999. Im Feuer vergoldet. Geschichte und Technik der Feuervergoldung und der Amalgamversilberung. AdRSchriftenreihe zur Restaurierung und Grabungstechnik 4. Stuttgart: heiss. ARRHENIUS, B., 1964. Patinering och kolorering, in W. Holmqvist (ed.), Nordiskt guldsmide under järnåldern. Stockholm, Historiska Museet, 38-40. ARRHENIUS, B. 1968. Sveagold und Wikingerschmuck. Römisch Germanisches Zentralmuseum Mainz Ausstellungskataloge 3. Mainz, Römisch Germanisches Zentralmuseum. AUFDERHAAR, I., 2005. Die Vergoldung im ersten Jahrtausend nach Christus. Technologien und ihre Relevanz im archäologischen Kontext untersucht anhand ausgewählter Fundgruppen. Unpublished Magisterthesis, University of Münster, Germany. A UFDERHAAR, I., 2009. Zu Entwicklungen in der Vergoldungstechnik im germanischen Raum während des 1. Jahrtausends nach Christus. Restaurierung und Archäologie 2: 31-46. AUFLEGER, M., 1996. Metallarbeiten und Metallverarbeitung, in A. Wieczorek, P. Perin, K. von Welck, W. Menghin (eds.), Die Franken. Wegbereiter Europas 2. Mainz, Philipp von Zabern, 618-628. BECKER, M., 2001. Bekleidung – Schmuck – Ausrüstung, in S. Fröhlich (ed.), Gold für die Ewigkeit. Das germanische Fürstengrab von Gommern. Landesamt für Archäologie Sachsen-Anhalt. Halle, Landesmuseum für Vorgeschichte, 127-147. BECKER, M., 2002A. Ein Germanenschild unter dem Mikroskop. Archäologie in Deutschland 6: 28-29. BECKER, M., 2002B. Der Fund des Monats Oktober 2002. Der Schild des Fürsten von Gommern. Germanische Kunst und Römische Beute (?) aus einem Grab des 3. Jahrhunderts. From: www.archlsa.de/funde-der-monate/10.02/, 12/2004. BECKER, M., FÜTING, M. and SCHNARR, H., 1996A. Vergoldete Silberpreßbleche der römischen Kaiserzeit aus materialkundlicher und technologischer Sicht, in S. Ostritz, R. Einicke (eds.), Terra & Praehistoria. Festschrift Klaus-Dieter Jäger. Beiträge zur Ur- und Frühgeschichte Mitteleuropas 9. Wilkau-Hasslau, Beier & Beran, 99-106. BECKER, M., FÜTING, M. and SCHNARR, H., 1996B. Mikroskopische und mikroanalytische Untersuchungen an Fundstücken aus dem „Fürstengrab“ von Gommern, Ldkr. Jerichower Land. Zweiter Teil. Jahresschrift für mitteldeutsche Vorgeschichte 78: 37-62. BECKER, M., 1998. Römische und germanische Bunt- und Edelmetallfunde im Vergleich. Archäometallurgische Untersuchungen ausgehend von elbgermanischen Körpergräbern. Bericht der Römisch-Germanischen Kommission 79: 204-216. From the goldsmith’s point of view: gilding on metals during the irst millennium AD… BÖHME, H.W., 1974. Germanische Grabfunde des 4. bis 5. Jahrhunderts zwischen unterer Elbe und Loire. Studien zur Chronologie und Bevölkerungsgeschichte. Münchner Beträge zur Vor- und Frühgeschichte 19. München, Beck. B ÖLCKOW, B., 2006. Quern-Scheersberg. Ein Friedhof im Feld, in R.-M. Weiss, W. Marnette (eds.), Frühes Gold aus Norddeutschland. Veröfentlichungen des Helms-Museums für Archäologie und die Geschichte Hamburgs 96. Hamburg, Helms-Museum, 66-67. BRANDT, J., 2006. Fürstenschmuck aus der Schublade. Die Scheibenibel von Tangendorf, in R.-M. Weiss, W. Marnette (eds.), Frühes Gold aus Norddeutschland. Veröfentlichungen des Helms-Museums für Archäologie und die Geschichte Hamburgs 96. Hamburg, Helms-Museum, 70-71. BREPOHL, E., 1987. heophilius Presbyter und die mittelalterliche Goldschmiedekunst. Wien, Köln, Graz, Böhlau. BREPOHL, E., 2003. heorie und Praxis des Goldschmieds. Leipzig and Köln, Fachbuchverlag. BRUNS, D., 2003. Germanic Equal Arm Brooches of the Migration Period. A study of style, chronology and distribution including a full catalogue of inds and contexts. BAR International Series 1113. Oxford, Archaeopress. B ÜHLER, B., 1998/99. Untersuchungen zu Guß, Oberlächenbearbeitung und Vergoldung an frühmittelalterlichen Bunt- und Edelmetallgegenständen. Archaeologia Austriaca. Beiträge zur Paläanthropologie, Ur- und Frühgeschichte Österreichs 82/83: 429-478. CALEY, E.R., 1926. he Leyden Papyrus X. An English Translation with brief notes. Journal of Chemical Education 2: 1149-1166. VON CARNAP-BORNHEIM, C., 1994. Zur Entwicklung des germanischen Gold- und Silberschmiedehandwerks vor und nach den Markomannenkriegen –Vergoldung, Filigran und Pressblech, in H. Friesinger, J. Tejral, A. Stuppner (eds.), Markomannenkriege. Ursachen und Wirkung. VI. Internationales Symposium “Grundprobleme der frühgeschichtlichen Entwicklung im nördlichen Mitteldonaugebiet”. Brno, Archäologisches Institut der Akademie der Wissenschaften der Tschechischen Republik, 99-107. VON CARNAP-BORNHEIM, C., 1998. Übernahme und Verbreitung innovativer Techniken und Verzierungsgewohnheiten bei germanischen Fibeln, in J. Kunow (ed.), 100 Jahre Fibelformen nach Oscar Almgren. Forschungen zur Archäologie im Land Brandenburg 5. Wünsdorf, Verlag Brandenburgisches Landesmuseum für Ur- und Frühgeschichte, 467-473. VON CARNAP-BORNHEIM, C., 2006. Mehr Schein als Sein, oder: Vergoldung in der Römischen Kaiserzeit im germanischen Barbaricum, in R. Bleile (ed.), Magischer Glanz. Gold aus archäologischen Sammlungen Norddeutschlands. Schleswig: Archäologisches Landesmuseum in der Stiftung SchleswigHolsteinische Landesmuseen Schloß Gottorf, 88-92. 251 COSACK, E., 1979. Die Fibeln der Älteren Römischen Kaiserzeit in der Germania libera (Dänemark, DDR, BRD, Niederlande, CSSR). Eine technologisch-archäologische Analyse. Teil I. Armbrustibeln, Rollenkappenibeln, Augenibeln. Neumünster, K. Wacholz. DRESCHER, H., 1955. Die Nachbildung der Scheibenibel aus Tangendorf, Kreis Harburg. Die Kunde 5: 25-33. EBERT, M., 1929. Vergolden. Reallexikon der Vorgeschichte 14. Berlin and New York: De Gruyter, 131-132. EGGERS, H.J., 1964. Die Kunst der Germanen in der Eisenzeit, in H.J. Eggers, E. Will, R. Jofroy, W. Holmqvist (eds.), Kelten und Germanen in heidnischer Zeit. Kunst der Welt. Ihre geschichtlichen, soziologischen und religiösen Grundlagen. BadenBaden, Holle, 5-87. EGGERS, H.J., 1974. Einführung in die Vorgeschichte. München, R. Piper & Co Verlag. EICHHORN, G., 1908. Der Grabfund zu Dienstedt bei Remda. Zeitschrift für Ethnologie 40: 902-914. ETHELBERG, P., 2000. Skovgårde. Ein Bestattungsplatz mit reichen Frauengräbern des 3. Jhs. n. Chr. auf Seeland. Nordiske Fortidsminder Serie B 19. Kopenhagen, Det Kongelige Nordiske Oldskriftselskab. FRANKE, H., 1987. Zur Herstellung der Kerbschnittibel von Groß-Köris, Kr. Königs Wusterhausen. Veröfentlichungen des Museums für Ur- und Frühgeschichte Potsdam 23: 237-241. FRICK, H.J., 1992/1993. Karolingisch-ottonische Scheibenibeln des nördlichen Formenkreises. Ofa 49: 243-463. FÜTING, M., 2001. Wo der Augenschein versagt – Mikroanalysen am archäologischen Objekt, in S. Fröhlich (ed.), Gold für die Ewigkeit. Das germanische Fürstengrab von Gommern. Halle, Landesamt für Archäologie Sachsen-Anhalt – Landesmuseum für Vorgeschichte, 164-167. GÖLDNER, H., 1987. Studien zu Rhein- und Moselfränkischen Bügelibeln I. Marburger Studien zur Vor- und Frühgeschichte 8-1. Marburg, Hitzeroth. GUSTAVS, S., 1987. Silberschmuck, Wafen und Siedlungsbefunde des 3. bis 5. Jh. u. Z. aus einem See bei Groß Köris, Kr. Königs Wusterhausen. Veröfentlichungen des Museums für Ur- und Frühgeschichte Potsdam 23: 215-236. HALLEUX, R., 2002. Les Alchimistes Grecs. Papyrus de Leyde, Papyrus de Stockholm, Recettes. Paris, Les Belles Lettres. HAMMER, P., 1998. Verfahrenstechnische Untersuchungen. Berichte der Römisch-Germanischen Kommission 79: 179-199. HAMMER, P. and VOSS, H.-U., 1998. Glossar metallkundlicher und herstellungstechnischer Fachbegriffe – erläutert für antike Handwerkstechnik. Bericht der Römisch-Germanischen Kommission 79: 314-330. H ARTMANN , A. and WOLF , R., 1975. Vergleichende Spektralanalysen an einigen frühmittelalterlichen Goldfunden und Goldblattkreuzen, in W. Hübener (ed.), Die ArcheoSciences, revue d’archéométrie, 33, 2009, p. 243-253 252 Goldblattkreuze des frühen Mittelalters. Veröfentlichungen des Alemannischen Institutes Freiburg i.Br. 37. Bühl, Konkordia. HASELHOFF, G., 1979. Römische Elemente in sächsischem Schmuck (am Beispiel der gleicharmigen Fibeln), in C. Ahrens (ed.), Sachsen und Angelsachsen. Veröfentlichungen des HelmsMuseums für Archäologie und die Geschichte Hamburgs 32. Hamburg, Helms-Museum, 153-161. HASELHOFF, G., 1990. Email im frühen Mittelalter. Frühchristliche Kunst von der Spätantike bis zu den Karolingern. Marburger Studien zur Vor- und Frühgeschichte Sonderband 1. Marburg, Hitzeroth. JANSSON, I., 1984A. Kleine Rundspangen, in G. Arwidsson (ed.), Birka II:1. Systematische Analysen der Gräberfunde. Stockholm, Almqvist & Wiksell, 58-74. JANSSON, I., 1984B. Kleine Rundspangen, in G. Arwidsson (ed.), Birka II:1. Systematische Analysen der Gräberfunde. Stockholm, Almqvist & Wiksell, 75-84. JANSSON, I., 1985. Ovala spännbucklor. En studie av vikingatida standardsmycken med utgångspunkt från Björkö-fynden. AUN 7. Uppsala, Societas Archaeologica Upsaliensis. JØRGENSEN, L. and VANG PETERSEN, P., 1998. Guld, magt og tro. Danske guldskatte for oldtid og middelalder. = Gold, power, belief. Danish gold treasures from Prehistory and the Middle Ages. Kopenhagen, Nationalmuseet. KENT, J.P.C., 1972. Gold Standards of the Merovingian Coinage, A.D. 580-700, in E.T. Hall (ed.), Methods of Chemical and Metallurgical Investigations of Ancient Coinage. Royal Numismatic Society Special Publication 8. London, Royal Numismatic Society, 69-74. KLEIN-PFEUFFER, M., 1993. Merowingerzeitliche Fibeln und Anhänger aus Preßblech. Marburger Studien zur Vor- und Frühgeschichte 14. Marburg, Hitzeroth. KOCH, A., 1998. Bügelibeln der Merowingerzeit im westlichen Frankenreich. Monographien des Römisch-Germanischen Zentralmuseums, Forschungsinstitut für Vor- und Frühgeschichte 41, 1 und 2. Mainz, Verlag des RömischGermanischen Zentralmuseums. KÜHN , H., 1981. Die germanischen Bügelfibeln der Völkerwanderungszeit in Mitteldeutschland. Akademische Druck- und Verlagsanstalt,Graz. KYHLBERG, O., 1986. Late Roman and Byzantine Solidi. An archaeological analysis of coins and hoards, in A. Lundström, H. Clarke (eds.), Excavations at Helgö X. Coins, Gold and Iron. Stockholm, Almqvist & Wiksel, 13-126. LINS, P.A. and ODDY, W.A., 1975. he Origins of Mercury Gilding. Journal of Archaeological Science 2: 365-373. LIPPMANN, E. von, 1919. Entstehung und Ausbreitung der Alchemie. Berlin, Springer. LUND HANSEN, U., 1971. Blik- og glasornamenterede Fibler af Mackeprang Type IX. Aarbøger for nordisk Oldkyndighet og Historie 1971: 72-119. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 243-253 Iris AUFDERHAAR L UND HANSEN , U., 1987. Römischer Import imNorden. Warenaustausch zwischen dem Römsichen Reich und dem freien Germanien während der Kaiserzeit unter besonderer Berücksichtigung Nordeuropas. Nordiske Fortidsminder Serie B 10. Kopenhagen, Det Kongelige Nordiske Oldskriftselskab. LUND HANSEN, U., 1995. Himlingøje – Seeland – Europa. Ein Gräberfeld der jüngeren römischen Kaiserzeit aus Seeland, seine Bedeutung und internationalen Beziehungen. Nordiske Fortidsminder Serie B 13. Kopenhagen, Det Kongelige Nordiske Oldskriftselskab. NØRLING-CHRISTENSEN, H., 1942. Une trouvaille de parures de l’ancien âge du fer romain faite a Vester-Mellerup, Vendsyssel. Acta Archaeologica 13: 332-355. ODDY, W.A. 1993. Gilding of Metals in the Old World, in Metal Plating and Patination, S. La Niece and P. Craddock (eds.), Butterworth-Heinemann Ltd, Oxford, 171-181. ODDY, W.A., LA NIECE, S., CURTIS, J.E. and MEEKS, N.D., 1981. Difusion-bonding as a Method for Gilding in Antiquity. MASCA Journal 1(8): 239-241. OLDEBERG, A., 1943. Metallteknik under förhistorisk tid II. Stockholm, Viktor Pettersons Bokindustri AB. OLDEBERG, A., 1966. Metallteknik under vikingatid och medeltid. Stockholm: Viktor Pettersons Bokindustri AB. PLATHER, U., ASTRUP, E. and STRAUME, E., 1995. Norwegian rosette-brooches of the 3rd century AD: their construction, materials and technique. Historical Metallurgy 29(1): 12-24. PLATHER, U. and SIMENSEN, C.J., 2002. Some characteristic features of gilded jewellery from the 3rd, 5th and 8th centuries found in Norway. Germania 80: 547-570. RADEMACHER, F., 1940. Fränkische Goldscheibenibeln aus dem rheinischen Landesmuseum in Bonn. München, Bruckmann. RAUB, C., 1993. How to coat objects with gold – Pliny, Leyden Papyrus X, Mappae Clavicula and heophilius seen with a modern chemist’s eyes, in C. Eluère (ed.), Outils et ateliers d’orfèvres des temps anciens. Antiquités nationales mémoire 2. Saint-Germain-en-Laye: Société des Amis du Musée des Antiquités Nationales et du château de Saint-Germain-en Laye, 101-110. RIEDERER, J., 1987. Archäologie und Chemie – Einblicke in die Vergangenheit. Berlin, Rathgen-Forschungslabor. ROGGENBUCK, P., 1988. Untersuchungen zu den Edelmetallfunden der römischen Kaiserzeit zwischen Limes, Nord- und Ostsee. BAR International Series 449. Oxford, Archaeopress. ROTH, H., 1986. Kunst und Handwerk im frühen Mittelalter. Archäologische Zeugnisse von Childerich I. bis zu Karl dem Großen. Stuttgart, heiss. SIEGMUND, F., 1998. Gold. Reallexikon der Germanischen Altertumskunde 12. Berlin and New York, De Gruyter, 308312. From the goldsmith’s point of view: gilding on metals during the irst millennium AD… THIEME, B., 1978. Filigranscheibenibeln der Merowingerzeit aus Deutschland. Bericht der Römisch-Germanischen Kommission 59: 381-500. THÖRLE, S., 1998. Formen und Verzierungen gleicharmiger Bügelibeln im westlichen Frankenreich. Acta Praehistorica et Archaeologica 30: 106-112. THÖRLE, S., 2001. Gleicharmige Bügelibeln des frühen Mittelalters. Universitätsforschungen zur prähistorischen Archäologie 81. Bonn, Habelt. THOMAS, S., 1966. Die germanischen Scheibenfibeln der Römischen Kaiserzeit im freien Germanien. Berliner Jahrbuch für Vor- und Frühgeschichte 6: 1-187. VIELITZ, K., 2003. Die Granatscheibenibeln der Merowingerzeit. Europe médiévale 3. Montagnac, Mergoil. VOSS, H.-U., HAMMER, P. and LUTZ, J., 1998. Römische und germanische Bunt und Edelmetallfunde im Vergleich. Bericht der Römisch-Germanischen Kommission 79: 107-382. 253 WEISGERBER, G., 2003. Quecksilber. Reallexikon der Germanischen Altertumskunde 24. Berlin and New York, De Gruyter, 9-11. WEISGERBER, G., 2007. Zinnober. Reallexikon der Germanischen Altertumskunde 34. Berlin and New York, De Gruyter, 572573. WERNER, J., 1981. Zu einer elbgermanischen Fibel des 5. Jahrhunderts aus Gaukönigshofen, Ldkr. Würzburg. Ein Beitrag zu den Fibeln vom „Typ Wiesbaden“ und zur germanischen Punzornamentik. Bayerische Vorgeschichtsblätter 46: 225-254. WOLTERS, J., 1981. Der Gold- und Silberschmied 1. Werkstofe und Materialien. Stuttgart, Rühle-Diebener Verlag GmbH. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 243-253 Gold foil covering of the handle of an iron knife from burial 2 of the Hunnic Period cemetery at Mukhino, in the Upper Don area L’ornement en or du manche d’un poignard en fer de la tombe 2 de la période des Huns de la nécropole de Moukhino, dans le Haut Don Irina Saprykina*, Yurii A. Teterin** and Robert Mitoyan*** Abstract: he chemical composition of 7 fragments of the gold foil covering the wooden (?) handle attached to an iron knife from the burial of a noblewoman in the Upper Don area was investigated by two methods – XRF and X-ray electronic spectroscopy (XRS). All samples showed that the gold foil was produced from an Au-Ag-Cu alloy (type III), with a high degree of Ag concentration in the XRS samples. he data suggests that this is the result of the process of silver difusion to the surface of the samples, but the reasons for this are not clear yet. he actual quality of the knife handle is low and, in comparison with other high-status objects, it cannot be included in the category of high-quality artefacts. hus, we cannot exclude the possibility that the knife from burial 2 at Mukhino is a local copy, or an item produced in the ‘metropolis’ especially for export to the ‘barbarian’ world. Résumé : Les compositions chimiques de 7 fragments de feuilles d’or couvrant le manche en bois (?) d’un couteau en fer provenant de la tombe d’une femme noble, trouvée dans le Haut Don, ont été déterminées par deux méthodes – FX et spectroscopie de difusions de rayons X (XRS). Tous les échantillons montrent que les feuilles d’or ont été produites à partir d’un alliage Au-Ag-Cu (type III), avec une forte teneur en Ag dans le cas des échantillons analysés par XRS. Ces données suggèrent qu’il s’agit du résultat d’un procédé de difusion d’argent vers la surface des échantillons, mais le mécanisme n’a pas encore pu être clairement déini. Le manche du couteau est de mauvaise qualité et, en comparaison avec d’autres objets de prestige, ne peut pas être inclus dans le groupe d’objets de bonne qualité. Pour cette raison, il ne peut pas être exclu que le couteau de la tombe 2 de la nécropole de Mukhino soit une copie de fabrication locale ou un objet produit dans la « métropole » spéciiquement pour exportation vers le monde « barbare ». Keywords: Burial of a noblewoman, gold foil covering of a knife handle, XRF analysis, X-ray electronic spectroscopy. Mots-clés : Tombe noble, feuille d’or, manche de couteau, XRF, XRS. 1. INTRODUCTION In 2002, a rich burial of a woman (burial 2) was discovered at the settlement of Mukhino, in the Upper Don region. So far, this is the only high-status burial in the Upper Don area (Southern Russia) dating to the Hunnic period. Judging by the grave goods, the burial dates to the second quarter of the 5th century AD, that is, to the end of period D2 (Untersiebenbrunn horizon: 380/400-440/450 AD), or to the beginning of period D2/D3 (Smolin-Kosino horizon: 430/440-460/480 AD) on the chronological scale of the European Barbaricum. Such graves mark certain power * RAS Institute of Archaeology – 117036, Dm.Ulyanova str., 19, Moscow, Russia. (dolmen200@mail.ru) ** he Kurchatov Institute – 123182, Kurchatov sq., 1, Moscow, Russia. (antonxray@yandex.ru) *** Geological Faculty of the MSU, Chair of Geochemistry – 119992, Leninsky Gory, 1, Moscow, Russia. (mitoyan@geol.msu.ru) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 255-257 256 Figure 1: (See colour plate) Fragment of gold foil covering. Figure 1 : (Voir planche couleur) Fragment de feuille de couverture. centres in the ancient pre-state formations of the East European ‘barbarians’ and also demonstrate that, geographical distances notwithstanding, the barbarian societies had a common hierarchical system. he rich grave goods from burial 2 at Mukhino cemetery are compatible with the well-known ‘princely’ inds from sites in Central and Western Europe, such as for example Untersiebenbrunn or Airan. Together with other inds from Eastern Europe, the materials from Mukhino represent a reliable basis for creating a uniied chronological scale for East European antiquities of the Great Migration period. he materials can also be useful for the study of social differentiation, as revealed by the quality and quantity of the grave goods. Burial 2 yielded over 30 gold fragments, including the covering of the handle of an iron knife, made of a thin metal plate (foil) of golden colour, with a characteristic ornament not encountered on any of the other artefacts from the burial. he iron knife with gold foil covering on the handle was found inside one of the samples of wood selected for radiocarbon analysis. he type of wooden item inside which the knife was found could not be identiied with certainty, yet the part which is still extant and contained the knife could be interpreted as a sheath or scabbard. he gold covering is fragmentary; yet, the overall amount of fragments and their location on the handle allow achieving a reliable reconstruction of the artefact. Fourteen fragments of the gold foil are extant, all of them representing a thin and lexible plate with a maximum thickness of 0.5 mm. he face is golden in colour and shows dark, almost black, spots. he reverse is dark and shows traces of some organic bonding agent. he foil must have been glued to the wooden handle with animal glue, since no traces of the chemical elements that indicate gilding have been found. he ornament was applied with a ArcheoSciences, revue d’archéométrie, 33, 2009, p. 255-257 Irina SAPRYKINA, Yurii A. TETERIN, Robert MITOYAN Faulenzerpunzen tool, which had a defect of the working surface. All fragments of the gold foil showed the same defect, which indicates that all the fragments belong to one and the same object. he other artefacts from burial 2 (plaques, mounts of plated metal) showed no analogies to the technique used for creating the knife handle covering. Nonetheless, similar coverings of sword handles and, less frequently, knife handles, are encountered in East European (Poland, Hungary) and North Caucasian cemeteries, and are characteristic of the Hun and post-Hun times. he metallurgy of gold-based alloys has been the subject of several studies (McDonald and Sistare, 1978; Rapson, 1990; Pinasco and Stagno, 1979). he main focus of our work was to study the chemical composition of the gold foil covering from burial 2 at the Mukhino 2 cemetery and to compare it with the available data on other similar items. 2. METHODS XRF was used to analyze the chemical composition of the metal (7 fragments of the gold foil have been analyzed). The analysis was carried out in the spectroscopy Laboratory of the Geological Faculty of Moscow State University (MSU) with a portable XRF analyzer: a laptop with special software, multichannel analyzer and sensor with radioisotope source (Am241 and Cd109 isotopes) (developed by R. Mitoyan, S. Koloskov, N. Eniosova, T. Saracheva). he measurement procedure is standardized, and the methodology of the experiment is provided in the study by N. Eniosova, R. Mitoyan and T. Saracheva (2008: 114-120). he resulting data is presented in Table 1. In order to obtain additional data regarding the gold and silver concentrations on the surface of the samples, their chemical composition was analyzed by X-ray electronic spectroscopy (XRS). he analysis was carried out in the precision spectroscopy laboratory of the Kurchatov Institute. he samples were foil fastened to an aluminum support plate. he X-ray spectrum of the investigated samples (191, 192, 195, 200, 205, 207, 208) were obtained using a MK II VG Scientiic spectrometer with AlKα X-ray in vacuum 1.3∙10-7 Pa at room temperature. he methodology of the experiment is provided in the study by A. Yu. Teterin, M.V. Ryzhkov et al. (2006). he surface of the samples was strongly contaminated. Gold foil covering of the handle of an iron knife from burial 2 of the Hunnic Period cemetery at Mukhino… Reference 191 192 195 200 205 207 208 Au % 15.15 16.67 15.63 5.99 2.20 1.98 12.66 XRS - (~5 nm) (surface) Ag % Cu % 63.64 55.00 51.56 3.12 65.27 61.90 3.08 70.30 55.70 - S% 21.21 28.33 29.69 28.74 32.82 27.72 31.64 XRF (>0.5 nm) (body) Au % Ag % Cu % 77.74 19.59 2.67 84.23 14.22 1.55 65.75 31.95 2.3 69.36 29.15 1.49 68.57 30.11 1.32 65.02 33.31 1.67 62.13 35.9 1.97 3. RESULTS AND DISCUSSIONS A comparison of the data obtained with the two methods shows signiicant diferences in the main concentrations of the metal (Table 1). hus, the chemical analysis of the metal showed a variation in Au and Ag content: the surface of the foil contains Au in the range of 1.98-16.67%, Ag 51.56-70.30%; the body of the sample shows other results: Au 62.13-84.23%, Ag 14.22-35.9%. he copper content also shows variations. No other elements have been discovered (sulphur excluded). Since the method of analysis is so well-developed that we can exclude the possibility of faulty methodology, other possible explanations for these results should be considered. he data suggests that the silver difused to the surface of the samples. his assumption is supported by the diferent Au and Ag concentrations on the surface and in the body of the samples. We assume that the initial Au and Ag content in the samples represented a medium value of the ‘extreme points’ presented in Table 1. hus, the samples could have initially contained Au and Ag in equal proportions. It is not clear why this particular type of gold alloy (type III, after Rapson 1990: 127-128) was used for the gold foil covering, since it is especially diicult to shape by pressure. As we recall, the knife originates from a rich burial of a woman who had a high social status (insofar as it can be conirmed by the results of analyzing other burials at Mukhino cemetery). On the colour scale, the alloy falls within the category of bleach alloys (Cretu and van der Lingen 1999), which are low-carat according to the modern classiication (under 14 carat). Our assumption is the following: during the Hunnic period, such gold-handled knives were status objects, certain power insignia, as far as the territory in question is concerned. Findings of such knives in the Upper Don area are extremely rare. Nonetheless, the actual quality of the knife handle is low, and, in comparison with other status objects, it cannot be included in the class of high-quality artefacts. 257 Table 1: XRF and XRS data for the investigated samples (data from analysis of the samples without prior treatment of the surface; concentrations are given in atomic %). Tableau 1 : Données par FX et XRS pour les échantillons analysés (correspondant à l’analyse des échantillons sans traitement au préalable de la surface, les concentrations sont présentées en %). hus, we cannot exclude the possibility that the knife from burial 2 at Mukhino is a local copy or an item produced in the ‘metropolis’ especially for export to the ‘barbarian’ world. Further research on the subject will probably involve isotope analysis, which should clarify the origin of the metal. Acknowledgements his research was produced with the support of RSHF grant 08-01-00013a. References CRETU, C. and VAN DER LINGEN, E., 1999. Coloured gold alloys. Gold Bulletin 32(4): 115-126. MCDONALD, A.S. and SISTARE, G.H., 1978. he metallurgy of some carat gold jewellery alloys. Gold Bulletin 11(3-4): 66-73. PINASCO, M.R. and STAGNO, E., 1978. Deformation and recrystallization of a jewellery white gold alloy. Gold Bulletin 12(2): 53-57. RAPSON, W.S., 1990. he metallurgy of the coloured carat gold alloy. Gold Bulletin 23(4): 125-134. TETERIN, YU.A., RYZHKOV, M.V., MASLAKOV, K.I., VUKCEVIC, L. and PANOV, A.D., 2006. Electronic structure of solid uranium tetraluoride UF4. Physical Review B 74(4): 045101 (1-9). ENIOSOVA, N., MITOYAN, R. and SARACHEVA, N., 2008. Methods for the study of nonferrous chemical compounds, in Nonferrous and precious metals and alloys in Medieval Eastern Europe. Moscow, (in Russian). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 255-257 he wire ‘at astragals’, or beaded wire, from medieval tradition to the technique and tools used by the Roman goldsmiths Castellani in the 19th century Le fil « à astragales », ou fil perlé, de la tradition médiévale à la technique et aux outils utilisés par les orfèvres romains Castellani au XIXe siècle Maurizio Donati* Abstract: his research describes the methods of construction of beaded wire (‘at astragals’), used as decoration by goldsmiths from medieval times until the 19th century, in the work of the Roman goldsmiths Castellani. his paper focuses on ifteen original steel tools from Alfredo Castellani’s Legacy, retrieved in 1978, restored and inventoried by the author. hese tools are used to describe the technique used in the Castellani workshop. his technique is compared with the technique described in the heophilus Presbyter’s treatise. Résumé : Ce travail démontre les méthodes de construction du il perlé (« à astragales ») utilisé par les orfèvres comme décoration dès la période médiévale jusqu’au XIXe siècle, période pendant laquelle ce il est utilisé par les orfèvres Castellani, à Rome. Cet article se concentre sur 15 outils originaux en acier, provenant de la Succession d’Alfredo Castellani, récupérés en 1978 et restaurés et inventoriés par l’auteur. Grâce à ces outils, il a été possible de décrire la technique de production utilisée dans les ateliers des Castellani. Cette technique est comparée à celle décrite dans le traité de héophile. Keywords: beaded wire, astragal, Castellani, tools, heophilus Presbyter. Mots-clés : il perlé, astragale, Castellani, outils, héophile. he purpose of this research on the production of wire ‘at astragals’, or beaded wire, is to establish a complete description of its construction methods, inherited from the medieval tradition and passed on by the German or Lombard monk heophylus Presbyter (end of the 11th-early 12th century) in his treatise “Diversarum artium schedula” (Dodwell, 1961). his wire was used for goldsmiths’ decorations in ancient times, fell into disuse after the Middle Ages, and continued in the 19th century, returning especially in vogue with the work of the Roman goldsmiths Castellani (Archivio di Stato di Roma (A.S.R.) Fam. Castellani, reg. 87), prompted by the inds from archaeological excavations. his research has made use of a theoretical approach informed by the medieval treatise, and of a practical one con- * Master goldsmith, emeritus professor at the Istituto Statale d’Arte di Roma 1, honorary member of the “Università e Nobil Collegio degli Orai Gioiellieri Argentieri dell’Alma Città di Roma” – Via Alessandro Stradella, 90, 00124, Roma, Italy. (donati.m@email.it) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 259-263 260 sisting of ifteen original steel tools suited for this wire (Fig. 1), part of a collection of 1327 tools, belonging to the Legacy of Alfredo Castellani to the “Museo Artistico Industriale di Roma” (M.A.I.), donated when he died in 1930, retrieved in 1978, restored and inventoried by the author. Alfredo Castellani, son of Augusto (Roma 1865-1930), left to the M.A.I. in 1930 some drawings, tools and many other objects issuing from his father’s workshop. he Legacy went partly lost, partly dispersed among the Museo di Palazzo Barberini, Museo della Civiltà Romana and the Instituto Statale d’Arte di Roma (I.S.A. 1) (Donati, 2005). he found tools lie at present in I.S.A. 1, with the following marking: Castellani Legacy, group Diverse / from entry 890 to 904, nn, V/36-V/50. Judging from the works created in the Castellani workshop, as one discovers in the registers of the inventories, beaded wire and also the tools suited for its making were already in use starting with the years 1852/1853, coinciding precisely with the proofs worked out for granulation. he wire was usually made by a workman, a certain Belli [Gaetano] (1832-1872), as can be read in A.S.R., Famiglia Castellani, Reg. 132: “Belli, oro a stampare sdragoli, novembre 1862”. Belli is cited in the Castellani registers from number 126 to number 137, and also in numbers 37, 38, 39, and 44. He was mostly responsible for melting gold, for the production of grains of gold, for moulding various motifs, for repairs of silver objects, for the preparation of welds and earths to produce the ‘colour’, or indeed for the ‘sdragola’. In the records of Castellani’s workshop (which are all kept in the A.S.R.), it is interesting to observe the following terms in the goldsmith’s lexicon: the term ‘astragal’ (derived through Maurizio DONATI the Latin astragalus from the Greek astravgalo~: rib, bone, dice) is misspelled as the dialect term ‘sdragola’ or ‘asdragola’, as repeatedly found in inventories and catalogues. Concerning the production of beaded wire, for many years various studies (Hofmann and Davidson, 1966; Lipinsky, 1975; Formigli, 1985) have indicated two likely methods of construction with tools referable to classical or medieval jewellery, or from conjecturing or referring to descriptions present in some treatises. Speciically regarding the system adopted by the Castellani, mere hypotheses have been put forward to date, failing objective comparisons (Ogden, 2004). herefore, the recovery of their original tools was important for pinpointing their methods, as they allowed the discovery of the technique used in their workshop to produce beaded wire. For the construction of these tools, the Castellani have certainly followed the description from heophilus Presbyter’s treatise (in Caput IX of Liber tertius, “De instrumento quod organarium dicitur”), considering that the entire technical concept corresponds to the medieval annotations made by heophilus, with the exception of the percussion: the Castellani used a rocker arm, whereas in heophilus it was made “cum malleo corneo”. he typology of the Castellani tools under discussion here varies according to their use: some of them are suitable to produce apical parts of beaded wires (for hanging details of necklaces, Castellani’s Legacy, tool entry 895, V/41), others to produce continuous beaded wires (Donati, 2007), a feature that was more frequently used by the Castellani (see Fig. 2). he matrices are also diferent in dimensions and in number, size, form, and diameter of the stamped sections. Figure 1: (See colour plate) Four original steel tools from the Alfredo Castellani Legacy to M.A.I. of Rome, entries 897-898, V/43, V/44; entries 899-900, V/45, V/46 (I.S.A. 1) (Photography by M. Donati). Figure 1 : (Voir planche couleur) Quatre outils en acier originaux appartenant à la Succession d’Alfredo Castellani, M.A.I. de Rome, entrées 897– 898, V/43, V/44; entrées 899 – 900, V/45, V/46. (I.S.A. 1) (photographie par M. Donati). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 259-263 he wire ‘at astragals’, or beaded wire, from medieval tradition to the technique and tools used… Figure 2: Detail of beaded wire, with typical faults: middle groove and irregularity of beads (Photography by M. Donati). Figure 2 : Détail d’un il perlé avec les anomalies typiques : sillon au milieu et l’irrégularité des perles (photographie par M. Donati). Normally they are used two by two (Castellani’s Legacy, tool entries 894, 896, 897, 898, 899, 900, 901, 902, 903, 904): one above and the other below, each with appropriate anchor seats drawn from the same tools, and through a rocker arm (Fig. 3). One half of the sections are printed in the upper stool, the other in the bottom one. he upper one has a support shank; the bottom one, parallelepiped, shows two ‘projec- Figure 3: Original Castellani steel tool (une seule partie présentée) with proof on silver, apical part of beaded wire. Entry 895, V/41 (Photography by M. Donati). Figure 3 : Outil en acier original de Castellani (sans l’autre) avec épreuve en argent, partie trillée du il perlé. Entrée 895, V/41 (photographie par M. Donati). 261 tions’, which are correlated with the two suitable anchor seats of the upper tool, so that they can be irmly joined together. For the procedure, the golden or silver wire was laid down between the two parts, usable in superimposed position, just in the grooves of their respective seats; afterwards, the wire was twisted on its axis and was simultaneously hammered with a rocker arm; its position was maintained with the aid of a buckle of wood or metal, in order to prevent the movement of the wire from their location. Once it reached the complete form of beads, it was continued for a segment, usually in proportion to the length of the tool, leaving out only one bead in the last seat for measuring, in order to secure a regular work without imperfections. he section of the wire, before beading, could have been diverse: round, square, hexagonal, octagonal. However, the initial diameter, or the distance between the faces, should have been slightly smaller than that of the completed bead, proportionately of the order of a few tenths of millimetres (in the Castellani’s pair of tools V/43 and V/44, the wire’s section had the following initial and inal diameters: 2.43 and 2.72 mm, respectively. he upper tool (entry 897) includes 15 cells, the bottom tool (entry 898) 13. A basic measurement gave the following dimensions: 30.7 x 21.7 mm for the upper tool; 38.4 x 30 mm for the bottom tool. For tools V/46 (entry 900) and V/47 (entry 899), the initial and inal wire’s diameters are 3.05 and 3.29 mm, respectively; for the upper tool: 10 cells, 31.5 x 21.7 mm; for the bottom tool: 9 cells, 41 x 30.5 mm), in order to prevent that the continuity of the beads was interrupted with an abnormal overlap related to the increase due to supericial waves of metal. his defect occurred always in cases of initial excess of metal and was usually irreparable. However, an approach to work with a scanty amount of metal could have been inadequate for an optimum result, because, due to the shortage, it could have created a furrow in the middle, a feature that nevertheless should have been remedied by actions repeated until the disappearance of the same furrow. Moreover, there could be other causes, not considered yet, of further defects, for example the fortuitous small displacement of the two tools. A considerable number of works with this type of decoration is currently on display on the shelves of the Modern Augusto Castellani Collection in the National Etruscan Museum of Villa Giulia in Rome: almost 106 jewels (Fig. 4), one sixth of the entire collection, divided in seven or eight parts, associated with various historical periods, referred to as: “Primigeno” (Primitive), “Tirreno” (Tyrrhenian), “Etrusco” (Etruscan), “Siculo” (Siculian), “Romano” (Roman), “Medievale” (Medieval), “Rinascimento” ArcheoSciences, revue d’archéométrie, 33, 2009, p. 259-263 262 Maurizio DONATI a Figure 4: (See colour plate) Brooch with cameo of George Washington. Made by Castellani; cameo by G. Girometti. Gold, sardonyx; frame with wire ‘at astragals’. Museo Nazionale Etrusco di Villa Giulia, Roma, 85211, from Soros and Walker, 2004: Fig. 4-21 (Photography by S.A.E.M.). Figure 4 : (Voir planche couleur) Broche avec camée de Georges Washington. Fabriqué par Castellani; camée par G. Girometti. Or, sardonyx, monture avec il aux astragales. Museo Nazionale Etrusco di Villa Giulia, Rome, 85211, d’après Soros et Walker, 2004 : Fig. 4-21 (photographie par S.A.E.M.). (Renaissance), “Moderno” (Modern). We note that in the shelf “Primigeno” there are no beaded works. Another typical example of working with beaded wire can be observed on a work prepared by Augusto Castellani to decorate a silver plate with niello, representing a crib and adopted in the form of ‘Peace’ for liturgical functions, a work that later – June 4, 1865 – was donated by him to the University of Goldsmiths of Saint Eligio in Rome, in memory of his father. his work is almost unique among those of Fortunato Pio (Figs. 5a and 5b). With these notes, we hope to inspire further studies, with the purpose of advancing more and more the understanding of the goldsmith’s art, a ield both interesting from a technical-scientiic point of view and fascinating from an aesthetic-anthropological one. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 259-263 b Figure 5: (See colour plate) Fortunato Pio Castellani: “Pace”, silver and niello. Gold(?)-mounted wire to astragals, by Augusto Castellani (1865). Archivio Storico, Università e Nobil Collegio degli Oreici Gioiellieri Argentieri dell’Alma Città di Roma (Photography by Di Giacomo): (a) obverse and (b) reverse. Figure 5 : (Voir planche couleur) Fortunato Pio Castellani: “Pace”, argent et niello. Montage en or (?) avec ils aux astragales par Augusto Castellani (1865). Archivio Storico, Università e Nobil Collegio degli Oreici Gioiellieri Argentieri dell’Alma Città di Roma (photographie par Di Giacomo): (a) avers et (b) revers. he wire ‘at astragals’, or beaded wire, from medieval tradition to the technique and tools used… References DODWELL, C.R., 1961. heophilus: he various arts (de diversis artibus). London: homas Nelson & Sons. DONATI, M., 2005. Materiali già appartenuti al M.A.I. di Roma e recuperati pressa l’I.S.A. di Roma, in G. Borghini (ed.), Storia del Museo Artistico Industriale di Roma. Roma, I.C.C.D., 223240. DONATI, M., 2007. Sur quelques aspects de l’orfèvrerie Castellani dans la seconde moitié du xixe siècle. L’atelier: des prototypes à la technique, in F. Gaultier, C. Metzger, Les bijoux de la collection Campana: de l’antique au pastiche. Paris, École du Louvre, 118-119. 263 FORMIGLI, E., 1985. Tecniche dell’oreiceria Etrusca e Romana. Firenze, Sansoni Ed., 94-95. HOFFMANN, H. and DAVIDSON, P.F., 1966. Greek gold: jewelry from the age of Alexander, ed. by A. von Saldern. Boston, Museum of Fine Arts. LIPINSKY, A., 1975. Il ilo perlinato, in A. Lipinsky, Oro, argento, gemme e smalti: tecnologia delle arti dalle origini alla ine del Medioevo 3000 a. C.-1500 d. C. Firenze, Leo S. Olschki, 206208. OGDEN, J., 2004. Revivers of the lost art: Alessandro Castellani and the quest for classical precision, in S.W. Soros, S. Walker (eds.), Castellani and Italian Archaeological Jewelry. New York, Bard Graduate Center, 181-200. SOROS, S.W. and WALKER, S. (eds.). Castellani and Italian Archaeological Jewelry. New York, Bard Graduate Center. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 259-263 he jewellery from the casket of Maria Pia of Savoy, Queen of Portugal, produced at Castellani’s workshop Les bijoux du coffret de Maria Pia de Savoie, Reine du Portugal, fabriqués à l’atelier Castellani Maria José Oliveira*, Teresa Maranhas**, Ana Isabel Seruya***, Francisco A. Magro*****, hierry Borel***** and Maria Filomena Guerra***** Abstract: he casket of Queen Maria Pia of Savoy, belonging to the collection of the Palácio Nacional da Ajuda, contains a set of thirty-three pieces of gold jewellery in the archaeological style created by the Castellani workshop in Rome. his family of jewellers, fascinated by the ancient techniques of gold working, restored but also reproduced several antique pieces, giving rise to the emergence of the so-called ‘archaeological jewellery’, very popular in the 19th century, using characteristic techniques of antique jewellery such as granulation, iligree, micro-mosaic and engraving of stones in intaglio. he main purpose of this paper is to present the results of the analyses of the base metals used in the manufacture of the jewellery and also to identify the production techniques, such as mounting, joining and decoration, used by the Castellani workshop in making the jewellery for the casket of Queen Maria Pia of Savoy. For this purpose, portable equipment of examination and analysis was brought to the museum. he results obtained in situ show the use of standard wires and granules, as well as of three diferent base alloys. One element of a hairpin pair which appears to have originated from a diferent production is discussed. Résumé : Le cofret ofert à la Reine Maria Pia de Savoie, appartenant aux collections du Palácio Nacional da Ajuda, contient trente trois pièces d’orfèvrerie en or fabriquées par les ateliers Castellani. L’attrait de cette famille d’orfèvres pour les techniques de l’orfèvrerie antique les a amenés à restaurer mais aussi à reproduire plusieurs pièces anciennes faisant émerger au XIXe siècle le style archéologique. Pour fabriquer leurs bijoux, les Castellani ont utilisé diférentes techniques caractéristiques de la décoration des pièces d’orfèvrerie ancienne, dont la granulation, le iligrane, la gravure de gemmes en entaille et la mosaïque sont des exemples. Dans cette étude nous présentons les résultats analytiques obtenus pour la composition des alliages utilisés dans la fabrication des pièces du cofret de la Reine Maria Pia ainsi que la description des techniques de production des objets – montage, assemblage, décoration – utilisées par les Castellani. À cette in, des équipements portables d’examen et analyse ont été transportés au musée pour efectuer une étude in situ. Les résultats obtenus montrent l’utilisation de ils et granules de forme homogène et de trois alliages d’or. Un élément d’une paire d’épingles à cheveux, dont la production semble être diférente, est discuté. Keywords: jewellery, gold, analysis, Castellani, XRF. Mots-clés : bijoux, or, analyses, Castellani, FX. * Laboratório de Conservação e Restauro José de Figueiredo, Rua das Janelas Verdes 37, 1249 – 018, Lisboa, Portugal. (mariaoliveira@imc-ip.pt) ** Palácio Nacional da Ajuda, Largo da Ajuda, 1349-021, Lisboa, Portugal. pnajuda.(teresamaranhas@imc-ip.pt) *** Centro de Física Atómica da Universidade de Lisboa, Av. Prof. Gama Pinto 2, 1649-003, Lisboa, Portugal. (anaseruya@gmail.com) **** Academia Portuguesa da História e Associação Numismática de Portugal, Rua Angelina Vidal 40, 1170-020, Lisboa, Portugal ***** Laboratoire du Centre de Recherche et de Restauration des Musées de France, UMR171 CNRS, 14, quai François-Mitterrand, 75001 Paris, France. (maria.guerra@culture.gouv.fr) rec. Oct. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 265-270 Maria José OLIVEIRA et al. 266 1. INTRODUCTION he work of the Castellani family on antique jewellery gave rise to the ‘archaeological style’ jewellery that was very much in fashion during the 19th century (Rudoe, 1986). In 1862, they created one of their most famous productions, the jewellery casket, a good example of their skill, to be ofered by the people of Rome to Queen Maria Pia of Savoy when she married King Luís I of Portugal. he casket of Queen Maria Pia of Savoy (PNA, inv. 52545/A), held in the Palácio Nacional da Ajuda in Lisbon, Portugal, contains 33 pieces of jewellery produced by the Castellani workshop in the archaeological style. his type of jewellery emerged as a consequence of the excavations of archaeological tombs in Egypt, Greece and Italy, namely in Pompeii, at the end of the 18th century. he techniques used by Etruscan goldsmiths (between the 7th and 3rd centuries BC) greatly fascinated the Castellani family. heir curiosity and innovation led them to the revival of the ancient production techniques, particularly granulation and iligree, in which the Etruscans were experts. he casket of the Palácio Nacional da Ajuda, created in the 19th century, is probably a copy of an original example that Augustus, the irst emperor of Rome, ofered to his daughter on the occasion of her marriage. he jewels of this casket are meant to replicate the needs of a noble woman’s toilette in ancient times. Some objects are stylistically very close to known ancient pieces of jewellery, others contain ancient gold and silver coins. hese items are inspired by the iconography and mythological themes of ancient Rome: a medallion, two pairs of culinks, a stickpin, three rings, a laurel diadem, fourteen hairpins, two pairs of earrings, a bulla pendant, a brooch, a hair comb, a bracelet with Roman coins, and a chain with Greek coins (Fig. 1). Castellani did not stamp all these pieces in a regular way. His maker’s punch – monogram “CC” within a cartouche, or just “CC” without the cartouche – appears only on some items. he ancient coins used by Castellani in these jewels were classiied and dated individually, their date of issue ranging from the 5th century BC to the 4th century AD. he chain has seven Greek silver coins, issued from the 5th to the 3rd century BC; the bracelet has seven Roman silver coins from the 1st century BC, and the bulla pendant has two Roman coins, one in gold from the 2nd century AD, and one in bronze, from the 4th century AD. In spite of the work dedicated to 19th century jewellery and more speciically to the work of the Castellani (Donati, 2006; Soros and Walker, 2004; Formigli, 1993), very few studies include analytical information on the techniques and ArcheoSciences, revue d’archéométrie, 33, 2009, p. 265-270 Figure 1: (See colour plate) Casket with the jewels of Queen Maria Pia of Savoy. Figure 1 : (Voir planche couleur) Cofret avec les joyaux de la Reine Maria Pia de Savoie. alloys used by the goldsmiths (Cesareo and Von Hase, 1976; Ogden, 2004; Swaddling et al., 1991). he aim of this work is to identify and classify the techniques used at Castellani’s workshop for the production of archaeological style jewellery by studying the casket ofered to Maria Pia with nondestructive analytical-based techniques carried out in situ. he results obtained in this work should provide references for the identiication of the assumed interventions of these goldsmiths on diferent ancient gold items. 2. METHODS All the jewellery items were studied in situ at the Palácio Nacional da Ajuda. he items were examined with a portable optical microscope and an X-radiography system, and analysed with a portable X-Ray Fluorescence (XRF) equipment. All these methods are non-invasive (Guerra and Calligaro, 2003). he study of the techniques of decoration and production used by Castellani in the execution of Maria Pia of Savoy’s jewellery was performed using an optical microscope Leica he jewellery from the casket of Maria Pia of Savoy, Queen of Portugal… 267 MZ6, with a magniication of up to 40x, a digital camera, Leica DC200, and a digital X-radiography system ArtXRay, NTB GmbH (X-Ray generator Y.MBS/160-F01). he portable XRF was an E.I.S. Srl system, model XRS38, with W anode (0.40 mA, 30 kV). he spectrum treatment and the quantiication of the results were carried out using the QXAS 3.6 program developed by the International Atomic Energy Agency. he results have been normalized to 100%. A set of ternary gold alloy standards were used to calibrate the equipment. heir composition is 75% Au, 12.5% Ag, 12.5% Cu, and 75.0% Au, 6.0% Ag, 19.0% Cu, respectively. 3. RESULTS Production techniques he examination of the jewels under the optical microscope allowed the identiication of four types of wire used in the manufacture of the iligree decoration patterns: plain circular section wire, rope or cable pattern wire (obtained by twisting two plain wires together), beaded wire and helicoidal wire. he plain circular section wires are of two diferent diameters: on average, 0.49 and 1.00 mm, respectively. he cable wires vary in diameter, averaging between 0.15 mm and 0.72 mm. he beaded wire has an average diameter of 0.72 mm, and the helicoidal wire of 0.53 mm. Modern wires were produced by drawing, which means passing the wires through the holes of a draw-plate in order to obtain the required diameter, contrary to what is typically assumed for ancient wire. Drawn wires can be recognised by the seams on the surface of the wire. hose seams are longitudinal and parallel to the axis of the wire. All the wires used in the production of the jewellery contained in Maria Pia’s casket were produced with drawn wire. Figure 2 shows the striation on the surface of one wire. In addition to iligree, the jewellery contained in the casket of Queen Maria Pia of Savoy presents patterns of granulation and evidence of other decorative techniques, such as micro-mosaic and engraving. he granules are of two diferent sizes: the smaller ones have an average diameter of 1.50 mm, while the larger ones have an average diameter of about 3.00 mm. Some of the pieces show repetition of motifs, such as the leaves on the diadem and the hair comb, which were executed by stamping each individual leaf with a matrix. he hairpins with ram’s heads show skilled decoration, in which a roughened surface efect is obtained by chasing. his technique was not used for any of the other pieces. Figure 2: Detail of the seams on the surface of the wire (PNA, inv. 52563). Figure 2 : Détail des sillons sur la surface du il (PNA, inv. 52563). All the elements of the earrings and pins were executed by a skilled goldsmith, using the same techniques and the same decorative elements. However, one pair of hairpins with an imperial eagle and the inscription SPQR presents signiicant diferences in terms of technical execution between the two elements of the pair. he decoration of these pieces is essentially obtained through the application of cable patterned wire and small gold foils in the form of leaves. he ropes consist of two wires twisted together (Fig. 3). Only one of Figure 3: Details of the iligree decoration of the pair of hairpins with the imperial eagle. Figure 3 : Détails de la décoration en iligrane d’une paire d’épingles à cheveux avec l’aigle impériale. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 265-270 Maria José OLIVEIRA et al. 268 the pins has a few granules fused to it (PNA, inv. 52570). On both hairpins, the inscription was applied using plain circular section wire, although the construction of the letters difers from one element of the pair to the other, as shown by the images obtained under the optical microscope for letters ‘Q’ and ‘R’ (Fig. 4). he letters shown in Figures 4e and 4f are of better quality. However, not only the letters are technically diferent between the two elements of the pair. he crowns of leaves and the paws of the eagles, for example, also difer between the two pins (Fig. 5). Again, the elements in Figures 5c and 5d demonstrate a higher level of technical execution than those in Figures 5a and 5b. Figure 4: (See colour plate) Details of the pair of hairpins with the imperial eagle. Figure 4 : (Voir planche couleur) Détails d’une paire d’épingles à cheveux avec l’aigle impériale. he radiographic images of these two pieces (Fig. 6) highlight the diferences between them (wings, head and paws of the eagles, drums, etc.). he diferences in radiographic density are related to the diferent thickness of the metal foils used in the execution of each piece. Base-alloys Table 1 presents the results obtained by portable XRF for all the individual jewellery items belonging to Maria Pia’s casket. We were able to identify three distinct groups according to the alloys used. he irst two groups include the jewellery with the simpler decoration, while the third group includes the jewellery with more complex decoration. Figure 6: (See colour plate) X-radiography of the pair of hairpins with the imperial eagle. Figure 6 : (Voir planche couleur) Radiographie aux rayons X d’une paire d’épingles à cheveux avec l’aigle impériale. Figure 5: (See colour plate) Details of the pair of hairpins with the imperial eagle. Figure 5 : (Voir planche couleur) Détails d’une paire d’épingles à cheveux avec l’aigle impériale. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 265-270 he jewellery from the casket of Maria Pia of Savoy, Queen of Portugal… Composition Alloy 3 Alloy 2 Alloy 1 Jewellery items Table 1: Characterization of the diferent alloys by XRF. Tableau 1 : Caractérisation des diférents alliages par XRF. 269 Au (%) Ag (%) Cu (%) Culink ROMA/AMOR (PNA, inv. 52685) 90.6 8.4 1.1 Culink ROMA/AMOR (PNA, inv. 52686) 92.7 6.7 0.6 Ring with lapis lazuli intaglio (PNA, inv. 52547) 93.1 6.5 0.4 Stickpin (PNA, inv. 52475) 92.6 6.2 1.2 Hairpin pointed head (PNA, inv. 52552) 90.0 9.1 0.9 Hairpin acorn head (PNA, inv. 52555) 85.0 10.7 4.4 Chain with Greek coins (PNA, inv. 52572) 90.8 8.2 1.0 Hairpin with imperial eagle (PNA, inv. 52571) 93.6 3.3 3.1 Ring Mercvril (PNA, inv. 52546) 86.7 11.6 1.7 Diadem (PNA, inv. 52548) 86.9 11.7 1.4 Hairpin spherical head (PNA, inv. 52549) 85.9 12.3 1.7 Hairpin spherical head (PNA, inv. 52550) 86.7 11.9 1.4 Hairpin pointed head (PNA, inv. 52551) 88.1 10.6 1.3 Earrings (PNA, inv. 52558) 88.4 10.7 1.0 Earrings (PNA, inv. 52559) 88.1 10.8 1.1 Earrings crotalia (PNA, inv. 52560) 88.0 10.5 1.5 Earrings crotalia (PNA, inv. 52561) 87.9 10.7 1.4 Hair comb (PNA, inv. 52565) 88.5 9.8 1.7 Bracelet with Roman coins (PNA, inv. 52568) 88.1 10.6 1.4 Culink EROS (PNA, inv. 52443) 79.1 17.3 3.6 Ring with cornelian intaglio (PNA, inv. 52545) 82.7 13.9 3.4 Pendant bulla (PNA, inv. 52563) 80.7 16.6 2.7 Brooch NODVS HERCVLEVS (PNA, inv. 52564) 82.8 15.2 2.1 Hairpin pin ram’s head (PNA, inv. 52566) 80.4 17.1 2.5 Hairpin pin ram’s head (PNA, inv. 52567) 78.7 18.9 2.4 Hairpin with imperial eagle (PNA, inv. 52570) 83.9 13.5 2.6 In contrast to all the other pairs of jewels – for example, the pair of ram’s head pins, with the composition 79.6% Au, 18.0% Ag and 2.4% Cu, the pair of hairpins with the imperial eagle also shows a diference in composition between the two elements. he second pin (Fig. 4d) has an average percentage of gold of approximately 94%, itting the irst group, while the other pin (Fig. 4a) has an average percentage of gold of about 84%, with higher silver contents, itting the third group. he jewellery in the irst group has a composition similar to the one of the ibulae in the set of copies made by Castellani and kept at the Villa Giulia museum in Rome (Cesareo and Von Hase, 1976). he limited number of analysis results available from objects in the Villa Giulia does not match our results. 4. DISCUSSION AND CONCLUSION The decorative elements of almost all the jewellery included in Maria Pia’s casket are identical in type and dimension. his may be the result of a continuous ‘assembly line’ type of production undertaken in the Castellani workshop, which is typical of 19th century manufacturing methods. he gold/metal alloys used in the fabrication of these objects have gold contents ranging from 73 to 98%, silver contents between 2 and 24%, and copper contents from 1 to 13%. However, the higher amounts of silver and copper are only present in a few particular objects. Although all the items in Queen Maria Pia of Savoy’s casket have been made by Castellani, we can now ascertain that diferent base alloys were used. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 265-270 270 All the jewellery denotes a high manufacturing skill. In spite of the diferent alloys employed in the manufacture of this jewellery, only the hairpin with the imperial eagle presents major diferences relative to the other element of the pair, and to the entire suite of jewellery. he jewellery collection of the Palácio Nacional da Ajuda includes other items in the archaeological style, namely a gold necklace with 23 beetles and a parure of gold iligree with cornelian settings (a bracelet, a bar-brooch and a pair of earrings), made by an unknown goldsmith. In the future, we intend to submit these pieces to a similar study in order to compare the results with those of the analyses of the Castellani jewellery. References CESAREO, R. and VON HASE, F.W., 1976. Analisi di ori etruschi del VII sec. a.C. con uno strumento portatile che impiega la tecnica luorescenza X eccitata da radioisotope. Atti dei Convegni Lincei 11: 259-296. DONATI, M., 2006. Les bijoux Campana et le fonds Castellani du Museo Artistico Industriale di Roma, in F. Gaultier, C. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 265-270 Maria José OLIVEIRA et al. Metzger (eds.), Trésors antiques, bijoux de la collection Campana. Paris, Musée du Louvre, 103-107. FORMIGLI, E., 1993. Einige Fälschungen antiken Goldschmucks im 19. Jahrhundert. Archäologischer Anzeiger 3: 299-332. GUERRA, M.F. and CALLIGARO, T., 2003. he analysis of gold: manufacture technologies and provenance of the metal. Measurement in Science and Technology 14: 1527-1537. OGDEN, J., 2004. Revivers of the lost art: Alessandro Castellani and the quest for classical precision, in S.W. Soros, S. Walker (eds.), Castellani and Italian Archaeological Jewelry. New York, Bard Graduate Center, 181-200. RUDOE, J., 1986. Elizabeth Barrett Browning and the taste for Archaeological-Style Jewelry. Philadelphia Museum of Art Bulletin 83(353): 22-23. SOROS, S.W. and WALKER, S. (eds.), 2004. Castellani and Italian Archaeological Jewelry. New York, Bard Graduate Center. SWADDLING, J., ODDY, A. and MEEKS, N., 1991. Etruscan and Other Early Gold Wire from Italy. Society of Jewellery Historians 5: 7-21. South America: gold studies in the New World Amérique du Sud : étudeS deS orS du NouveAu moNde Archéologie précolombienne et analyses scientifiques : la figurine d’El Angel, une œuvre composite d’orfèvrerie de la culture La Tolita Tumaco (Équateur-Colombie) Pre-Columbian archaeology and science-based analysis : the figurine of El Angel, a composite goldwork from La Tolita Tumaco culture (Ecuador – Colombia) Jean-François Bouchard* et Maria Filomena Guerra** Résumé : Une petite igurine en or provenant du site El Angel est associée aux traditions orfèvres Tumaco La Tolita, alors qu’elle provient d’une région hors de l’aire culturelle de cette civilisation, à l’extérieur de la frange côtière. L’étude, au moyen de méthodes non-destructives, des techniques de mise en forme, assemblage et décoration utilisées pour sa fabrication montre que la igurine a subi un remontage moderne, mais que la grande majorité de ses parties principales est authentique. La composition élémentaire des alliages et surtout la présence de fortes teneurs en Fe, Pd et Pt sont typiques des ors utilisés pour l’orfèvrerie La Tolita Tumaco. La igurine provenant du site d’El Angel, en Sierra, serait ainsi un des rares exemples d’objets exportés loin de son aire géoculturelle. Abstract: A small gold igurine found at the site of El Angel, far from the coast threshold, is usually associated with the Tumaco La Tolita goldsmith traditions, in spite of a location outside the cultural area of this civilisation. he study of the igurine by non-destructive scientiic methods, evidencing the diferent steps of production, such as joining and decoration, shows that the object was remounted. However, the majority of its main parts are genuine. he composition of the alloys and especially the presence of Fe, Pd and Pt are typical of the gold used in the production of La Tolita Tumaco objects. he analytical results attest that the igurine from El Angel, in Sierra, could have been one of the exceptional objects exported far away from its geo-cultural area. Mots-clés : La Tolita Tumaco, analyse, composition, orfèvrerie, El Angel. Keywords: La Tolita Tumaco, analysis, composition, goldwork, El Angel. * Archéologie des Amériques, UMR 8096 CNRS – Maison René Ginouvès, 21 allée de l’Université, 92023Nanterre Cedex. (jean-francois.bouchard@ mae.u-paris10.fr) ** Laboratoire du Centre de Recherche et de Restauration des Musées de France, UMR171 CNRS, 14, quai François Mitterrand, 75001 Paris, France. (maria.guerra@culture.gouv.fr) rec. Oct. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 273-279 274 1. INTRODUCTION La métallurgie préhispanique est surtout devenue célèbre à travers les pillages des conquistadors. À peine arrivés dans le Nouveau Monde, ceux-ci ont mis à sac les « trésors » des indigènes. Par la suite, les fouilles clandestines des pillards, visèrent surtout les sépultures et lieux funéraires préhispaniques. Divers pays, comme la Colombie, l’Équateur, le Pérou, la Bolivie, pour ne citer que ceux de la tradition métallurgique andine, ont perdu une grande part de leur orfèvrerie, souvent fondue en lingots car les pillards n’y voyaient le plus souvent que des objets bien « monnayables » dans tous les sens du mot. Ces mêmes pillards étaient les premiers sur place pour ouvrir les tombes et les chercheurs n’y avaient accès qu’une fois le pillage terminé. Commençait alors un traic occulte où les œuvres inissaient par être vendues discrètement et se retrouvaient dans des collections privées et publiques. Ce n’est qu’une très inime partie des œuvres d’orfèvrerie de tous les musées publics qui échappent à cette règle générale. Ce rappel permet de comprendre pourquoi les archéologues sont souvent mal informés sur ces œuvres. Parmi les traditions orfèvres des Andes, l’une des plus anciennes est celle de la civilisation La Tolita Tumaco qui s’est épanouie sur le littoral Paciique de la Colombie et de l’Équateur (600 av. J.-C.-400 apr. J.-C.). La métallurgie de La Tolita Tumaco a été étudiée par divers auteurs, archéologues et archéomètres (Bergsoe, 1937 ; Verneau et Rivet, 1912-1922 ; Scott et Bray, 1994 ; Plazas, 1998 ; Bray, 2000 ; Bouchard et Usselmann, 2003 ; Scott, 2004). Un certain nombre de musées possèdent dans leurs collections des œuvres d’orfèvrerie provenant de la région La Tolita Tumaco. Le musée du Quai-Branly, Paris, nous montre ainsi une petite igurine en or, pour laquelle nous avons pu mener une étude particulière. Bien que provenant d’une région hors de l’aire culturelle Tumaco La Tolita (site El Angel, au sud-ouest de Tulcán, dans les environs du volcan Chiles, province du Carchi, Equateur), elle est considérée comme ayant été réalisée par les orfèvres de cette culture (Fig. 1). L’œuvre est composée d’un assemblage de 12 éléments majeurs, façonnés en feuilles d’or martelées, puis mis en forme. Ils sont assemblés par des ils d’aspect métallique et de couleurs diférentes. Sur le visage, deux éléments en amande représentant les yeux sont ixés et entourés de ils torsadés. L’absence de données précises dans les enregistrements du musée rendait hypothétique l’attribution culturelle suggérée pour cette œuvre. L’examen et l’analyse de la igurine du site El Angel au LC2RMF, études peu usuelles en archéologie préhispanique, mettent en évidence les techniques de mise en forme, assemblage et décoration utilisées pour sa fabrication mais aussi ArcheoSciences, revue d’archéométrie, 33, 2009, p. 273-279 Jean-François BOUCHARD, Maria Filomena GUERRA la présence de restaurations modernes par addition de certaines parties. L’analyse des alliages a aussi révélé le type d’or utilisé et permis la comparaison avec d’autres productions d’orfèvres de cette aire géoculturelle, de façon à entreprendre une recherche sur la provenance de l’or. 2. RÉSULTATS ANALYTIQUES L’examen de la igurine d’El Angel a été réalisé sous loupe binoculaire, par radiographie X et par microscopie électronique à balayage (MEB) avec un Philips XL30 ESEM en mode SE à 20 kV. L’analyse élémentaire a été efectuée à l’accélérateur AGLAE au moyen des techniques PIXE et PIGE (particle induced X-ray emission et γ-ray emission) avec un faisceau de protons de 3 MeV extrait à l’air. La technique PIXE utilise deux détecteurs de Si(Li), un dédié à la détection des éléments majeurs et l’autre, avec un iltre sélectif de 75 μm de cuivre, dédié à la mesure des éléments mineurs et traces ; la technique PIGE utilise un détecteur HPGe avec une eicacité de 30 % placé à 45° avec le faisceau (Guerra et Calligaro, 2004 ; Guerra, 2004). Le Tableau 1 présente les résultats obtenus pour les régions plates centrales des diverses parties de la igurine d’El Angel. Ces valeurs sont le résultat de deux mesures avec balayage de 200 μm2 à 500 μm2, selon les dimensions de la partie analysée. Les modes de fabrication et les alliages employés La radiographie X pratiquée sur la igurine permet de comprendre le montage des diférentes plaques obtenues par martelage et des ils utilisés (Figure 2). La plus grande épaisseur des plaques constituant la jambe et le pied gauches de la igurine semble indiquer un ajout ou un remontage postérieur à la découverte à l’aide de deux éléments modernes copiés sur la jambe droite originale. Cette hypothèse est confortée par l’absence de platine dans l’alliage des plaques suspectes. Les yeux sont formés chacun de deux éléments : un contour d’œil en forme d’amande, réalisé en il d’or torsadé et un élément, lui aussi de forme ovale, réalisé en métal de couleur gris-moyen et d’aspect plutôt terne. Dans l’enregistrement d’origine, il est seulement spéciié que l’un des yeux est en platine. Notre analyse a montré que la plaque blanchâtre de l’œil droit a bien été fabriquée avec un alliage or-platine, obtenu par frittage (en anglais sintering, p. ex. Scott et Bray, 1994). Par contre, l’analyse de la plaque noirâtre de l’œil gauche montre qu’elle est composée d’argent et de cuivre, ce serait un ajout postérieur efectué lors du remontage de la jambe gauche. Archéologie précolombienne et analyses scientiiques : la igurine d’El Angel… 275 Figure 1 (a) et (b) : La igurine composite d’El Angel (or, argent, platine). Dimensions : hauteur 13 cm ; largeur 6 cm (© C2RMF, D. Bagault). Figure 1 (a) et (b): he composite igurine found at the site of El Angel (gold, platinum, silver). Dimensions: 13 cm high; 6 cm large (© C2RMF, D. Bagault). Ces résultats sont en accord avec les conclusions de d’Harcourt publiées en 1948. Par ailleurs, la composition de l’œil droit se situe dans la région chimique déinie par les teneurs en platine, or et argent obtenues par Bergsoe (1937) et par Scott and Bray (1994) pour des objets attribués à La Tolita. La teneur en fer de l’alliage est de 3,3 % ce qui se situe dans les valeurs types avancées par Scott (1998) pour les grains de platine natif d’Amérique du Sud, typiquement de 1 à 4 %. Il semble donc que l’œil droit ait une composition conforme à celle qu’on est en droit d’attendre pour un objet Tumaco La Tolita. L’observation détaillée des yeux sous loupe binoculaire met en évidence les deux types de fabrication des ils torsadés formant le contour de l’œil. Malgré l’utilisation de ils lisses pleins dans les deux cas, l’œil en or-platine est entouré d’un il façonné par martelage, dont l’alliage possède en moyenne environ 2 % de platine, alors que l’œil en argent-cuivre est entouré d’un il fabriqué par tréilage, avec un alliage sans platine, et il est donc nécessairement moderne. Les diférents éléments constituant la igurine sont assemblés à l’aide de ils. Dans certains cas, l’assemblage est assez discret, comme pour la jonction de parties des jambes, mais, ArcheoSciences, revue d’archéométrie, 33, 2009, p. 273-279 276 plaque iligrane plaque œil gauche iligrane pied jambe gauche jambe cuisse pied jambe droite jambe cuisse avant torse arrière bras bras gauche main bras bras droit main nez tête menton main / bras gauche jambe gauche ils d’assemblage bras gauche / torse bras droit / torse œil droit Jean-François BOUCHARD, Maria Filomena GUERRA Au % Ag % Cu % 54,6 82,7 0,4 76,9 76,5 76,3 79,4 78,0 77,3 78,1 77,6 78,2 78,6 78,9 79,7 78,1 71,1 77,7 8,6 12,7 49,5 17,3 17,2 17,3 10,4 10,8 11,8 10,9 10,5 12,9 9,0 9,9 10,7 10,6 12,0 10,7 1,0 2,7 47,1 4,8 5,0 5,5 7,2 8,3 8,5 8,2 8,3 5,8 9,5 8,6 7,9 8,6 6,4 8,8 74,1 20,4 4,4 61,9 75,4 18,0 Zn % Pt % 32,3 1,8 2,8 2,8 2,3 2,7 2,0 3,0 2,6 2,4 1,6 2,7 9,7 2,6 Ru ppm 810 40 60 140 190 40 40 90 28 826 Rh ppm 6010 90 380 460 430 340 330 370 220 210 115 215 3230 270 Pd ppm 3175 200 210 260 300 320 325 390 225 270 185 290 725 355 Sn ppm Sb ppm 1650 320 170 390 85 145 30 170 50 162 168 36,2 795 645 5,5 62,8 Fe ppm 33000 385 200 650 40 730 35 50 800 45 45 550 1220 520 540 490 6735 485 770 34,4 ailleurs, il peut être plutôt grossier et, même, fait au moyen de ils tortillés qui dépassent à l’extérieur de la igurine, en particulier à la jonction des bras et des épaules (igure 3b). D’après Scott (Scott, 2004), les ils utilisés par les orfèvres de la civilisation La Tolita Tumaco sont typiquement façonnés par martelage quand le but est de produire des attaches de parties de fabrication indépendante. En ce qui concerne les ils de décoration, ils sont généralement pleins-tordus et martelés, mais aussi parfois tubulaires-tordus (ou en très ine tôle d’or roulée sur elle-même). Les ils de la igurine d’El Angel peuvent être lisses pleins, d’aspect noirâtre (igure 3a) ou jaune, et aussi tubulaires de couleur jaune. Leur analyse montre que les ils jaunes sont fabriqués à partir d’alliages d’or et les ils noirâtres sont en laiton. Ces caractéristiques suggèrent l’hypothèse d’un remontage des parties principales, après la découverte de l’œuvre avec un matériau d’origine non archéologique, au moins en partie. Le fait que le dernier type d’alliage soit inconnu à l’époque préhispanique conirme bien l’hypothèse d’un remontage moderne. Hypothèses de provenance de l’or La présence d’éléments du groupe du platine dans les ors de la région du Nord-Ouest de l’Amérique du Sud, ainsi que la composition du platine natif, sont connus depuis de longues années, grâce aux analyses efectuées au xixe siècle ArcheoSciences, revue d’archéométrie, 33, 2009, p. 273-279 104 1014 610 Tableau 1 : Résultats obtenus par PIXE pour les diférentes parties de la igurine d’El Angel. Table 1: Compositional results on all the analysed parts of the El Angel igurine by PIXE. par Berzelius pour la Colombie (Berzelius, 1928) et par Wolf pour Esmeraldas (Wolf, 1912). L’analyse des diférents éléments supposés originaux de la igurine d’El Angel montre qu’ils ont été fabriqués à partir d’un alliage dont la composition moyenne est de 77,7±2,1 % d’or, 10,8±1,0 % d’argent, 8,0±1,0 % de cuivre et 2,5±0,4 % de platine (remarquons la présence d’inclusions de platinoïdes, identiiées au MEB-EDS et constituées d’Ir, Pt et Os). La comparaison de cet alliage avec la composition des objets publiés par Bergsoe, 1937 ; Scott et Bouchard, 1988 ; Rovira, 1994 ; Scott et Bray, 1994 ; Estevez, 1998 ; Barrandon et al., 2004 et Valdez et al., 2005, montre que la igurine s’approche des réalisations des orfèvres de la civilisation La Tolita Tumaco. Il s’avère donc que cet alliage est bien diférent de celui utilisé dans la fabrication des parties que nous supposons être des ajouts modernes. En efet, les ils d’assemblage, le il de l’œil gauche et les plaques de la jambe et pied gauches ont une composition moyenne de : 75,8±1,1 % d’or, 18,1±1,3 d’argent et 5,1±0,5 % de cuivre ; le platine ne pourrait être présent qu’à des quantités inférieures aux limites de détection. La présence de platinoïdes dans les parties jugées antiques semble ainsi suisante pour indiquer que la plupart des éléments constituant la igurine paraissent bien originaux. De façon à vériier l’attribution de l’or utilisé dans la fabrication des parties originales aux gisements alluvionnaires de la côte, nous avons comparé nos résultats pour les éléments Archéologie précolombienne et analyses scientiiques : la igurine d’El Angel… 277 Figure 3 : (Voir planche couleur) Images sous loupe binoculaire des ils : (a) noirâtres et (b) tortillés (© C2RMF, D. Bagault). Figure 3: (See colour plate) Low magniication micrographs under the binocular of the wires : (a) blackish and (b) wriggled (© C2RMF, D. Bagault). Figure 2 : (Voir planche couleur) Radiographie aux rayons X de la igurine d’El Angel (© C2RMF, T. Borel). Figure 2: (See colour plate) X-radiography of the igurine from El Angel (© C2RMF, T. Borel). traces avec les compositions de divers objets de La TolitaTumaco et d’Esmeraldas publiés par Barrandon et al., 2004 et par Valdez et al., 2005. Parmi les éléments traces les plus signiicatifs nous avons sélectionné le palladium et le platine. Sur la Figure 4 ces deux éléments sont corrélés (corrélation typique des gisements alluvionnaires) pour les objets de La Tolita, des objets fabriqués par les civilisations côtières, ainsi que pour la igurine d’El Angel. Les objets provenant des régions de la Sierra possèdent des teneurs plus basses et plus dispersées en palladium et platine. Ceci peut être expliqué par l’utilisation des gisements aurifères de type primaire ainsi que par un possible mélange d’or alluvionnaire avec de l’or ilonien. L’or de la statuette d’El Angel s’approche bien de l’or des alliages utilisés par les diférents groupes d’orfèvres qui ont travaillé le métal à La Tolita. Non seulement la statuette d’El Angel présente des analogies stylistiques de cette région côtière, comme il ressort de l’étude morphologique et technologique, mais surtout les parties originales de cette œuvre sont très proches en composition des divers produits d’orfèvrerie réalisés à partir des matières premières obtenues dans les mêmes types de gisements aurifères. 3. CONCLUSION Les analyses et examens menés sur la igurine d’El Angel suggèrent bien qu’elle a fait l’objet de diverses interventions modernes, en particulier un remontage maladroit et l’ajout de parties non préhispaniques. Sont suspects : l’œil gauche (en alliage d’argent-cuivre) et le il du contour de cet œil gauche, la jambe gauche et, sans doute, le pied gauche, la ArcheoSciences, revue d’archéométrie, 33, 2009, p. 273-279 278 Jean-François BOUCHARD, Maria Filomena GUERRA Figure 4 : Représentation des concentrations en Pd et en Pt (en ppm) pour la igurine d’El Angel et pour divers objets de La Tolita Tumaco et d’Esmeraldas publiés par Barrandon et al., 2004 et par Valdez et al., 2005 et des monnaies sud américaines (Guerra, 2004b). Figure 4: Pd and Pt contents (in ppm) for the igurine from El Angel and for objects from La Tolita Tumaco and Esmeraldas published by Barrandon et al., 2004 and Valdez et al., 2005 and South America coins (Guerra, 2004b). plupart des ils d’assemblage (voire tous). Nous pouvons aussi remettre en question l’articulation et la mobilité des parties de la igurine : elles n’étaient qu’une simple conséquence de la restauration moderne. À l’origine, elle avait certainement été assemblée fermement et il n’était pas possible de lui donner diverses postures en déplaçant ses membres ou sa tête. Il n’y avait donc pas d’articulation mais un assemblage. Cela est d’ailleurs plus conforme au montage d’autres igurines La Tolita Tumaco connues. Pour tenter d’évaluer l’époque de l’intervention moderne, on notera que les parties suspectes, de par leurs compositions élémentaires, correspondent à une métallurgie européenne. L’analyse a donc bien conirmé des soupçons sur quelques éléments de la igurine d’El Angel, mais, la grande majorité des parties principales semblent authentiques et typiques par leur composition élémentaire de cette orfèvrerie La Tolita Tumaco. La igurine, si elle provient bien du site d’El Angel, en Sierra, n’a donc pas été trouvée dans la frange côtière correspondant au territoire La Tolita Tumaco. Pourtant par sa composition, l’or utilisé dans la fabrication de la igurine possède les éléments caractéristiques espérés pour un travail d’orfèvrerie de cette zone culturelle. Cette igurine serait ainsi un des rares exemples d’objets exportés loin de son aire géoculturelle. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 273-279 Dans l’état actuel de nos connaissances, il reste néanmoins inexplicable qu’elle soit ainsi parvenue dans les hautes terres. Remerciements Nous remercions Christiane Naffah, alors chargée du Chantier des Collections du Musée du Quai Branly, MarieFrance Fauvet-Berthelot, responsable des collections Amérique du Musée de l’Homme, et hierry Borel, Dominique Bagault et l’équipe AGLAE du C2RMF, qui ont rendu possible l’étude. Bibliographie BARRANDON, J.-N., ESTÉVEZ, P. et VALDEZ, F., 2004. Identiicación mineralogica de las Fuentes del oro precolombino utilizado en la metalurgia prehispanica del Ecuador. Anejos de AEspa XXXII, 405-416. BERZELIUS, J.-J., 1838. Traité de chimie, Tome 1er, Société Typographique Belge, A. Wahle et C., Bruxelles. BERGSOE, P., 1937. he Metallurgy and Technology of Gold and Platinum among the precolumbian Indians. (Translated by F. C. Reynolds). Ingeniorvidenskabelige Skrifter n° A44, Archéologie précolombienne et analyses scientiiques : la igurine d’El Angel… Naturvidenskabelige Samfund i Kommission hos GEC Gad, Copenhagen. BOUCHARD, J.-F. et USSELMANN, P., 2003. Trois millénaires de civilisation entre Colombie et Équateur, La région Tumaco La Tolita, Paris, CNRS. BRAY, W., 2000. Malagana and the goldworking tradition of Southwest Colombia. Colombian Gold : technology, style and iconography, he Trustees of he British Museum, 94-111. D’HARCOURT, R., 1948. Archéologie d’Emeraldas et de Manabí, Equateur, Journal de la Société des Américanistes 37, 319-325. ESTEVEZ DE ROMERO, P., 1998. Platino en el Ecuador precolombino, Boletín Museo del Oro 44-45: 159-182. GUERRA, M. F., 2004a. Fingerprinting ancient gold with proton beams of diferent energy, Nucl. Instrum. and Methods B 226, 185-198. GUERRA, M. F., 2004b. he circulation of South American precious metals in Brazil in the end of the 17th century, Journal of Archaeological Sciences, 31: 1225-1236. GUERRA, M. F. et CALLIGARO, T., 2004. Gold traces to trace gold. Journal of Archaeological Sciences, 31: 1199-1208. PLAZAS, C., 1998. Cronología de la metalurgia colombiana, Boletín Museo del Oro 44-45: 3-78. ROVIRA, S., 1994. Pre-hispanic goldwork from the Museo de América, Madrid : a new set of analysis. Archaeometry of Pre-Columbian 279 Sites and Artefacts, D. A Scott and P. Meyers ed., he Getty Conservation Institute, Los Angeles: 323-350. Scott, D. A., 2004. Gold and platinum metallurgy of La Tolita : a metalworking centre of the Paciic lowlands of Ecuador, Anejos de AespA XXXII, 63-82. SCOTT, D. et BRAY, W., 1980. Ancient Platinum Technology in South America, Platinum Metals Review 24: 147-157. SCOTT, D. A., 1998. Technical examination of ancient South American metals: some examples from Colombia, Peru and Argentina, Boletín Museo del Oro 44-45: 79-115 SCOTT, D. et BOUCHARD, J.-F., 1988. Orfebreria prehispánica de la llanuras del paciico de Ecuador y Colombia, Boletín Museo del Oro 22: 2-16. SCOTT, D. A. et BRAY, W., 1994. Pre-Hispanic platinum alloys: their composition and use in Ecuador and Colombia. Archaeometry of Pre-Columbian Sites and Artefacts, D. A. Scott and P. Meyers ed., he Getty Conservation Institute, Los Angeles: 285-322. VALDEZ, F., GRATUZE, B., YÉPEZ, A. et HURTADO, J., 2005. Evidencia temprana de metalurgia en la Costa Paciica ecuatorial, Boletín Museo del Oro 53: 1-9. VERNEAU, R. et RIVET, P., 1912-1922. Ethnographie de l’ancien Equateur, (2 vol.) Ministère de l’instruction publique, Paris. WOLF, T. H., 1892. Geografía y geología del Ecuador, Casa de la Cultura Ecuatoriana. 2e ed. 1975: 363-366. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 273-279 Pre-Columbian alloys from the royal tombs of Sipán and from the Museum of Sicán: non-destructive XRF analysis with a portable equipment Alliages précolombiens des tombes royales de Sipán et du musée de Sicán : analyse non-destructive avec un système FX portable Roberto Cesareo*, Angel Bustamante**, Julio Fabian**, Cristina Calza***, Marcelino dos Anjos***, Ricardo T. Lopes***, Walter Alva****, Luis Chero****, Fidel Gutierrez****, Maria del Carmen Espinoza****, Rosendo Rodriguez****, Marco Seclen****, Victor Curay*****, Carlos Elera***** and Izumi Shimada***** Abstract: On the north coast of present-day Peru, approximately between 50 and 700 AD, lourished the Moche civilization. It was an advanced culture, and the Moche were sophisticated metalsmiths. he Moche metal working ability was impressively shown by the excavations of the “Tumbas Reales de Sipán”, discovered by W. Alva and co-workers in 1987. he Sicán culture is a successive civilization (750-1375 AD) which extended as far as present day Piura in the north and Trujillo in the south. he Sicán culture was strongly inluenced by the Moche culture, particularly in its metallurgical development (Shimada and Griin, 1994). he metal objects from the Museums of Sipán and Sicán were analyzed with a portable equipment which uses energy-dispersive X-ray luorescence (XRF). his portable equipment is composed of a small size X-ray tube and a Si-PIN thermoelectrically cooled X-ray detector. It was determined that the analyzed artefacts are composed of gold, silver and copper alloys, of gilded copper, of silvered gold, and of tumbaga, the last being a poor gold alloy enriched at the surface by depletion gilding, i.e. removing copper from the surface. In the case of gold, silver and copper alloys, their composition was determined by EDXRF analysis employing standard alloys. In the case of gilded copper, silvered copper and of tumbaga, the ratio Cu(Kα/Kβ) was accurately determined from the X-ray spectra, irst in order to clearly distinguish them, and subsequently in order to determine the thickness of the gilding. Mean values of 0.4 μm and 2.7 μm were determined for gilded copper and tumbaga, respectively. For gilded silver, the ratio Ag(Kα/Kβ) was measured. he ratios Au-Kα/Cu-Kα, Ag-Kα/Cu-Kα and Au-Kα/Ag-Kα (gilded copper, silvered copper, and gilded silver, respectively) also depend on the thickness of gilding or silvering, and were employed to this aim. Résumé : Sur la côte nord de l’actuel Pérou, entre environ 50 et 700 AD, prospérait la civilisation Moche. Cette culture était très avancée et les Moches étaient des métallurgistes avertis. Leur talent pour travailler le métal est illustré par les fouilles du site des “Tumbas Reales de Sipán”, découvert par W. Alva et collaborateurs en 1987. La culture Sicán est une civilisation qui se succède de 750 à 1375 AD et qui s’étend au nord jusqu’à Piura et au sud jusqu’à Trujillo. La culture Sicán a été fortement inluencée par la culture Moche, et tout particulièrement leur métallurgie (Shimada et Griin, 1994). Les objets métalliques des musées de Sipán et Sicán ont été analysés avec un spectromètre portable à luorescence X à dispersion d’énergie. Cet équipement est constitué d’un tube à rayons-X miniaturisé et d’un détecteur de rayons-X de type Si-PIN à refroidissement thermoélectrique. Nous avons pu montrer * Dip. di Matematica e Fisica, Università di Sassari – Sassari, Italy. (cesareo@uniss.it) ** Universidad Nacional Mayor de San Marcos – Lima, Perù. (angelbd1@gmail.com) *** COPPE, Universidade Federal do Rio de Janeiro – Rio de Janeiro, Brasil. (ricardo@lin.ufrj.br) **** Museo “Tumbas Reales de Sipán” – Lambayequee, Perù. (museosipan@museosipan.com) ***** Museo de Sicán” – Ferrañafe, Perù. (museosican@speedy.com.pe) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 281-287 282 Roberto CESAREO et al. que les objets sont fabriqués avec des alliages d’or, argent et cuivre, avec du cuivre et de l’argent doré ainsi qu’avec du tumbaga, alliage d’or enrichi par mise en couleur, c’est-à-dire par appauvrissement en surface du cuivre. Les compositions des alliages en or, argent et cuivre ont été déterminées par ED-FX au moyen de standards d’or. Pour diférencier les cuivres et les argents dorés ainsi que les tumbaga et ensuite déterminer l’épaisseur de la dorure, le rapport Cu(Kα/Kβ) a été déterminé avec précision à partir des spectres à rayons X. Des valeurs moyennes de, respectivement, 0,4μm et 2,7μm ont été déterminées pour les cuivres dorés et les tumbaga. Pour les argents dorés, c’est le rapport Ag(Kα/Kβ) qui a été mesuré. Les rapports Au-Kα/Cu-Kα, Ag-Kα/Cu-Kα et Au-Kα/Ag-Kα (cuivres dorés, cuivres argentés et argents dorés, respectivement) dépendent aussi de l’épaisseur de la dorure et de l’argenture. Keywords: X-ray luorescence, gold leaf thickness, Moche culture. Mots-clés : Fluorescence à rayons X, épaisseur de feuille d’or, culture Moche. 1. IntroductIon Objects belonging to the Moche civilization (Vetter Parodi, 2006; Alva, 2006) were analyzed in the past using various destructive techniques (Andrade et al., 2005; Burger, 1992; Lechtman, 1998; Ruvalcada Sil, 2005; Saettone et al., 2003; Scott, 2000); fragments of gold and silver artefacts (and also of silvered gold) from Loma Negra, Peru, were accurately studied and analyzed by Schorsch (1998) by employing EDXRF attached to a scanning electron microscope, and wave length dispersive X-ray Spectrometry. he gold objects showed the following composition: Au ~ 80%, Ag = 10-20%, Cu = 5-15%. he silver objects showed a high Ag content, of about 97-99%. Copper represents the rest of the composition, to 100%. An interesting and unusual feature is the case of silvered gold, with a silver sheet measured to have a thickness of about 5 μm. Fragments from 17 Moche objects from the “Museo Tumbas Reales de Sipán” have been analyzed by Hörz and Kallfass (2000), using various techniques. hese authors were able to identify: a) gilded copper objects: they are characterized by a thin gold ilm (2-6 μm). he coatings consist of a gold-copper alloy containing some silver; b) coppersilver alloys: fragments from several human head shaped beads have been analyzed, showing a mean composition of 79% Cu, 20% Ag, and 1% Au; c) copper-gold-silver alloys (tumbaga): fragments from a headdress, a chin ornament, an ornamental disc, and ornamental beads were analyzed; the average compositions were calculated to be varying between 20-60% Cu, 35-65% Au and 6-15% Ag. he alloy composition is strongly dependent on the distance from the surface. For the systematic analysis of Sipán and Sicán metal objects, the use of non-destructive techniques may be proposed, and, among them, energy-dispersive X-ray luorescence analysis (EDXRF) appears to be the most suited, ArcheoSciences, revue d’archéométrie, 33, 2009, p. 281-287 because it is non-destructive, multi elemental, reliable, rapid, and may be carried out in situ (Cesareo et al., 2004). his method is able to quantify the composition of a gold or silver alloy when standard samples are used. EDXRF analysis provides reliable results regarding the concentration of high carat gold alloys with a reduced quantity of copper, and also in the case of high concentration copper or silver alloys. In the case of tumbaga, of copper-rich gold alloys, of gilded copper, and of copper-rich silver alloys, the results provided by EDXRF are incomplete or erroneous. EDXRF is a surface analysis, and it is useful in this case because the respective thickness of the alloy is of the order of a few μm to a maximum of tens of μm; because of this peculiarity, EDXRF is able to distinguish a gold alloy from gilded copper or tumbaga, and a silver alloy from gilded silver, by using the internal ratio of Cu and Ag-lines. he method is also able to determine the gold thickness value in the case of gilded copper or tumbaga gold. 2. ExpErImEntal sEt-up he portable equipment employed for the analysis of Sipán and Sicán alloys is composed of an X-ray tube and a Si-PIN detector, both manufactured by AMPTEK. he X-ray tube has a weight of 300 g, length of 17 cm – including the collimator – and diameter of 3.7 cm. It has an Ag-anode, and works at 30 kV and 100 μA maximum voltage and current. X-rays from the tube irradiate areas of about 5 mm diameter, when the object is at a distance of ~2 cm. he X-ray beam intensity is too high when alloys are analyzed, because the Si-PIN detector is able to process only a few thousands of photons/sec without losing energy resolution. he X-ray beam is therefore collimated with a brass cylinder 2 cm long and with an internal hole of 2 mm 283 Pre-columbian alloys from the royal tombs of Sipán and from the Museum of Sicán… 3. thEorEtIcal background Quantitative analysis of gold, silver and copper alloys Figure 1: Experimental setup, showing, on the left, the Eclipse II X-ray tube and the Si-PIN detector (both collimated with a brass cylinder), measuring a Moche golden mask in the Museum “Tumbas Reales de Sipán”. Figure 1 : Coniguration expérimentale avec, sur la gauche, le tube à rayons-X Eclipse II et le détecteur Si-PIN (les deux collimatés avec un cylindre en laiton) pendant l’analyse du masque d’or Moche au musée « Tumbas Reales de Sipán ». in diameter. Furthermore, in order to excite silver in a more eicient manner, the X-ray tube output is also iltered with about 0.1 mm Ti. he X-ray detector is a thermoelectrically cooled Si-PIN, with 300 μm thickness, 7 mm2 area of the Si-crystal, and a thin Be-window. his detector has an energy resolution of about 180 eV at 5.9 keV, and an eiciency of 90%, 25% and 8% at 10, 20, and 30 keV, respectively. It has a weight of 150 g and a length of 14 cm. It is also collimated. he measuring time was of approximately 100 sec, according to sample composition and geometry. Standard gold and silver alloys, with certiied Au, Ag and Cu concentrations, were employed for calibration and for the quantitative determination of the alloy composition. In order to measure the gilding thickness of gilded gold or silver, the Cu(Kα/Kβ) or Ag(Kα/Kβ) ratios, and the (Au-Lα/Cu-Kα) or (Au-Lα/Ag-Kα)-ratios were employed. Au leaves and Ag foils were employed for calibration (the foils were each 0.125 μm and 0.28 μm thick for Au and Ag, respectively). hick sheets of pure Cu and Ag were also employed. hickness values were tested by transmission measurements with mono-energetic X-rays. Artefacts of very diferent sizes, compositions and surfaces were analyzed. It is therefore very diicult to reproduce a ixed geometry. For this reason, determination of the alloy components on the basis of the fundamental parameters method is diicult, and an alternative approach was preferred, using the intensity ratio of two components (for example Cu/Au, assuming that Au(%)+Cu(%)+Ag(%) = 100), which is not dependent on the geometry. Other elements present are determined by using fundamental parameters. Gilding thickness in gilded Cu (or Ag) and in tumbaga through determination of (Kα/Kβ), (Lα/Lβ)- or (Au-Lα/Cu-Kα)-ratio, and (Au-Lα/ Ag-Kα)-ratios (Kα/Kβ), (Lα/Lβ), (Lα/Lγ)-ratios altered by self-attenuation The Kα/Kβ, Lα/Lβ and Lα/Lγ-ratios are tabulated (Cesareo, 2000; Markowicz, 1992). For example, for copper, Kα/Kβ = 7.4, and for gold, Lα/Lβ = 1. hese values are valid for an ininitely thin sample, corresponding to a thickness < 1μm, < 0.5μm and < 0.2μm for Cu, Ag and Au, respectively. For larger thicknesses, self attenuation efects must be considered (Cesareo and Brunetti, 2008). (Kα/Kβ), (Lα/Lβ), (Lα/Lγ)-ratios of an element altered by attenuation by a second element When a sheet of metal (a), for example copper, of ininite thickness, is covered by a sheet of another metal (b), for example gold, then the ratio (Kα/Kβ)a or (Lα/Lβ)a is altered because of the attenuation of the covering sheet: (Kα/Kβ) = (Kα/Kβ)s.a.exp-[μ1-μ2] d (1) (Kα/Kβ)s.a. is the ratio of metal (a) according to possible self-attenuation; μ1 and μ2 (cm-1) are the attenuation coeficients of element (b) at the energy of Kα and Kβ rays of element (a); d is the thickness (in cm) of the sheet of element (b) (Cesareo and Brunetti, 2008; Cesareo et al., 2009). Similar equations may be calculated for Lα/Lβ and Lα/Lγ-ratios. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 281-287 284 Figure 2: Gilded copper: attenuation of Cu(Kα/Kβ) ratio by Au (or Au alloys) vs Au thickness. Figure 2 : Cuivre doré: atténuation du rapport Cu(Kα/Kβ) par l’Au (ou les alliages d’Au) versus l’épaisseur d’Au. (Au-Lα/Cu-Kα)-ratio versus Au thickness Another way to experimentally determine, from the X-ray spectrum, the thickness of the second element (b) assuming that the irst element (a) has an ininite thickness is the use of the X-ray ratio of the two elements, for example, the ratio (Au-Lα/Cu-Kα). his ratio, for two generic elements at ixed incident energy and geometrical arrangement, is provided by the following formula: N b /N a ∝ (µ a0 +µ aa )/(µ b0 +µ bb )[1-exp-{µ 0b +µ bb }ρ b d b ] (2) exp{µ0b+µba}ρbdb −µa0 or µb0 (in cm2/g) are the total mass attenuation coeficients of elements (a) and (b) at incident energy; µab (in cm2/g) indicates the mass attenuation coeicient of element (a) at energy of the involved line of element (b); ρb (in g/ cm3) is the density of element b. In Equation (3), the proportionality should be quantitatively determined according to experimental data. Figure 3 represents theoretical values (from Eq. 3) and experimental measurements values for the (Au-Lα/Cu-Kα)-ratio versus Au thickness, assuming Cu with ininite thickness. 4. rEsults Alloys from “Tumbas Reales de Sipán” About 50 objects from the “Tumbas Reales de Sipán” were analyzed, made of gold, gilded copper, tumbaga, silver and ArcheoSciences, revue d’archéométrie, 33, 2009, p. 281-287 Roberto CESAREO et al. Figure 3: heoretical (squares) and experimental values for (AuLα/Cu-Kα) ratio vs Au thickness, assuming Cu with ininite thickness. Figure 3 : Valeurs théoriques (carrés) et valeurs expérimentales du rapport (Au-Lα/Cu-Kα) versus l’épaisseur d’Au, en assumant une épaisseur ininie pur le Cu. copper alloys, the majority of them originating from the tomb “Señor de Sipán”. Gold objects composition Several objects are made of gold (an example is shown in Fig. 4) and are characterized by Au, Cu and Ag as main components. In some samples, traces of Fe, Zn and Br are visible. he mean results of the EDXRF analysis carried out on gold objects are the following: Au = 69.5 ± 7%; Ag = 21 ± 4%; Cu = 9.5 ± 5%. Gilded copper: analysis and Au-thickness measurement Only a few objects are surely of gilded copper. hey were identiied by the exclusive presence of Cu in some analyzed areas, and by a deteriorated surface. In many cases, it was possible to clearly determine the Au leaf thickness from Cu (Kα/Kβ) and (NAu-L/NCu-K) ratios. Several sheets of gilded copper were analyzed. hey are characterized by a Cu(Kα/Kβ) ratio of 6.1 ± 0.1, corresponding to a gilding thickness of 1.2 ± 0.5 μm. From the (Au-Lα/Cu-Kα) ratio of 0.1 results a value of ~ 0.5 μm, however. A beautiful mask of gilded copper was analyzed in detail, showing the following composition: Au ~97.5%, Ag ~2.5%. he gilding thickness was measured to be ~ 0.5 μm. Cu could not be determined, because it was present below the gilding. Pre-columbian alloys from the royal tombs of Sipán and from the Museum of Sicán… Figure 4: Peanut heads made of gold, belonging to a necklace. he average composition is: Au = 58%, Ag = 26%, Cu = 16%. Figure 4 : Perles en forme de cacahuète exécutées en or et appartenant à un collier. D’après la Ref. (2). La composition moyenne est: Au = 58%, Ag = 26%, Cu = 16%. Tumbaga (or gilded copper) mean composition and gold thickness measurement he majority of the gold alloys were identiied as tumbaga (an example is shown in Fig. 5), which behaves in a similar manner as gilded Cu for EDXRF analysis. he ‘gold-equivalent’ surface thickness can be determined from Cu-Kα/ Kβ and from (NAu-L/NCu-K) ratios (see Section 3). he mean Au-Cu-Ag concentration and Au thickness values are: Au = 60 ± 10%; Cu = 30 ± 9%; Ag = 10 ± 4%. he (NAu-L/NCu-K) ratio = 1.44 ± 0.7, corresponding to an Au thickness of 3.10 ± 0.7 μm; he (Cu-Kα/Kβ) ratio = 5.35 ± 0.5, corresponding to an Au thickness of 2.5 ± 1.4. Silver objects he EDXRF analysis of silver objects shows that the silver content is relatively high, and that it systematically contains gold. he following mean concentration values were determined: 285 Figure 5: Leg protector made of tumbaga. he ratio Cu(Kα/Kβ) is 4.8, corresponding to an Au thickness of 3.1 μm Figure 5 : Protecteur de jambe en tumbaga. Le rapport Cu(Kα/Kβ) est 4,8, correspondant à une épaisseur d’Au de 3,1 μm Turquoises Many of the gold objects include turquoise, which is a hydrated phosphate of aluminium and copper. he turquoise shows an occasional presence of Zn, Fe and Cr as impurities, resulting in deviations from the blue colour. Results pertaining to Sipán turquoise show a systematic presence of Fe and Zn, at an average concentration of 10% and 8.5%, respectively. Alloys from the museum of Sicán About 20 objects from the Museum of Sicán (Shimada and Griin, 1994) were analyzed, the majority of which were of gilded copper; others were made of gold, tumbaga, silver and copper alloys. Objects made of gold he following mean values could be determined: Au = 62 ± 4% ; Ag = 32 ± 3% ; Cu = 6 ± 2%. Ag = 92 ± 4%; Cu = 5 ± 2.5%; Au = 3 ± 1.5%. Copper objects Parts of a few objects are made of copper, composed of about 99% Cu, and traces of Fe and Ni. Objects made of gilded copper he following mean values could be determined for the gilding: Au = 67%; Ag = 33%. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 281-287 286 Roberto CESAREO et al. Objects made of tumbaga Acknowledgements A few objects are of uncertain composition; they could either be made of gilded copper or of tumbaga. In fact, the gilded copper objects are identiied because of the altered Cu(Kα/Kβ) ratio, and because of the presence of highly corroded areas on almost pure copper. In other cases, the ratio Cu(Kα/Kβ) is altered, but no corroded areas were detected. One artefact, a beautiful mask, is certainly made of tumbaga, whose concentration and characteristic thickness parameters are the following: his work was partially carried out within the framework of the project IAEA-CRP (G4.20.02/1371) “Uniication of nuclear spectrometry: integrated techniques as a new tool for material research”. J. Fabian expresses his gratitude to the International Centre for heoretical Physics Abdus Salam for a 5 months grant at the University of Sassari. Au = 34 ± 6%; Cu = 57 ± 5%; Ag = 7 ± 1.5%. he Cu(Kα/Kβ) ratio = 4.9 ± 0.3, corresponding to an Au thickness of 3.8 ± 0.9 μm. he (Au-L/Cu-K) ratio = 1.05 ± 0.02, corresponding to an Au thickness of 4.5 ± 0.8 μm. Objects made of silver Only one object is made of silver, i.e. a brooch in the form of a monkey, with the following composition: Ag = 94%, Cu = 3.7%, Au = 0.5%, Pb = 0.8%, Br = 1%. Objects made of copper he following mean values could be determined: Cu = 98 ± 1%; Fe = 1.2 ± 0.5%; As = 0.8 ± 0.5%. 5. conclusIons EDXRF analyses of precious objects from Sipán and Sicán demonstrate their complexity and variety; they are made of the following alloys: gold, gilded copper, tumbaga, silver, silvered copper, silvered gold, copper, and so on, and their nature is not always easy to identify. From the metallurgical point of view, a comparison between the Moche (50-700 AD) and Sicán (750-1375) civilizations shows that: – Moche precious objects seem to be more sophisticated in terms of beauty and technology; – he average compositions of gold and silver are similar; – he Moche civilization largely used tumbaga-gold (representing more than 50% of the ‘gold’ artefacts), while Sicán used more gilded copper; – he Moche employed almost pure copper; the Sicán civilization had a knowledge of arsenical copper. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 281-287 References Alva,W., 2006. SIPAN : descrubrimiento e investigation. Lima: Quebecor World Perù S.A. andradE, E., murIllo, g., polIcronIadEs, r., acosta, l., Zavala, E.p., rocha, m.F. and cEntEno, S.A., 2005. IBA analysis of some precolumbian gilded copper samples. Nuclear Instruments and Methods in Physics Research B 240: 570-575. BurgEr, R.L., 1992. Chavín and the origin of Andean civilization. London, hames and Hudson Ltd. cEsarEo, R., 2000. X-Ray Physics, La Rivista del Nuovo Cimento. Bologna, Ed. Compositori. CEsarEo, r., brunEttI, a., castEllano, a. and rosalEs, m.a., 2004. Portable equipment for X-ray luorescence analysis, in K. Tsuji, J. Injuk, R.E. van Grieken (eds.), X-Ray Spectrometry: Recent Technological Advances. Chichester J. Wiley & Sons, 307-341. CEsarEo, r. and brunEttI, A., 2008. Metal sheets thickness determined by energy-dispersive X-ray luorescence analysis. Journal of X-Ray Science and Technology 16(2): 119-130. CEsarEo, r., rIZZutto, m.a., brunEttI, a. and rao, D.V. Metallocation and thickness in a multilayered sheet by measuring Kα/Kβ, Lα/Lβ and Lα/Lγ X-ray ratios. Nuclear Instruments and Methods in Physics Research B 267: 2890-2896. hőrZ, g. and kallFass, M., 2000. he treasure of Au and Ag artefacts from the Royal Tombs of Sipán, Peru. Materials Characterization 45: 391-420. lEchtman, H., 1998. New perspectives on Moche Metallurgy: techniques of gilding copper at Loma Negra, Northern Peru. American Antiquity 47(1): 3-30. MarkowIcZ, A.A., 1992. X-ray Physics, in R.E. van Grieken, A.A. Markowicz (eds.), Handbook of X-ray Spectrometry. New York: M. Dekker Inc., 1-92 Ruvalcada sIl, J.L., 2005. PIXE analysis of pre-Hispanic items from ancient America, in M. Uda, G. Demortier, I. Nakai (eds.), X-rays for archaeology, Dordrecht, Springer, 123-149. saEttonE , E.a.o., matta , J.a.s., a lva , w., c hubacI , J.F.o., FantInI, m.c.a., galvão, r.m.o., kIyohara, p. and tabacnIks, M.H., 2003. Plasma cleaning and analysis of archaeological artefacts from Sipán. Journal of Physics D: Applied Physics 36: 842-848. Pre-columbian alloys from the royal tombs of Sipán and from the Museum of Sicán… schorsch, D., 1998. Silver and gold Moche artefacts from Loma Negra. Metropolitan Museum Journal 33: 109-136. scott, D.A., 2000. A review of gilding techniques in ancient South America, in T. Drayman-Weisser (ed.), Gilded Metals: History, Technology and Conservation. London, Archetype Publications in association with he American Institute for Conservation of Historic and Artistic Works, 203-222. 287 ShImada, I. and grIFFIn, J.A., 1994. Precious metal objects of the Middle Sicán. Scientiic American 270(4): 60-67. VEttEr parodI, l., 2006. Gold of ancient Peru. Lima, Roberto Gheller Doig Ed. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 281-287 Technological and material features of the gold work of Mesoamerica Caractéristiques technologiques et matérielles du travail de l’or en Mésoamérique José Luis Ruvalcaba Sil*, Gabriela Peñuelas Guerrero**, Jannen Contreras Vargas**, Edith Ortíz Díaz*** and Eumelia Hernández Vázquez**** Abstract: he metallurgical work in Pre-Columbian America achieved a high degree of development. In the Mesoamerican area, the development of metallurgy started very late, during the post-Classic period (800 AD). Nevertheless, various techniques, such as lost wax casting and false iligree, were rapidly developed and improved. Most of the gold metallurgy was developed in the Oaxaca region, in the South of Mexico, by the Mixtec people. A limited number of gold alloy based artefacts survived after the conquest of Mexico. Few collections are related to archaeological contexts and may be used to obtain reliable information about the technical features and materials used in the Mesoamerican gold metallurgy. For this reason, non-destructive analyses have been performed on some artefacts of the most important collections of the Mexican museums using portable XRF and PIXE in the laboratory. In this work, a full review of the main technical features and alloys of the analyzed gold artefacts from the Mesoamerican area collections are presented. his database may be used to study collections of gold artefacts from the Mesoamerican areas. Résumé : Le travail du métal a atteint un niveau élevé en Amérique précolombienne. Dans l’aire Mésoaméricaine, ce développement débute très tard, pendant la période postclassique (800 AD). Néanmoins, diférentes techniques, notamment la fonte à la cire perdue et le faux iligrané, rapidement se développent et se perfectionnent. La majeure partie de la métallurgie de l’or a été développée par le peuple Mixtèque dans la région d’Oaxaca, au sud du Mexique. Seul un nombre limité d’objets fabriqués avec des alliages d’or ont survécu à la conquête du Mexique. Quelques rares collections possèdent des objets dont les contextes archéologiques sont connus et peuvent ainsi être utilisées pour obtenir des informations iables sur les procédés techniques et les matériaux utilisés en Mésoamérique dans le cas de la métallurgie de l’or. Pour cette raison, une sélection d’objets appartenant aux plus importantes collections des musées mexicains a été analysée d’une façon non-destructive au moyen d’une FX portable et par PIXE au laboratoire. Ce travail présente un bilan complet des procédés technologiques et des alliages des objets mésoaméricains en or étudiés. La base de données obtenue peut être appliquée à l’étude de collections d’objets en or fabriqués dans l’aire mésoaméricaine. Keywords: Mesoamerica, gold, depletion gilding, PIXE, XRF. Mots-clés : Mésoamérique, or, mise en couleur, PIXE, XRF. * Instituto de Fisica, Universidad Nacional Autónoma de México – Apdo. Postal 20-364 Mexico DF 01000, Mexico. (sil@isica.unam.mx) ** Escuela Nacional de Conservación, Restauración y Museografía-INAH – General Anaya 187 Col. San Diego Churubusco, Coyoacán, cp, 04120. México D.F. (gabrielapenuelas@gmail.com), (jannencontreras@gmail.com) *** Instituto de Investigaciones Antropológicas, Universidad Nacional Autónoma de México – Circuito Exterior s/n, Ciudad Universitaria, Mexico DF 04510, Mexico. (edithd@servidor.unam.mx) **** Instituto de Investigaciones Estéticas, Universidad Nacional Autónoma de México – Circuito Mario de la Cueva s/n, Ciudad Universitaria, Mexico DF 04510, Mexico. (euihernandez@gmail.com) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 289-297 290 1. INTRODUCTION he main cultural area extending approximately from contemporary central Mexico to Honduras and Nicaragua is called Mesoamerica (Fig. 1). In this area, several civilizations and cultures developed from 1500 BC until the Spanish conquest in the 16th century AD. Despite the fact that Mesoamerican cultures achieved a signiicant degree of development in agriculture, mathematics, astronomy, calendaric and technological knowledge, metallurgy appeared in speciic areas quite late, around 800 AD. It is generally accepted that the irst metallic artefacts and metalwork knowledge arrived to Western and Southern Mexico and the Yucatan peninsula from Central America, Colombia and Peru by several paths (Fig. 1). First, the artefacts arrived as a result of exchange. Some of the basic forms were copied and afterwards modiied following local and cultural inlu- José Luis RUVALCABA SIL et al. ences (Hosler, 1995). Later, the metalwork focus in Western Mexico and in Oaxaca – South of Mexico – developed rapidly, with local and original contributions that gave rise to a high level of technical manufacturing of artefacts. Copper metallurgy was developed especially in Western and Central Mexico, while gold and silver were mainly worked in the Oaxaca area, in the South of Mexico. Apparently, the mineral and metal sources available in these regions determined these speciic developments. Gold was obtained mainly from rivers, as powder and nuggets. Mining and gold extraction were not developed until the conquest and Spanish colonization. By the end of the Mesoamerican Age, copper and gold metallurgy became widespread in Western and Central Mexico, and the metallic artefacts were traded and used by the elites of the Mesoamerican peoples both as symbols of power and for ritual purposes in oferings and burials. Only Figure 1: Map of the areas of Mesoamerican and South American gold work traditions. he possible routes of contact are shown. Figure 1 : Carte des aires de tradition orfèvre mésoaméricaine et sud-américaine. Les routes probables de contact sont visibles. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 289-297 291 Technological and material features of the gold work of Mesoamerica high level rulers and warriors were allowed to wear metallic artefacts. Gold and silver were considered symbols of the sun and moon. Nevertheless, precious metals did not have monetary value. In fact, other materials such as green stone objects or fully coloured feathers had a high value and were much appreciated by the Mesoamerican cultures. Most of the usual artefacts were ornaments: chin ornaments, ear ornaments, rings, pendants, necklaces and foils. 2. GOLD COLLECTIONS FROM MESOAMERICAN AREAS Few gold based artefacts from Mesoamerica survived the European destruction and the ambition for gold and silver. In Europe and North America, there are few items that are representative of the goldsmiths’ technological skills and developments (Matos Moctezuma and Solis, 2003). he Mexican collections are composed of items recovered mainly from archaeological excavations of burials and oferings. hey may be used to infer reliable information about the technical features and materials used in the Mesoamerican gold metallurgy. he Mesoamerican collections of Mexico correspond to the Maya area, West Mexico, Aztec gold artefacts, and the sets from the Oaxaca region (Solis, 2004). he Maya he gold artefacts discovered in the Maya area originate mainly from the Yucatan region. In fact, most of the artefacts were recovered from the Chichen Itza Cenote during the irst decade of the 20th century. hese items were part of oferings and ritual ceremonies. heir chronology corresponds to the post-Classic period (800-900 AD), but, judging by the style, shape and size of the artefacts, it is clear that these zoomorphic pieces, bells and pendants were traded from the Central American peoples Diquis and Veraguas (Fernández and Segura, 2004). Only a few foils in the shape of serpents may have been manufactured in the Maya area. his represents the biggest collection of gold artefacts discovered in Mesoamerica, but we have to take into consideration the fact that it is not representative of the local metal work technology. Most of the artefacts are currently on display in the Peabody Museum. Part of the collection was given back to Mexico, and they are exhibited in the Maya room of the National Museum of Anthropology and History of Mexico (MNAH), and in the Yucatan Museum of Palacio Canton. Artefacts from West Mexico One of the major loci of metallurgy in Mesoamerica was the Western region, ruled by the Tarascan, rivals of the Aztecs. hey used extensively copper based artefacts for ornaments and weapons. Nevertheless, a limited number of gold artefacts have also been discovered in this region. Most the artefacts are bells, plates, zoomorphic items and ear ornaments with clear stylistic inluences from South America. Aztec gold artefacts he artefacts belonging to the Aztec culture were discovered during excavations in Mexico City and among the oferings found in the remains of the archaeological site of the main temple of Tenochtitlan, the capital city of the Aztecs, in the historic centre of Mexico City (Matos Moctezuma and Solís, 2003). he sets of artefacts are part of the collections of MNAH and the Museum of the Great Temple of Tenochtitlan (MTM). hey are small plates, pendants and bells decorated with false iligree, and correspond to the period of 1325-1521, the dates of the foundation and conquest of Tenochtitlan, respectively. he gold raw materials, such as powder and ingots, or even foils, were brought from the Oaxaca and Guerrero regions, as tributes from the peoples under Aztec rule. he historical registers indicate the amounts of gold provided by the diferent regions of the Aztec Empire. Manufactured pieces may also be originating from tributes, but pendants and other items were especially ofered as presents for rulers, priests or elite warriors. he historical sources indicate that goldsmiths from the Oaxaca area worked in the Aztec workshops to produce artefacts for the Aztec rulers and ceremonies. he Florentine Codex, written under the supervision of the Franciscan priest Sahagun in the second part of the 16th century, describes in the ninth book the manufacturing process and technology employed by the goldsmiths to produce the gold artefacts by lost wax casting and depletion gilding (Sahagun, 2000). he most outstanding artefacts were discovered among the MTM oferings: a few pendants with spherical beads with a clay nucleus; however, most of the items are gathered in practically three oferings only (#3, #34 and #126). Unfortunately, most of the smaller items were melted, since the items of the ofering were often ired, and they appeared as small solid beads in the archaeological context. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 289-287 292 Mixtec artefacts he Mixtecs were the most skilled goldsmiths of Mesoamerica. heir civilization developed in the Oaxaca regions during the post-Classic period (from 900 AD until the conquest) and replaced the Zapotecs, the dominant culture during the Classic period in Oaxaca, in various sites of the Central Valleys. About 80% of the existing Mesoamerican gold artefacts belong to the Mixtec tradition. he most important collections are in the Oaxaca Room of the MNAH, in the Museo de las Culturas de Oaxaca (MCO), and in the Fisherman’s Treasure in the Baluarte de Santiago in Veracruz (BSV). he collection of MNAH is composed of artefacts from various regions of the Oaxaca area: the Central Valleys and the Sierra. he collection includes pendants, chin ornaments (bezotes), zoomorphic pieces, necklaces, rings and ear ornaments, bells, foils, etc. Some of the artefacts were acquired at the beginning of the last century and they lack an archaeological context. Nevertheless, the main gold artefacts from the tombs of Zaachila and other artefacts discovered in archaeological contexts are also on display. Among the Zaachila pieces, a huge solar pendant made by the lost wax technique, fully decorated ring ornaments and chin ornaments, as well as ear pieces, are exhibited. he Papantla pendant, discovered in the Gulf of Mexico area but belonging to the Oaxaca tradition, is an exceptional piece, similar in terms of technological manufacturing to the inds of Tomb 7 of Monte Alban. he most outstanding gold treasure of pre-Hispanic Mexico was discovered in 1934 at Monte Alban, in a tomb reused by the Mixtec people (Caso, 1969). In this multiple burial, artefacts made of green stones, turquoises, shells, bones, gold and silver were found. About 121 gold artefacts and 24 silver pieces were found. All types of artefacts were discovered, such as sets of remarkable pendants, fully decorated rings, bells, rich necklaces, strings of beads, ear pieces, chin ornaments, etc. he collection includes sets of almost identical pendants and inger ornaments, with practically the same decoration in gold and silver alloys (Fig. 2, a-c and Fig. 3). he third collection is located in the port of Veracruz, on the Gulf coast of Mexico. It was recovered from the sea by a isherman who sold part of the items to a jeweller. Fortunately, an important number of artefacts were not melted. he treasure is attributed to the shipwreck of the boat on which the entrusted Figueroa was returning to Spain after looting tombs in the Sierra of Oaxaca. he collection of 40 pieces includes beautiful pendants, seven bracelets, fully decorated beads, zoomorphic pieces, discs, as well as two ingots. According to their style and shape, the artefacts undoubtedly share features with the Oaxaca artefacts of the MNAH, and with the MCO pieces. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 289-297 José Luis RUVALCABA SIL et al. Following the main discoveries described above, few gold artefacts have been found in the Oaxaca area in recent times. One such case is that of the San Francisco Caxonos site in the Sierra region, where a pendant was recovered (Fig. 2d). his artefact was fully studied using various non-destructive techniques, since it is remarkable in terms of the technology used for its gilding (Ortiz Diaz and Ruvalcaba, 2009). 3. MAIN METALLURGICAL TECHNIQUES AND TECHNOLOGICAL FEATURES he amount of gold artefacts from Mesoamerica is limited in comparison with South American metallic objects. Nevertheless, it is possible to infer some patterns related to the materials used, as well as technological features. First, all the types of gold artefacts from Mesoamerica were manufactured in combination with other precious materials, such as green stones, turquoise, obsidian, crystal rock and sea shells. In very few cases, two examples actually, symmetric bimetallic artefacts, with one half made from a silver-rich alloy and the second half made using a gold-rich alloy, were cast. he irst example, a pendant from Teotitlan, Oaxaca, is currently on display in the Oaxaca room of the MNAH collection, and the second case is that of the set of two discs from Tomb 7 of the Monte Alban collection in the MCO. he main diiculties in the study and characterization of the artefacts are, irst of all, the limited number of items, and, second, the fact that, since they belong to museum collections, it is generally not possible to move them to the laboratory for technological analyses. Moreover, sampling is not allowed or is limited. hus, the analytical approach that we have considered in our research is that of performing non-destructive analyses in situ by imaging techniques, in order to detect restorations and mounting materials, followed by microscopic studies with a 40-80X stereoscope, and inally by elemental analysis by XRF (Peñuelas Guerrero, 2008). he main techniques used to fabricate the Mesoamerican artefacts are hammering for the foils and plates, and lost wax casting using a nucleus of a mixture of clay and carbon. his manufacturing technique allowed the optimization of the amount of gold alloy required for the casting. In fact, native gold was not as abundant in Mesoamerica as we may expect. Undoubtedly, the false iligree technique was developed and improved in comparison with the South American artefacts. All the inest details employed to simulate the threads were produced by lost wax casting. his technique was also used to produce granulations (false gra- 293 Technological and material features of the gold work of Mesoamerica Figure 2: (See colour plate) Various artefacts from Tomb 7 of Monte Alban: a) Pendant with dates, b) zoomorphic pendants, c) Ball game pendant, and d) Warrior pendant from Caxonos, Sierra of Oaxaca. Figure 2 : (Voir planche couleur) Diférents objets de la Tombe 7 de Monte Alban: a) Pendentif avec dates, b) pendentifs zoomorphes, c) pendentif au jeu de bale et d) pendentif au guerrier de Caxonos, Sierra of Oaxaca. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 289-287 294 José Luis RUVALCABA SIL et al. Figure 3: hree of ive of the Xochipilli pendants set of Tomb 7 of Monte Alban. Details of false iligree modelling are shown. Figure 3 : Trois des cinq pendentifs Xochipilli de la Tombe 7 de Monte Alban. Détails du faux iligrane. nulation) and mesh decorations. After hundreds of analyses using XRF and PIXE (Cesareo et al., 1994; Ruvalcaba et al., 1995; Peñuelas Guerrero, 2008), as well as SEM-EDS measurements carried out on the artefacts (La Niece and Meeks, 2000), no indication of soldering has ever been observed in the Mesoamerican gold artefacts. he collection of Tomb 7 of Monte Alban provides outstanding information about the modelling and lost wax casting of the metal work. Since this collection includes sets of almost identical pendants and inger ornaments, from a careful examination it is clear that they were produced in series in the same workshops, following the same iconography, size and decoration. However, the back parts show small diferences in the wax modelling of the threads, and ArcheoSciences, revue d’archéométrie, 33, 2009, p. 289-297 the unions of the various parts in the original model (Fig. 3). hus, each object is in fact unique, despite their similarity. Also, the composition of the gold alloys used, determined by XRF, indicates that the same alloys were used to cast one set of pieces, like in the case of the small zoomorphic pendants (78% Au, 13% Ag, 9% Cu) of Figure 2b. Nevertheless, the sets may have diferent compositions as well. For example, four of the ive pendants of the Xochipilli goddess in Figure 3 have the same composition (47% Au, 33% Ag, 20% Cu), but for the ifth piece the elemental contents are diferent (54% Au, 26% Ag, 20% Cu). he diference in the composition of these sets may suggest two diferent workshops. he elemental analysis of the huge ball game pendant (Fig. 2a) indicates that three alloys were used to manufacture this Technological and material features of the gold work of Mesoamerica artefact; the irst four sections have a similar composition (59% Au, 26% Ag, 15% Cu), while for the feathers section (5), the mean composition is completely diferent (50% Au, 38% Ag, 12% Cu), and it also changes slightly for the bits and bells sections 6 and 7 (50% Au, 36% Ag, 15% Cu). his behaviour has also been observed for the eagle chin ornament from the Zaachila tombs and for the Caxonos pendant (Fig. 2d). From these results, we may infer that the casting of this kind of pieces was carried out by rows, sometimes using the same alloy, and in other cases diferent alloys, as a function of the amount of available materials. A comparison of the mean composition of the artefacts of Tomb 7 of Monte Alban and other collections indicates that the Mixtec goldsmiths were able to produce a signiicant number of alloys with similar composition, but also that they were using similar native gold sources to manufacture the artefacts. On the other hand, in terms of inishing the items, the historical sources report the use of a depletion gilding technique (mise-en-couleur) by successive oxidation and cleaning of a gold-poor alloy. According to this technique, the amount of gold increases gradually at the surface due to the elimination of copper and the consequent nucleation of gold (Bray, 1993; Ruvalcaba et al., 2004); however, so far this was only identiied in one piece from Mesoamerica. his is the case of the warrior pendant discovered at the San Francisco Caxonos site in Oaxaca (Fig. 2d). When this kind 295 of depletion gilding is used, we may expect that the surface composition presents both lateral and depth heterogeneity because of the irregular loss of copper by pickling and the dynamics of gold nucleation during heating. X-ray based techniques, such as EDS, PIXE and XRF, provide the mean elemental composition, and only by changing the incident beam energy and performing the analysis on several regions of the surface it is possible to detect this heterogeneity and the use of a depletion gilding technique (Demortier and Ruvalcaba, 1998). For the Caxonos pendant, diferential PIXE combined with RBS was applied to determine the gold enrichment depth proile, ranging from 50% gold content in the bulk up to practically 100% gold concentration at the upper surface, on the irst 4 µm depth. Copper concentration may reach more than 35% in the bulk (Demortier and Ruvalcaba, 2005; Ortiz Diaz and Ruvalcaba, 2009). According to the data reported in analytical studies, it is possible to show in Figure 4 the composition of the gold alloys of the Mesoamerican items by regions (Franco Velazquez and Torres Montes, 1989; Cesareo et al., 1994; Ruvalcaba et al., 1995; Ortiz Diaz and Ruvalcaba, 2009; Peñuelas Guerrero, 2008). In this igure, the data for the Maya artefacts imported from Central American regions show high amounts of gold (higher than 90%) and do not it with the composition of other Mesoamerican regions. On the other hand, the data for the Aztec collection overlaps Figure 4: Comparison of the elemental composition of gold artefacts from the main Mesoamerican areas. Figure 4 : Comparaison de la composition élémentaire d’objets d’or des principales aires mésoaméricaines. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 289-287 296 with that pertaining to the artefacts from the Central Valleys of Oaxaca, made by the Mixtec goldsmiths. he compositions of the gold alloys range from 60 up to 85% gold, and may reach about 20% copper; the silver contents rises from 10 to 40%. Nevertheless, the artefacts from the Sierra of Oaxaca (Caxonos and Fisherman’s Treasure) do not it very well with the composition of the Central Valley and Aztec pieces. heir gold alloys are richer in copper (from 10 to 40%) and silver amounts vary from 10 to 35%. here is a diference of about 60° C in the melting temperatures of the alloys. Also, depletion gilding may have been used, since the amount of copper is signiicantly higher. For these reasons, the hypothesis of two loci of metal work, one including the Oaxaca area and the Central Valleys, and the other the Sierra, has been proposed. hese data may be used as a reference for Mesoamerican style artefacts without an archaeological context or reported provenance. For instance, the composition of the gold artefacts from the British Museum its perfectly our reported data for the Central Valley artefacts. Finally, a comparison of the reported data for Mesoamerican pieces with the gold alloys measurements carried out for the Panama traditions (Veraguas and Cocle) and some Colombian cultures (Calima, Tolima and Muisca) indicates an overlap between the Veraguas items and the Maya artefacts from the Cenote of Chichen Itza, as it may be expected, while the Colombian pieces show some similarity with the pieces from Oaxaca areas. 4. FINAL REMARKS Despite the fact that metallurgy developed quite late in Mesoamerica – after 800 AD – a quick development and high technological level were achieved in the manufacturing of gold artefacts after this date. A limited number of objects (less than 500 pieces) have been preserved. From the analyses of these items, we can infer reliable information about the work and skills of Mesoamerican goldsmiths; in particular, they were masters of the false iligree and false granulation techniques. here are very few cases of artefacts inished by depletion gilding. his fact may be due to the low amount of copper in the gold alloys. Other gilding techniques, such as electrochemical replacement plating, have been observed in imported artefacts recovered in the Maya region of Yucatan (Contreras et al., 2007). So far, there is no technical evidence of soldering techniques. Since the collections are unique, it is diicult to access the items. hus, in situ analysis represents a convenient irst step ArcheoSciences, revue d’archéométrie, 33, 2009, p. 289-297 José Luis RUVALCABA SIL et al. that might justify further analyses in the laboratory, and/ or establish a suitable strategy for sampling with minimal damage to the artefact. Acknowledgements his research is part of the MOVIL project Non destructive methodologies for the study of the Mexican cultural heritage, with the inancial support of the CONACyT Mexico U49839-R. References BRAY, W., 1993. Techniques of gilding and surface enrichment in preHispanic American metallurgy, in S. La Niece, P. Craddock (eds.), Metal Plating and Patination: Cultural, Technical and Historical Developments. Oxford: Butterworth Heinemann, 182-192. CASO, A., 1969. El Tesoro de Monte Albán. Memorias del Instituto Nacional de Antropologia e Historia III. México: Instituto Nacional de Antropologia e Historia. CESAREO, R., GIGANTE, G.E., IWANCZYK, J.S., ROSALES, M.A., ALIPHAT, M. and ÁVILA, P., 1994. Non-destructive analysis of pre-Hispanic gold objects using energy dispersive X-ray luorescence. Revista Mexicana de Física 40(2): 301-308. CONTRERAS, J., RUVALCABA SIL, J.L. and ARENAS ALATORRE, J., 2007. Non destructive study of gilded copper artifacts from the Chichen-Itza Cenote, in Proceedings of the XI International Conference on Particle Induced X-rays Emission and its Analytical Applications, Puebla, Mexico, May 25-29, 2007. México, Universidad Nacional Autonóma de México (CD-ROM). DEMORTIER, G. and RUVALCABA SIL, J.L., 1998. Diferential PIXE Analyses of Mesoamerican Jewelry Items. Nuclear Instruments and Methods in Physics Research B 118: 352-355. DEMORTIER, G. and RUVALCABA SIL, J.L., 2005. Non-destructive ion beam techniques for the depth proiling of elements in Amerindian gold jewellery artefacts, in R. Van Grieken, K. Janssens (eds.), Cultural Heritage Conservation and Environmental Impact Assessment by Non-destructive Testing and Microanalysis. London, A.A. Balkema Publishers, 91-100. FERNÁNDEZ, P. and SEGURA, J., 2004. Metalurgia del sureste de Costa Rica: Identiicación de Producciones Locales basadas en evidencia Tecnológica y estilistica, in A. Perea, I. Montero, O. García-Vuelta (eds.), Primer Simposium Internacional sobre Tecnología del Oro Antiguo SITOA. Anejos de Archivo Español de Arqueología XXXII. Madrid, CSIC, 49-61. FRANCO VELAZQUEZ, F. and TORRES MONTES, L., 1989. La orfebrería prehispánica en el Golfo de México y el Tesoro del Pescador, in C. Aguilar (ed.), Orfebrería prehispánica. México, Corporación Industrial San Luis, 217-271. Technological and material features of the gold work of Mesoamerica HOSLER, D., 1995. Sound, color and meaning in the metallurgy of ancient west Mexico. World Archaeology 27(1): 100-115. LA NIECE, S. and MEEKS, N., 2000. Diversity of Goldsmithing Traditions in the Americas and the Old World, in C. McEwan (ed.), Pre-Columbian Gold, Technology, Style and Iconography. London, Fitzroy Dearborn Publishers, 220-239. MATOS MOCTEZUMA, E. and SOLIS, L.F., 2003. Aztecs. London, Royal Academy Books. ORTIZ-DIAZ, E. and RUVALCABA SIL, J.L., 2009. An historical approach to a gold pendant: he study of diferent metallurgic techniques in ancient Oaxaca, Mexico, during the late postclassic period, in Archaeometallurgy in Europe: 2nd International Conference, Aquileia, Italy, 17-21 June 2007: selected papers. Milano, Associazione Italiana di Metallurgia, 511-518. PEÑUELAS GUERRERO, G., 2008. Caracterización por Medio de Análisis Instrumentales de los Materiales Constitutivos de la Orfebreria de la Tumba 7 de Monte Albán, Oaxaca. México, 297 Esc. Nal. de Restauración. Conservación. y Museografía, Instituto Nacional de Antropologia e Historia. RUVALCABA SIL, J.L., DEMORTIER, G. and OLIVER, A., 1995. External beam PIXE analysis of gold pre-Hispanic Mexican jewelry. International Journal of PIXE 5: 273-288. RUVALCABA SIL, J.L., TORRES, L., FRANCO, F. and ORTIZ DÍAZ, E., 2004. Artifacts rich surfaces: Depletion gilding or natural surface corrosion? Study of corrosion and oxidation of gold alloys, in A. Perea, I. Montero, O. García-Vuelta (eds.), Primer Simposium Internacional sobre Tecnología del Oro Antiguo SITOA. Anejos de Archivo Español de Arqueología XXXII. Madrid, CSIC, 41-47. SAHAGUN, B., 2000. Historia General de las Cosas de la Nueva España. México, Conaculta, Colección Cien de Mexico. SOLÍS, L.F. and CARMONA, M., 2004. Oro Precolombino de Mexico Colecciones Mixteca y Azteca. IXE -Landucci-Mogilner eds. Hong Kong. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 289-287 Technological study of pre-Columbian bimetallic discs from Monte Alban, Oaxaca, Mexico Étude technologique des disques bimétalliques précolombiens de Monte Alban, Oaxaca, Mexique Gabriela Peñuelas Guerrero*, Jannen Contreras Vargas*, José Luis Ruvalcaba Sil** and Ángel García Abajo* Abstract: One of the most remarkable examples of metalwork from Mesoamerica is the collection from Monte Alban’s Tomb 7. Among its items, we encounter extraordinary examples of bimetallic artefacts; a pair of solar representations whose main particularity is that they are formed by two diferent alloy sheets: one rich in silver and the other rich in gold. he silver-rich zones show deterioration characterized by brittleness, fractures and the loss of some fragments; this fragility is apparently due to dislocations caused by repeated hammering and insuficient annealing during the manufacturing process. Because sampling of the original items was not allowed, test specimens were prepared replicating the manufacturing techniques and alloy compositions, based on data previously obtained by X-ray Fluorescence Analysis of the original pieces. Examination of these test specimens was undertaken in order to understand the thermodynamic behaviour of the metallic microstructure and its relation to the alteration processes. Samples of each alloy were then taken and analyzed by metallographic methods. Résumé: Un des plus remarquables exemples du travail du métal en Mésoamérique est l’ensemble trouvé dans la Tombe 7 de Monte Alba. Parmi ces objets, certains sont des exemples extraordinaires de productions bimétalliques, comme une paire de représentations solaires dont la principale particularité est l’utilisation de deux feuilles d’alliages diférents : un riche en argent et l’autre riche en or. Les régions riches en argent montrent une dégradation caractérisée par des cassures, fractures et perte de fragments; cette fragilité est apparemment due aux dislocations causées pendant leur fabrication par des coups de marteau répétés avec une chaufe insuisante. Comme il est impossible de prélever des objets originaux, des analogues ont été préparés en utilisant les mêmes techniques de fabrication et des alliages identiques, révélés par les analyses réalisées précédemment par FX sur les pièces originales. L’examen des analogues a été réalisé de façon à comprendre le comportement thermodynamique de la microstructure métallique et son rapport aux procédés d’altération. Des échantillons de chacun des alliages ont été prélevés pour efectuer une analyse métallographique. Keywords: Bimetallic artefacts, gold alloys, silver alloys, metallography, X-ray luorescence. Mots-clés : objets bimétalliques, alliages d’or, alliages d’argent, métallographie, FX. 1. INTRODUCTION In October 2007, part of the jewellery collection which constitutes the ofering of Tomb 7 of Monte Alban, Oaxaca, Mexico, was analyzed by means of portable X-ray Fluorescence (XRF) with the purpose of characterizing the alloys and manufacturing techniques, and of obtaining relevant information on the alteration phenomena, * Escuela Nacional de Conservación, Restauración y Museografía-INAH – General Anaya 187 Col. San Diego Churubusco, Coyoacán, México D.F. CP. 04120., Mexico. (gabrielapenuelas@gmail.com), (jannencontreras@gmail.com) ** Instituto de Física, Universidad Nacional Autónoma de México – Apdo. Postal 20-364, México D.F 01000, Mexico. (sil@isica.unam.mx) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 299-302 300 in order to prepare a conservation proposal (Peñuelas Guerrero, 2008). his ofering represents one of the most representative, varied and rich metalworking traditions of the Mixtec area, and even of the entire Mesoamerican area (Caso, 1969). Even though some analyses of a few gold objects were carried out before (Camacho Bragado, 2005), this was the irst one using portable XRF to obtain elemental composition data, enabling one third of the entire metalwork collection to be studied. Moreover, this analytical approach was the irst one employed for characterizing the silver alloys of the artefacts. Among the jewellery items, there are two small and unique artefacts with solar representations in relief, made with two diferent alloys: one rich in silver and the other rich in gold. hese pieces are known as bimetallic discs (Fig. 1). In Mexico, besides these pieces, only one other bimetallic pendant originating from Teotitlan del Camino, Oaxaca (Fig. 2) is known. It is conserved in the Museo Nacional de Antropologia e Historia – INAH (Carmona Macias, 1997). he main diference between the Monte Alban discs and the Teotitlan pendant is that the pendant was clearly made by lost wax casting while, because of their thinness, the discs were probably made by hammering. he production of bimetallic items with gold in preColumbian America is not as uncommon as we may think, but platinum was used instead of silver. Fray Bernardino de Sahagun, one of the most representative chroniclers of New Spain, describes the extraordinary objects that Aztecs were able to make by combining gold and silver (Sahagun, 2000). he irst archaeologist to analyze the manufacture technique of the bimetallic discs was Easby (Caso, 1969), who proposed that the discs were made by hammering two different sheets together: one of an alloy rich in silver, and the other of an alloy rich in gold, achieving a difusion boun- Gabriela PEÑUELAS GUERRERO et al. Figure 2: (See colour plate) he Teotitlan del Camino, Oaxaca, bimetallic pendant. Figure 2 : (Voir planche couleur) Le pendentif bimétallique Teotitlan del Camino, Oaxaca. dary of the foils. Besides the unusual manufacture technique, the bimetallic discs present a speciic conservation problem: the silver-rich zones are brittle, show fractures and have lost some fragments; this fragility is apparently due to dislocations caused by repeated hammering with insuicient annealing. he actual study was carried out to understand their deterioration phenomena as well. 2. METHODOLOGY Figure 1: (See colour plate) he bimetallic discs from Tomb 7 of Monte Alban, Oaxaca, Mexico. Figure 1 : (Voir planche couleur) Les disques bimétalliques de la Tombe 7 de Monte Alban, Oaxaca, Mexique. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 299-302 Previous XRF measurements of the two discs provided information on their composition. For the irst disc, the mean composition of the part rich in gold is 85.8% Au, 13.7% Ag and 0.6 % Cu, while the silver-rich alloy section has an average composition of 91.1% Ag, 6.4% Au and 2.5% Cu (Peñuelas Guerrero, 2008). For the second disc, the average compositions for the gold and silver sections are 81% Au, 18.2% Ag, 0.7% Cu, and 91.3% Ag, 5.5% Au and 3.2% Cu, respectively. Based on Easby’s hypothesis and knowing the elemental composition of each alloy from XRF measurements, replica alloys were designed to test his hypothesis. he irst step involved the reproduction of the alloys with a mean composition about 85% Au, 14% Ag and 1% Cu, Technological study of pre-Columbian bimetallic discs from Monte Alban, Oaxaca, Mexico and 91% Ag, 6% Au and 3% Cu, and the casting of each part individually. Subsequently, the homogeneity of the elemental composition of the replicas was veriied using a portable XRF spectrometer with a Mo X-ray tube with a voltage of 45 kV, a current of 0.2 mA, and a CT detector. he exciting X-ray beam had a diameter of 1.5 mm on the sample surface and each measurement lasted 30 sec. Once the foils were reproduced, we proceeded to try to join them by hammering and heating. Although we were unable to join them according to this procedure, after obtaining a similar thickness to that of the original discs, we proceeded to take samples of each alloy for metallographic examination. 3. RESULTS After a long hammering and annealing process, it proved impossible to join the two sheets, due to their diferent behaviour. Without soldering, higher temperatures were required to join them, but the diferent melting points of each alloy (the gold-rich alloy melts at 1010 °C while the silver-rich alloy melts at 954 °C), in combination with the rapid heat difusion caused by the limited thickness of the foil (0.8 mm), caused the silver-rich alloy to begin to fuse instead of joining to the gold one. he fusing of the silver alloy was veriied by means of metallographic analysis; while the structure of the gold alloy showed crystals and macles characteristic of cold work, the 301 silver-rich alloy had a cast-like structure, even though both metals were worked at the same temperature (Fig. 3). 4. CONCLUSION Based on observations of the tests performed on the replica alloys and the study of the other existing bimetallic artefact, the Teotitlan pendant made by lost wax casting, a hypothesis regarding the manufacture of the Monte Alban bimetallic discs is proposed. he Teotitlan pendant was probably cast by irst casting the higher melting point alloy and then completing the mould with the other lower melting point alloy. We believe that the alloys of the bimetallic discs were likely cast in a similar manner and were then joined mechanically, with heat. Afterwards, the goldsmith must have hammered the cast objects to obtain the desired thickness, size and design. he beating and heating should have strengthened the join and produced a slight silver diffusion into the gold-rich alloy. We can conclude that the brittleness of the silver-rich alloy is due to an excess of dislocations resulting from the diferent behaviour of the two alloys that were cold worked (hammered) together. Nevertheless, the next step of this experiment is to reproduce the lost wax casting and hammering in order to obtain further information regarding the deterioration phenomena. Figure 3: (See colour plate) Metallographic plates of the silver and gold alloys made for the bimetallic replica test (200X and 400X). Figure 3 : (Voir planche couleur) Planches métallographiques des alliages d’argent et d’or réalisées pour efectuer les analogues bimétalliques (200X and 400X). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 299-302 302 Acknowledgements his research is part of the MOVIL project Non destructive methodologies for the study of the Mexican cultural heritage, with the inancial support of the CONACyT Mexico U49839-R. References CASO, A., 1969. El Tesoro de Monte Albán. Memorias del Instituto Nacional de Antropologia e Historia III. México, Instituto Nacional de Antropologia e Historia. CARMONA MACIAS, M., 1997. Entre Crisoles y Dioses: La orfebreria prehispánica de Oaxaca, in M. Dalton Palomo, V. Loera y Chávez Castro (eds.), Historia del Arte de Oaxaca, Arte ArcheoSciences, revue d’archéométrie, 33, 2009, p. 299-302 Gabriela PEÑUELAS GUERRERO et al. Prehispánico, vol. 1. Oaxaca, México, Gobierno del Estado de Oaxaca, Instituto Oaxaqueño de las Culturas, 286-307. CAMACHO BRAGADO, A., ORTEGA, M., VELASCO, M. and JOSÉYACAMÁN, M., 2005. A microstructural study of gold treasure from Monte Alban’s Tomb 7. Journal of the Minerals, Metals and Materials Society 57(7): 19-24. PEÑUELAS GUERRERO, G., 2008. Caracterización por Medio de Análisis Instrumentales de los Materiales Constitutivos de la Orfebreria de la Tumba 7 de Monte Albán, Oaxaca. México: Esc. Nal. de Restauración. Conservación. y Museografía, Instituto Nacional de Antropologia e Historia. SAHAGUN, B., 2000. Historia General de las Cosas de la Nueva España. México, Conaculta, Colección Cien de Mexico. Location of gold placers in Oaxaca during the late pre-Hispanic period and early Colonial times: 1250 to 1550 AD Localisation de placers aurifères à Oaxaca pendant la période préhispanique tardive et la période coloniale récente : 1250 à 1550 apr. J.-C. Edith Ortiz Díaz * Abstract: During the late post-Classic period (1250-1521 AD), the Mesoamerican region of Oaxaca was well known for the pre-Hispanic gold metallurgy developed by the Mixtecs and Zapotecs. In this epoch, the most important burials were adorned with exceptional jewels, such as the well know items of Tomb 7 from the archaeological site of Monte Alban, and from other sites, like Zaachila. In addition to the Central Valley of Oaxaca and the Mixtec area, there are other zones in Oaxaca where the ancient goldsmiths manufactured exquisite gold objects, like the Northern Sierra of Oaxaca, or the Isthmus of Tehuantepec. Nevertheless, there is a huge lack of information regarding the location of the principal gold placers of Oaxaca. he objective of this paper is to situate the most important gold placers of Oaxaca during the late post-classic period and the irst decades of the Colonial times. Résumé : La région mésoaméricaine d’Oaxaca est reconnue par la métallurgie préhispanique de l’or, développée par les Mixtèques et les Zapotèques pendant la période postclassique récente (1250-1521 AD). À cette époque, les sépultures les plus importantes étaient remplies de pièces d’orfèvrerie exceptionnelles telles les fameux objets de la Tombe 7 du site archéologique de monte Alban ou encore des objets d’autres sites comme Zaachila. En plus de la vallée centrale d’Oaxaca et de la zone d’inluence des Mixtèques, d’autres régions à Oaxaca, comme le nord de la Sierra d’Oaxaca ou l’isthme de Tehuantepec, ont dévoilé des objets en or exceptionnels, manufacturés par les anciens orfèvres. Néanmoins, les informations sur la localisation des principaux placers d’Oaxaca sont quasi inexistantes. Pour répondre à cette question, ce travail se propose de situer les principaux placers d’Oaxaca exploités pendant la période postclassique tardive et les premières décades de la période coloniale. Keywords: colonial, locate, Mexico, Oaxaca, gold placers, pre-Columbian. Mots-clés : colonial, localisation, Mexique, Oaxaca, placers, précolombien. 1. INTRODUCTION Techniques of gold and silver smithing, introduced in Mesoamerica from South America since the 8th century AD, were admirably assimilated and even improved by the Mixtec and Zapotec artisans of Oaxaca. However, the Mixtec and Zapotec goldsmiths of the Mixtec area and the Central Valley of Oaxaca were not the only artisans in this region. Documentary evidence indicates that the Chinantecs (Díaz del Castillo, 1979) and the highland Zapotecs of the *Instituto de Investigaciones Antropológicas, Universidad Nacional Autónoma de México, Circuito Exterior s/n – Ciudad Universitaria, 04510, México D.F., Mexique. (edithd@servidor.unam.mx) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 303-307 Edith ORTIZ DÍAZ 304 Northern Sierra of Oaxaca were creating gold jewels (Ortiz and Ruvalcaba, 2007) by the end of the pre-Hispanic period (1250-1524 AD), and also that the Zapotecs of the Isthmus of Tehuantepec were ine artisans (Saville, 1920). Based on this information, we can establish at least three centres of gold manufacture in Oaxaca before the Spanish conquest. In consequence, in no other part of Mexico have so many objects of gold been found as in Oaxaca. Nevertheless, to date we do not have any information about the gold production process, or about who controlled the access to the gold placers. Even if these questions are critical for understanding the complexity of past societies, archaeological and historical research have left aside such inquires until now, as well as the issue of what happened with the ancient gold placers after the Conquest. It is clear that we cannot answer all these questions in the present work. As a result, in this paper we are going to focus our attention on locating the ancient gold placers of the Mixtec area, Central Valley of Oaxaca, and the Northern Sierra (Highlands and Lowlands) (Fig. 1). 2. METHODS he alluvial placers of Oaxaca have been exploited since approximately the 10th century AD. Even if this activity became very important for Mixtecs and Zapotecs, we do not have too much archaeological evidence regarding it. he main reason for this is that most of the ancient artefacts that were employed to extract gold from the rivers have not been found until today in the archaeological record. his is partly because some of these ancient tools used to extract gold from the rivers were crafted from organic materials, like wood vessels. We also have to consider the possibility that the Spanish exploitation of the ancient native gold placers could have altered the pre-Columbian evidence pertaining to gold working. Yet another inconvenience in terms of situating the placers is that the rivers could have moved and lood stages have changed since ancient times. hus, in order to achieve our objective, we will examine early historical records from Oaxaca and central Mexico. In addition, we will incorporate in this study a brief review of the geological information pertaining to each area under consideration, in order to determine the gold potential that actually exists in selected zones. 3. RESULTS Gold placers in the Mixtec area he valley of Nochixtlan-Yanhuitlan is part of the Mixtec area. According to the Codex Yanhuitlan, which was painted after the Spanish conquest, we can clearly identify two place signs, or toponyms, on the eleventh plateau (Fig. 2). Some scholars have proposed a location for each place sign. First, on the left side of the bottom of the plateau there is a place sign that can be translated as tlachtepec or tepetlachco, which means “ballgame on the hill”. Nineteen kilometres to the south of the modern town of Yanhuitlan, there is a settlement whose name is “Tlachitongo”. Based on the ethno-historical records, it is Figure 1: he studied area. Figure 1 : La région étudiée. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 303-307 Location of gold placers in Oaxaca during the late pre-Hispanic period and early Colonial times: 1250 to 1550 AD 305 Figure 2 : Codex Yanhuitlan, Eleventh Plate indicating the toponymic of the river. Figure 2 : Codex Yanhuitlan, onzième planche indiquant la toponymie du leuve. possible to think that this town could be the old Tlachtepec or Tepetlachco (Jiménez Moreno and Mateos Higuera, 1940). he second place sign which appears on another scene of the codex has been identiied as “house” or “place site” (Smith, 1973). Smith (1973) argues that this toponym is not necessarily attached to a speciic place, but could also indicate a boundary. his ancient gold placer was probably situated on the border of the old polity of Yanhuitlan. After the Spanish conquest, Europeans exploited the gold placers that the Mixtec Indians had in use in Yanhuitlan, but placers were abandoned by the second half of the 16th century. Perhaps the reason was that those gold placers were not as rich as they might seem. In a recent survey, the Mexican Geological Service (SGM) did not report this Mixtec area as part of an active gold mineral region (Servicio Geológico Mexicano, 2005). Gold placers in the Central Valley of Oaxaca he Central Valley of Oaxaca was occupied by Mixtec and Zapotec populations by the end of the 14th century. In 1486 the Aztecs conquered the Central Valley and forced the Mixtecs and Zapotecs to pay tribute to the Aztec Empire (Marcus, 1983). A list of the tributes paid by the towns of the Valley of Oaxaca is shown on Folio 43v. of the Codex Mendoza (Berdan and Anawalt, 1992). Saville (1920) indicated that one of the most signiicant things of this Folio is that the place named in nahuatl Teocutlan or Teocuitlan can be translated as “the place where gold is found” (Fig. 3). Even if we were not able to locate this town with certainty in present day Mexico, the settlement could have been located in the south of the Valley of Oaxaca. Figure 3. (See colour plate) Codex Mendoza, Folio 43v. he town of Teocuitlai. Figure 3 : (Voir planche couleur) Codex Mendoza, Folio 43v. La ville de Teocuitlai. As it happened in the Mixtec area, as soon as the Spanish conquered this Valley, they began to wash the gold rivers, trying to get as much gold as they could, but they found that this activity was far from being productive because gold was rare (Chance, 1982). However, the Mexican Geological Service actually indicates that the Central Valley is one of the two principal mineral regions of the State. According to the SGM, the main mineral district of this region yields between 2.2 and 13 g/t of Au and between 82 and 146 g/t of Ag (Servicio Geológico Mexicano, 2005). Gold placers in the Northern Sierra In order to study the Northern Sierra gold placers, we are going to divide the mountain range in two diferent areas: the highlands, including the mountains that are higher than 1000 m.a.s.l., and the lowlands, including the mountains that are below 1000 m.a.s.l. and the coastal plain. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 303-307 Edith ORTIZ DÍAZ 306 Highlands: he Northern Sierra was inhabited by three diferent ethnic groups: Highland Zapotecs, Highland Chinantecs, and Mixes. In the beginning of the 16th century, Moctezuma II informed Cortes that there was gold in this area, but that this region was not subjected to the Aztec Empire. Some of the main watercourses of Mexico have their origin in the Northern Sierra, and the rivers were well known in the 16th century for representing sources of precious metal. Based on early historical records, we know that the Zapotecs and their neighbours, the Chinantecs, used to pay gold as tribute to the Spanish Crown after the conquest. he native populations obtained the gold from various rivers. Nevertheless, colonial authorities complained about the amount of gold they obtained from the rivers and about the quality of gold artefacts that they received from the native populations because they were very thin (Chance, 1998). Between 1520 and 1530, the conquerors tried to establish gold and silver mines in the Mixe and Highland Zapotec areas, but they did not prosper at all (Chance, 1998). Lowlands: As mentioned above, the Aztecs conquered the Mixtec area and the Central Valley of Oaxaca, but they also conquered the northern part of the actual state of Oaxaca and the south of Veracruz. According to the Codex Mendoza, the towns in this area, named the province of Tuxtepec, used to pay their tribute to the Aztecs in ine gold objects, among other items. Díaz del Castillo (1979) emphasized the skill of the Chinantecs in working metal. he author does not provide a speciic name for the rivers in this area, but we can identify two possibilities: the Tuxtepec and the Malinaltepec rivers. Both streams were part of the Papaloapan causeways that have their origin in the Highlands of the Northern Sierra. After 1524 AD, the Spanish tried to exploit the gold placers, but by the end of the irst half of the 16th century the Tuxtepec area was practically uninhabited by native populations, because of epidemic diseases. his lack of a native workforce radically afected the exploitation of gold in the lowlands (Acuña, 1984). According to the Mexican Geological Service, four diferent mineral districts currently exist in the Northern Sierra (highlands and lowlands). he values for precious metals are in the range of 0.5 to 6.4 g/t of Au, and of 200 to 280 g/t of Ag. 4. DISCUSSION AND CONCLUSION As we have seen in this paper, the native populations of Oaxaca obtained gold from placer mining. hose placers were formed in watercourses, creeks, rivers, lood plains and ArcheoSciences, revue d’archéométrie, 33, 2009, p. 303-307 deltas. Even if in some areas it is not possible to determine where these gold placers were exactly located, we begin to have a clearer idea of their position. According to the actual percentage of gold reported by the Mexican Geological Service, the Central Valley of Oaxaca area is the richest in gold. he Northern Sierra occupies the second place, and the Mixtec area is in the third place. In pre-Columbian times, the Mixtec region paid a tribute to the Aztecs of twenty vessels of gold dust. Mixtecs and Zapotecs from the Valley of Oaxaca also sent to Moctezuma twenty large gold plaques. he plaques had the size of an average plate and the thickness of a thumb (Saville, 1920). According to the Codex Mendoza, the towns that were settled in the lowlands of the Northern Sierra or the province of Tuxtepec used to pay tribute to the Aztecs in inished objects, even if the Central Valley of Oaxaca was the richest in gold resources. Nevertheless, during the last century, archaeologists have recovered more gold objects from the Central Valley of Oaxaca. he cause of this could be the lack of archaeological exploration in the Northern Sierra. In contrast, archaeologists have focused their attention to the Central Valley of Oaxaca and the Mixtec area. On the other hand, the Spanish conquest made it evident that the gold sources of Oaxaca were not as plentiful as they had thought. he gold and silver was not suicient to support an intensive exploitation, and it was only adequate for the creation of a small number of luxury gold objects for the pre-Columbian societies. Finally, the importance of establishing the location of gold placers in pre-Columbian times is related to the reconstruction of the entire gold work process, from placer mining to the ancient gold workshops. In order to achieve this goal, an interdisciplinary combination of archaeology, history, gold objects characterization, geomorphology, geology, and geochemical information pertaining to each area must be employed. References ACUÑA, R., 1984. Relaciones geográicas del siglo XVI: Antequera. México City: Instituto de Investigaciones Antropológicas, Universidad Nacional Autonóma de México. BERDAN, F.F. and ANAWALT, P.R. (eds.), 1992. Codex Mendoza. Berkeley, University of California Press. CHANCE, J.K., 1982. Razas y clases de Oaxaca colonial. México City, Instituto Nacional Indigenista. CHANCE, J.K., 1998. La conquista de la Sierra. Españoles e indígenas de Oaxaca en la época colonial. México City, Instituto Oaxaqueño de las Culturas, CIESAS. Location of gold placers in Oaxaca during the late pre-Hispanic period and early Colonial times: 1250 to 1550 AD DÍAZ DEL CASTILLO, B., 1979. Historia verdadera de la conquista de la Nueva España. México City, Editorial Promexa. JIMÉNEZ MORENO, W. and MATEOS HIGUERA, S., 1940. Códice de Yanhuitlan. Facsimilar edition with a preliminary study. México City, Secretaria de Educación Pública, Instituto Nacional de Antropología e Historia. MARCUS, J., 1983. Aztec military campaign against the Zapotecs: the documentary evidence, in K. Flannery, J. Marcus (eds.), he cloud people. Divergent evolution of the Zapotec and Mixtec civilizations. New York, Academic Press, 314-318. ORTIZ-DIÁZ, E. and RUVALCABA SIL, J.L., 2007. An historical approach to a gold pendant: he study of diferent metallurgic techniques in ancient Oaxaca, Mexico, during the 307 late post-classic period, in Archaeometallurgy in Europe: 2nd International Conference, Aquileia, Italy, 17-21 June 2007. Milano: Associazione Italiana di Metallurgia, CD-ROM. SAVILLE, M., 1920. he goldsmith’s art in ancient México. New York, Museum of the American Indian and Heye Foundation. SERVICIO GEOLÓGICO MEXICANO, 2005. Carta Geológico Minera E14-9 Oaxaca, Puebla Esc 1: 250,000. Aguascalientes, Ags. México: Instituto Nacional de Estadística, Geografía e Informática (INEGI). SMITH, M.E., 1973. Picture Writing from Ancient Southern Mexico: Mixtec Place Signs and Maps. Norman, University of Oklahoma Press. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 303-307 he irst gold coins struck in Brazil: myth or reality? Les premières monnaies en or frappées au Brésil : mythe ou réalité ? Mathieu Duttine*, Maria Filomena Guerra**, Rejane Maria Lobo Vieira***, Rosa B. Scorzelli****, Carlos Eduardo Pereira***** and Carlos A. Perez****** Abstract: Besieged in Pernambuco by the Portuguese, the Dutch issued in 1645 and 1646, to pay their soldiers, the irst coin inscribed “BRASIL”. Named obsidional, it is said to have been fabricated by melting either African gold or gold tableware. It is only in 1694 that the Brazilian itinerant mint was created in Bahia, and successively closed and transferred to Rio de Janeiro in 1698, to Pernambuco in 1700, and back to Rio de Janeiro in 1702. his itinerary is related to the exhaustion of the local metal supplies, until the discovery of gold in Brazil in the late 1600s. SR-XRF analyses of a small set of coins issued by the Dutch West Indies Company and the irst Rio de Janeiro mint show the use of diferent gold alloys and the ratios of trace elements allow advancing several assumptions on the provenance of the gold. Résumé : Ain de payer la solde de leurs soldats assiégés par les Portugais à Pernambuco, les Hollandais frappèrent, en 1645 et 1646, les premières monnaies portant l’inscription « BRASIL », dites obsidionales, à partir de la fonte d’or africain ou de vaisselle d’or. Ce n’est qu’en 1694 que le premier atelier itinérant brésilien fut créé à Bahia; il fut ensuite fermé et transféré à Rio de Janeiro en 1698, puis à Pernambuco en 1700 avant de s’installer de nouveau à Rio de Janeiro en 1702. Ces délocalisations successives apparaissent liées à l’épuisement des stocks de métal jusqu’aux découvertes de gisements d’or au Brésil à la in du XVIe siècle. L’analyse par SR-XRF de quelques monnaies émises par la Compagnie Hollandaise des Indes Occidentales et d’autres frappées par le premier atelier de Rio de Janeiro indique que l’utilisation de diférents alliages d’or et les teneurs en certains éléments traces apportent quelques indices quant à la provenance de l’or. Keywords: Gold, Brasil, obsidional, coin, SR-XRF. Mots-clés : or, Brésil, monnaie, obsidional, SR-XRF. 1. INTRODUCTION Until the end of the 17th century, many Spanish and Portuguese coins circulated in Brazil with countermarks, which were applied during temporary operating periods in the Capitanias mints. Only emergency issues were struck in 1645 and 1646 by the Dutch to pay their soldiers, besieged on the Pernambuco coast by the Portuguese. he obsidional coins are the very irst coins having Brasil inscribed on the reverse and G.W.C. (Geoctroyeerde Westindische * Institut de Chimie de la Matière Condensée de Bordeaux. UPR 9048 CNRS – 87, Avenue du Docteur-Albert-Schweitzer, 33608 Pessac cedex, France. (m.duttine@icmcb-bordeaux.cnrs.fr) ** Laboratoire du Centre de Recherche et de Recherche et de Restauration des Musées de France, UMR171 CNRS – 14, quai François-Mitterrand, 75001 Paris, France. (maria.guerra@culture.gouv.fr) *** Museu Histórico Nacional, Acervo de Numismática – Praça Marechal Âncora, s/n° 20021-200 Rio de Janeiro-RJ, Brazil. **** Centro Brasileiro de Pesquisas Físicas – Rua Dr. Xavier Sigaud 150, Urca-Rio de Janeiro, 22290-180, Brazil. (scorza@cbpf.br) ****** Instituto Nacional de Tecnologia (INT) – Rua Venezuela 82, 20081-312, Rio de Janeiro, Brazil. ******* Laboratório Nacional de Luz Síncrotron-LNLS/CNPq – Caixa Postal 6192, 13038-970 Campinas, Brazil. (perez@lnls.br) rec. Oct. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 309-312 310 Compagnie), indicating the West Indies Company, on the obverse. hey could have been issued using African gold brought by the ships circulating between the Netherlands, the African coast (to take gold), and Brazil (to take sugar and pau-brasil) or by simple melting of gold and silver tableware (Vieira et al., 2007). It was only on the 8th of March 1694 that a royal decree by Peter II (1667-1706) created the Brazilian mint in Bahia, which was successively closed and transferred to Rio de Janeiro in 1698, to Pernambuco in 1700, and back to Rio de Janeiro in 1702 (Lima, 2005). his itinerary is supposed to be related to the exhaustion of the metal supplies, until the discovery of gold in the state of Minas Gerais in 1695 (Noya Pinto, 1979). In a previous work, it was shown that the irst Bahia mint (1694-1698) struck a mixture of Colombian and other Latin American gold, certainly part of the old supplies, while the second Bahia mint (after 1714) struck a gold typical of the new Brazilian sources in Minas Gerais (Guerra, 2004). he aim of the present study is to conirm whether the same practice was observed in the irst Rio de Janeiro mint, issuing coins in 1699 and 1700, and to provide valuable information regarding the gold used to issue the obsidional gold coins. Mathieu DUTTINE et al. Figure 1: (See colour plate) (a) 4.000 réis gold coin from the mint of Rio de Janeiro (1699); (b) III lorins obsidional gold coin struck in Brazil (1645) for the Dutch West Indies Company. Figure 1 : (Voir planche couleur) (a) Monnaie en or de 4 000 réis frappée par l’atelier de Rio de Janeiro (1699) ; (b) monnaie obsidional en or frappée au Brésil (1645) pour la Compagnie Hollandaise des Indes Occidentales. 2. METHODS AND RESULTS All the gold coins studied in this work belong to the collection of the Museu Histórico Nacional (MHN), Rio de Janeiro. Six coins of 1.000, 2.000 and 4.000 réis issued in 1699 and 1700 by the Rio de Janeiro mint (Fig. 1a) and four obsidional coins (III, VI and XII lorins) struck by the Dutch West Indies Company in 1645 and 1646 (Fig. 1b) were selected for analysis by Synchrotron Radiation X-ray Fluorescence (SR-XRF) at the Laboratório Nacional de Luz Síncrotron (LNLS), Campinas, Brazil. Micro-SR-XRF analyses were performed in air with an incident photon energy of 4.2 keV provided by a Si(111) double crystal (channel-cut type) monochromator (energy resolution ∆E/E=3.10-4 in the 4-14 keV energy range). he photon lux was about 4-x-109 photons/s. he characteristic X-rays were collected in energydispersive mode by a Ge(Li) detector (150 eV FWHM at 4.2 keV) positioned at an angle of 90° with respect to the incident beam. he SR-XRF data was analysed with the AXIL software (Van Espen et al., 1977) in order to evaluate the contribution to ED-XRF spectra of several elements, such as Au, Ag, Cu, Pb, Hg, Pt, Pd, Sn, Sb and Zn. Samples of known composition were used as calibration standards to estimate the atomic concentrations of these elements in the analysed gold coins. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 309-312 Unfortunately, some experimental problems (parasitic X-ray emissions of unidentiied origin) rendered impossible the measurement or even estimation of the Pt and Hg contents for all the analysed coins. Table 1 shows the results obtained for the gold coins analysed by SR-XRF. he ternary diagram in Figure 2a shows that the base alloys used for the fabrication of the obsidional coins difer from those used in the Rio de Janeiro and Bahia mints; these two mints were issuing coins, except one, of equivalent ineness and close silver and copper contents. Figure 2b shows the concentrations of Sn, Sb and Pd (in ppm) normalised to the concentrations of gold in % and to 100%, for the coins issued in Rio de Janeiro analysed in this work and the coins issued in Bahia, published in a previous work (Guerra, 2004). It is clear from this diagram that the metal used in Rio de Janeiro is close to the gold used in the irst Bahia mint. However, the impossibility to quantify the Pt content for the Rio de Janeiro issues does not allow conirming the use of Latin American gold. he obsidional coins were fabricated with a diferent gold. However, the concentrations of Sn, Sb and Pd are clearly distinct from the results obtained for both the Islamic coins struck in Northern Africa with local gold (Godonneau and he irst gold coins struck in Brazil: myth or reality? 311 Au % s Ag % s Cu % s Pd ppm s Sn ppm s Sb ppm s XII lorins 1645 86,1 0,7 12,57 0,25 0,48 0,06 167 38 201 50 151 VI lorins 1645 87,2 0,5 11,20 0,51 0,76 0,03 111 27 258 62 159 III lorins 1645 87,0 0,6 11,45 0,11 0,53 0,02 210 47 229 59 186 30 935 60 III lorins 1646 87,9 0,8 10,59 0,23 0,86 0,05 120 33 271 42 201 27 1250 101 4000 réis 1699 92,6 0,6 4,52 0,04 2,60 0,13 72 11 142 31 108 16 309 26 4000 réis 1700 93,2 0,7 3,73 0,39 2,82 0,14 75 19 82 25 122 10 195 23 2000 réis 1699 91,9 0,6 5,61 0,13 2,34 0,10 99 28 129 27 149 35 307 77 2000 réis 1700 92,2 0,5 4,84 0,07 2,49 0,11 91 21 121 24 107 12 199 15 1000 réis 1699 91,7 0,5 5,36 0,04 2,35 0,12 87 16 178 39 98 19 215 16 1000 réis 1700 92,3 0,8 4,87 0,43 2,51 0,11 101 27 197 27 190 37 357 55 Type of coin Fe ppm s 19 985 43 31 1129 58 Obsidional Rio de Janeiro Table 1: Compositions of the gold coins analysed by SR-XRF (Bahia coins were published in Guerra, 2004). Tableau 1 : Résultats des analyses élémentaires par SR-XRF des monnaies en or étudiées (pour les monnaies émises in Bahia cf. Guerra, 2004). a b Figure 2: Ternary diagram for the analysed coin representing the concentrations of (a) the major elements Au-Ag-Cu (in %) and (b) the trace elements Sn-Sb-Pd (in ppm) normalised to Au (in %) and to 100%. Figure 2 : Diagramme ternaire présentant les teneurs en (a) éléments majeurs Au-Ag-Cu (in %) et (b) éléments traces Sn-Sb-Pd (en ppm) normalisées à la teneur en Au (en %) et à 100%. Guerra, 2002) and from the Portuguese coins struck during their control of São Jorge da Mina on the African Coast (Guerra, 2005). hese results seem to indicate that African gold was not used to fabricate the obsidional coins. he second assumption – melting of gold tableware – can only be veriied by analyses of Brazilian and Dutch gold tableware from the period under consideration. 3. DISCUSSION AND CONCLUSION he analysis of the irst coins struck in Brazil, issued by the United West India Company in Pernambuco, showed the use of a base alloy of poorer quality than the monetary alloy used in the irst itinerant Brazilian mint. Both the ArcheoSciences, revue d’archéométrie, 33, 2009, p. 309-312 Mathieu DUTTINE et al. 312 Bahia (1694-1698) and Rio de Janeiro (1699-1700) mints issued coins made of an equivalent alloy of good quality. he measurement of trace elements characteristic of the gold’s provenance, and the comparison with results previously obtained for Brazilian, Portuguese, Latin American and North African gold coins, showed the similarity of the gold used in the irst Rio de Janeiro and Bahia mints. However, only the quantiication of the Pt contents would allow drawing a conclusion concerning the use of the same mixture of South American gold. he high contents of Sn, Sb and Pd measured for the obsidional coins do not conirm the assumption that gold carried by the Dutch ships circulating from the African Coast to Brazil and to Netherlands was melted to fabricate this coinage. Further analyses appear, however, necessary in order to conirm these results and to show whether gold Brazilian or Dutch tableware could have been used for this purpose. Acknowledgements he authors wish to thank Dr. Martin Radtke for his assistance and helpful comments regarding data calculation and are grateful to the CNRS (France) and to the CNPq (Brazil) for inancial support. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 309-312 References GONDONNEAU, A. and GUERRA, M.F., 2002. he circulation of precious metals in the Arabic Empire: the case of the Near and the Middle East. Archaeometry 44(4): 473 & 599. GUERRA, M.F., 2004. he circulation of South American precious metals in Brazil at the end of the 17th century. Journal of Archaeological Science 31(9): 1225-1236. GUERRA, M.F., 2005. he circulation of gold in the Portuguese area from the 5th to the 18th century, in A. Perea, I. Montero, O. Garcia-Vuelta (eds.), Ancient gold technology: America and Europe. Madrid: Anejos de AespA XXXII, CSIC, 423-431. LIMA F.C.G.C., 2005. Uma Análise Crítica da Literatura Sobre a Oferta e a Circulação de Moeda Metálica no Brasil nos Séculos XVI e XVII. Estudos Econômicos 35(1): 169-201. NOYA PINTO, V., 1979. O ouro brasileiro e o comércio anglo-português. Brasiliana 371. São Paulo, Nacional Edição. VAN ESPEN, P., NULLENS, H. and ADAMS, F., 1977. A computer analysis of X-ray luorescence spectra. Nuclear Instruments and Methods 142 (1-2): 243-250. VIEIRA, R.M.L., GUERRA, M.F., SCORZELLI, R.B., SOUZA AZEVEDO, I., DUTTINE, M. and BRITO PEREIRA, C.E., 2007. Estudo preliminar de algumas moedas holandesas da colecção Museu Histórico Nacional do Rio de Janeiro. Revista Brasileira de Arqueometria, Restauração e Conservação 1(6): 296-300. Precious metals determination in ancient coins by portable ED-XRF spectroscopy with a 238Pu source Détermination des métaux précieux dans les monnaies anciennes par spectroscopie ED-XRF portable avec une source de 238Pu Paulo S. Parreira*, Carlos R. Appoloni*, Rejane M. Lobo Vieira**, Rosa B. Scorzelli***, Laetitia Le Corre*** and Maria Filomena Guerra**** Abstract: A portable X-ray luorescence system (pXRF) was employed to analyse a set of gold, silver and billon coins from the collection of the Museu Histórico Nacional do Rio de Janeiro (MHN), struck during the Brasil Colônia period, under the Dutch occupation, and in Portuguese mints under the reigns of Fernando I (1367 to 1383) and his successor João I (1383 to 1433). he experimental setup consisted of a mini Si-PIN detector and a radioactive source of 238Pu for the excitation of the samples. he aim of the study was to test the possibilities and limitations of this portable system for the determination of the coins’ base alloys, and to verify whether the results obtained were consistent with historical facts. he Brasil Colônia period coins show about 80.5% gold and 96.8% silver. For the Portuguese coins, we could identify a possible monetary debasement during the reign of João I. No matrix corrections were made for these preliminary results. Résumé : Un équipement portable à luorescence X (pFX) a été utilisé pour l’analyse de monnaies en or, en argent et en billon appartenant à la collection du Museu Histórico Nacional de Rio de Janeiro (MHN). Certaines de ces monnaies ont été frappées au Brésil pendant la période Brasil Colônia et sous l’occupation hollandaise, alors que d’autres ont été frappées au Portugal sous les règnes de Ferdinand I (1367-1383) et de son successeur Jean I (13831433). Le montage expérimental est constitué d’un mini détecteur Si-PIN et d’une source radioactive de 238Pu, dont la radiation émise est utilisée comme source d’excitation. Le but de ce travail est double : vériier les possibilités et les limites de ce système portable lors qu’il s’agit de déterminer les alliages constitutifs de monnaies en métal précieux et vériier les rapports entre l’évolution des compositions des monnaies et les diférents faits historiques. Les monnaies appartenant à la période Brasil Colônia sont constituées de 80,5% d’or et de 96,8% d’argent et les monnaies frappées au Portugal montrent une dévaluation sous Jean I. Keywords: coins, ineness, gold, silver, X-ray luorescence. Mots-clés : monnaies, titre, or, argent, FX. * Dept. de Física/CCE, Universidade Estadual de Londrina, Cx.P. 6001, 86051-990 Londrina/PR, Brazil. (parreira@uel.br), (appoloni@uel.br) ** Museu Histórico Nacional,Rio de Janeiro/RJ, Brazil. (arjlobo@terra.com.br) *** Centro Brasileiro de Pesquisas Físicas, Rio de Janeiro/RJ, Brazil. (scorza@cbpf.br) **** Laboratoire du Centre de Recherche et de Restauration des Musées de France, UMR171 CNRS, France. (maria.guerra@culture.gouv.fr) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 313-318 314 1. INTRODUCTION Among the analytical techniques used in the studies of cultural heritage objects, energy dispersive X-ray luorescence (ED-XRF) is a very versatile one due to its portable coniguration allowing in situ measurements. In situ experimental setups (Cesareo et al., 2007; Karydas, 2007) have been used to perform analyses in speciic situations, such as in the case of large sculptures, frescos, paintings, etc., and to characterise metallic alloys, as well as several pigments. In the present work we used a handmade portable X-ray luorescence system (pXRF) (Neiva et al., 2005; Appoloni et al., 2007) to analyse a set of gold, silver and billon coins from the collection of the Museu Histórico Nacional do Rio de Janeiro (MHN). The analyses were carried out in the Numismatic Department of the MHN, and the aim of the study was to test the possibilities and limitations of this portable system for the determination of the coins’ base alloys, and to verify whether the results obtained were consistent with historical facts. he assessment of this portable system was accomplished by comparing some of the compositions obtained in this work with the compositions obtained for the same coins with a ixed ED-XRF, as well as results from the literature on similar coins. Figure 1: (See colour plate) Images of the gold coin of 4000 Réis from the Brasil Colônia period (top) and silver coin from the reign of Fernando I, Portugal (bottom), showing the obverse and reverse sides, respectively. Figure 1 : (Voir planche couleur) Images de l’avers et du revers des monnaies d’or de 4 000 Réis de la période Brasil Colônia (en haut) et d’argent du règne de Ferdinand I, Portugal (en bas). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 313-318 Paulo S. PARREIRA et al. he base alloys were determined for coins struck during the Brasil Colônia period (from 1500 to independence, in 1822) and during the Dutch occupation (since the invasion of Bahia in 1624 to the insurrection of Pernambuco in 1654), as well as for Portuguese coins struck in Portugal during the reigns of Fernando I (1367 to 1383) and his successor João I (1383 to 1433). Figure 1 shows the pictures of a 4000 Réis gold coin (Brasil Colônia period, Bahia, 1695) and a silver coin of Real ‘with FR’ (Reign of Fernando I). For the coins struck in Brazil, it was intended to verify whether a change in composition is visible with the moving of the royal mint from Bahia to Rio de Janeiro in 1698, to Pernambuco in 1700, and to Rio de Janeiro in 1702, before the discovery of gold in Minas Gerais, as well as to obtain the irst results on the quality of the coins struck by the Dutch during the siege of Pernambuco, the so-called ‘obsidional coins’, the irst coins bearing the name of Brazil. Concerning the Portuguese coins, it was under Fernando I that a irst reform of the Portuguese monetary system was observed. his reform, related to the war in Spain, is said to have resulted in a general debasement of the coins that was strengthened under João I, Fernando’s successor. he analysis of coins struck under these two kings should allow verifying this assumption. For a irst approach to those questions, a set of the coins and their dates of issue were selected in collaboration with the Numismatic Department of the MHN as follows. Precious metals determination in ancient coins by portable ED-XRF spectroscopy with a 238Pu source Brasil Colônia period Four coins minted in Bahia (BA), Pernambuco (PE) and Rio de Janeiro (RJ) were analysed. One is a silver coin of 640 Réis, and three are gold coins of 4000 Réis, struck under Pedro II of Portugal. relative values, based on Millazzo’s work (Milazzo, 2004), with the diference that in the present case, instead of the pure massive standards for the determination of the elements’ concentrations, experimental sensitivity curves, Si, were used. Wi = Dutch occupation period he obsidional coins, also called necessity currencies, were the irst emissions minted in Brazilian territory with the inscription of the name Brasil (Vieira et al., 2007). hese coins are rectangular plates. hree gold and three silver coins of diferent values were analysed. Fernando I Five silver coins of Real ‘with FR’, one coin of half Forte, and one coin of half Real ‘with F’ were analysed. João I For this period, a total of six billon coins were analysed: two coins of Real of three and half libras, one coin of half Real of ten soldos, one coin of a quarter Real Cruzado, and two coins of half Real Cruzado. 2. EXPERIMENTAL he experimental setup consisted of a 7 mm2 Si-PIN detector (FWHM of 265 eV for the 6.4 keV Fe line) and a 95 mCi annular radioactive source of 238Pu (13.6 and 17.2 keV lines) for the excitation of the samples, arranged in a speciic handmade system (PXRF-LFNA-01). he focus of the radiation beam strikes an area of approximately 180 mm2, while the coins have surfaces with areas ranging from 314 to 962 mm2; thus, the focus of the radiation beam covers between 57% and 19% of the coins’ area, respectively. For each piece, one measurement was carried out at the centre of the obverse and another one at the centre of the reverse, with 1000 s of excitation-detection time. he spectra were processed by AXIL (IAEA, free access software) and the results were expressed in weight percent. he Au and Ag contents were determined by using their Lα lines and Cu by using its Kα line. he quantitative analysis was performed applying the well-known formula of the fundamental parameters to all the identiied elements, without the matrix correction term. he results were scaled to the unit, equation 1, to obtain 315 Ci ∑Cn (1) Where Ci is the concentration of the i element; Wi is the relative concentration of i element (%). Experimental sensitivities Si were obtained for each of the elements that constitute the gold and silver coins, employing MICROMATTER™ standards (thin ilm standards of the elements and compounds for use in the calibration of X-ray luorescence (XRF) equipment). he sensitivities obtained for the K and L lines are shown in Figures 2 and 3, respectively. In the present work, the calculations were performed with no matrix corrections. his means no corrections for the self absorption of the characteristics X-rays by the matrix, and no corrections for the enhancement efects (luorescence produced by heavier elements over lighter elements in the matrix). his approach does not signiicantly interfere with the quantiication of the major elements Au and Ag of the coins, the main objective of this work, since they are present in high concentrations in the gold and silver coins, respectively, as can be seen in the following discussion of the results. Figure 2: Calculated sensitivity for the K lines. Figure 2 : Sensibilité calculée pour les lignes K. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 313-318 316 Paulo S. PARREIRA et al. Figure 3: Calculated sensitivity for the L lines. Figure 3 : Sensibilité calculée pour les lignes L. 3. RESULTS AND DISCUSSION Figures 4 and 5 show typical spectra of gold and silver coins where the Cu lines and the lines of the excitation source are also presented. he average detection limits for the major elements Au, Ag and Cu were 3.2%, 36.1%, and 1.3% respectively for Au, Ag and billon coins matrix. he results obtained for the Au, Ag and Cu contents are shown in Table 1. he reported results are the average of the two measurements performed on each coin, and its standard deviation. It was also possible to verify the occurrence of some other elements in the X-ray luorescence spectra, such as Ca, Fe, Hg and Pb, even if in much smaller concentrations, a matrix correction being necessary for their correct quantiication. Table 1 shows that all gold coins struck under the Dutch occupation have a homogeneous composition, containing on average 90.3% Au, 8.3% Ag and 0.6% Cu. he value of the coin is obtained by simply changing the gold weight. For example, for a coin of III Florins we could measure 1.87 g, while for a coin of VI Florins we could measure 3.72 g. he same situation was veriied for the silver coins. Measurements of Ag and Au contents for the coins struck during the Dutch occupation were also carried out using the ixed ED-XRF system of the Instituto Nacional de Tecnologia (INT). he values obtained for gold and silver contents, shown in Table 2, are on average 87.9% and 94.8%, respectively, while the values determined with the portable system were 90.4% and 87.7%. It can be seen that the results obtained with the PXRF system for Au in the coins of III and VI Florins show a diference of 2 to 4% from those reported by ArcheoSciences, revue d’archéométrie, 33, 2009, p. 313-318 Figure 4: Typical spectrum obtained for gold coins. Figure 4 : Spectre typique des monnaies en or. Figure 5: Typical spectrum obtained for silver coins. Figure 5 : Spectre typique des monnaies en argent. the INT, while for the content of Ag in the coin of X soldos, the diference was about 7%. his comparison shows that the simpliied choice of no matrix correction was appropriate for the purpose of this work. For the gold and silver coins struck during the Brasil Colônia period (1695-1702), an average of 96.8% and 80.5%, for gold and silver contents, respectively, were Precious metals determination in ancient coins by portable ED-XRF spectroscopy with a 238Pu source Origin III Florins Dutch occupation 1645-1654 Brasil Colônia 1695-1702 Portugal King Fernando I 1367-1383 Date of issue 1646 1645 1645 1654 1654 1654 1702 1695 1699 1701 — — — — — — — — — — — — — Denomination 1 2 VI Florins XXXX Soldos 1 2 X Soldos 4000 Réis (PE) 4000 Réis (BA) 4000 Réis (RJ) 640 Reis (PE) Real with FR ½ Forte ½ Real with F Real of 3 ½ libras Portugal King João I 1383-1433 ½ Real of 10 soldos ¼ of Real cruzado ½ Real cruzado Origin Denomination 1 III Florins Dutch occupation 1645 - 1654 2 VI Florins XXXX Soldos 1 Weight g Au % Ag % Cu % 1.83 1.87 3.72 12.04 — — 8.17 8.03 8.01 18.39 3.5 3.47 3.51 3.55 3.49 1.64 1.84 2.8 1.84 1.9 0.44 2.18 1.89 89.6±0.3 91.0±0.3 90.5±0.3 0.70±0.04 1.8±0.1 0.5±0.1 98.8±0.2 95.8±0.2 95.7±0.2 0.4±0.1 4.1±0.1 1.20±0.04 3.90±0.04 1.6±0.1 1.2±0.1 0.7±0.1 1.1±0.1 — — 0.20±0.04 — — — 9.6±1.5 7.8±1.7 7.6±1.4 88.3±5.7 86.7±5.2 88.1±2.2 — 1.7±0.5 — 81±16 84±14 77±2 79,8±1,5 76±3 82.5±2.1 83.0±3.7 82.6±2.8 14.4±0.8 2.7±0.6 39.2±1.2 15.4±1.4 10.4±0.9 6.8±0.7 0.49±0.04 0.39±0.04 0.80±0.04 10.0±0.1 9.7±0.1 10.9±0.1 0.67±0.02 1.70±0.02 2.19±0.03 17.1±0.1 8.7±0.1 21.0±0.1 14.7±0.1 21.2±0.1 15.2±0.1 14.6±0.1 15.1±0.1 84.7±0.1 93.8±0.1 59.9±0.1 83.7±0.2 79.7±0.1 90.3±0.1 Au % Ag % Cu % 88.09 10.75 0.45 88.95 10.25 0.35 86.82 12.08 0.61 0.19 94.77 4.55 Table 2: Results obtained with the INT stationary EDXRF system. Tableau 2 : Résultats obtenus avec le systèmes ixe de l’INT. determined, while for the period from 1695 to 1698, the values obtained by Guerra (2004) were 91.5% and 93.3%, for coins minted in Bahia. When the results for Cu concentrations are compared, those obtained with PXRF vary from 0.7 to 2.2% for gold coins, while those obtained by Guerra (2004) vary from 0.9 to 1.7%. hese last results were obtained by PAA (Proton Activated Analysis) and TFNAA (hermalised Fast Neutron Activated Analysis), which are absolute techniques with detection limits in the range of the ppm; however, both techniques use the induction of nuclear reactions and require the use of a particle accelerator to carry out the analysis. Concerning the question of whether there was a change in composition of the coins struck in Brazil with the moving of the royal mint from Bahia to Rio de Janeiro in 1698, to Pernambuco in 1700, and to Rio de Janeiro in 1702, it can be observed in Table 1 that there is no diference between gold coins from 1695 and 1699, but that they are diferent from the gold coin from 1702, which has a greater Au content. 317 Table 1: Results obtained for Au, Ag and Cu concentrations with pXRF. Tableau 1 : Résultats obtenus pour les concentrations d’Au, d’Ag et de Cu par pFX. Regarding the coins struck in Portugal under Fernando I and João I, in the work of Norma Botelho Portugal (2000), it is assumed that due to large expenditures incurred because of wars with Castile, internal crises and conquest of the colonies in Africa, the debasement observed during the reign of João I was more important than the debasement observed under his predecessor Fernando I. his monetary debasement was experimentally demonstrated in this work and can be seen in Table 1. Hence, the average concentrations of Ag and Cu, 80.6% and 15.8%, respectively, during the reign of King Fernando I, changed to 14.8% and 82.0%, respectively, during the reign of King João I. hese results show the replacement of silver by copper in the monetary alloys used under the reign of King João I. For some of the analysed coins, it was also possible to verify the presence of Ca, Fe, Hg and Pb, but the concentrations of these elements are close to the detection limits of our system and are strongly afected by the matrix, taking into account the expected higher deviations of pXRF systems (Kump et al., 2005). herefore, the lack of precision in these measurements would also inluence the accuracy of the results for the major elements Au and Ag, since in this methodology the relative determination takes into account all the elements observed in the X-ray spectra. In addition to this, the presence of Ca is generally related to the cleaning processes performed in museums, and the Fe contents may be afected by many other phenomena than metallurgical processes. Only Hg and Pb could provide useful information on the metals and on the extraction techniques. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 313-318 Paulo S. PARREIRA et al. 318 4. CONCLUSIONS References he irst results obtained with our portable XRF system for the measurement of major element (Au, Ag and Cu) concentrations in gold and silver monetary alloys showed a good agreement with the results obtained for the same coins by ED-XRF (Energy Dispersive X-Ray Fluorescence) using a ixed system, as well as with those previously published for coins from the same period, analysed with nuclear activation techniques. We must however denote that the advantage of this portable system is the fact that in situ measurements can be carried out with a relatively high analytical speed without moving the object from the museum to the laboratory. his portable system proved to be a viable technique in the study of coins, but the results may still be improved by considering matrix corrections, i.e. corrections for self-absorption and enhancement efects. he more signiicant limitation of this system (and any portable system) is the fact that it is not possible to obtain reliable results for minor elements, due to the low activity of the radioactive source (or X-ray tube power of other pXRF systems). his limitation could, in part, be improved for future analysis by applying matrix correction. he irst results related to the quality of the X soldos coin struck by the Dutch during the siege of Pernambuco were obtained, results which can be included among similar data for Ag and Au coins already measured (Vieira 2007). hese coins represent the so-called ‘obsidional coins’, the irst coins bearing the name of Brazil. Results indicates that there was no change in the composition of the coins struck in Brazil with the moving of the royal mint from Bahia to Rio de Janeiro in 1698 and to Pernambuco in 1700, but there was an increase in gold content after the returning of the royal mint to Rio de Janeiro in 1702. he results obtained for the silver coins struck in Portugal in the 14th-15th century veriied the assumption that a higher monetary debasement was observed during the reign of João I than during the reign of his predecessor, Fernando I. APPOLONI, C.R., BLONSKI, M.S., PARREIRA, P.S. and SOUZA, L.A.C., 2007. Study of the pigments elementary chemical composition of a painting in process of attribution to Gainsborough employing a portable X-rays luorescence system. Nuclear Instruments and Methods in Physics Research A 580: 710-713. CESAREO, R., FERRETTI, M., GIGANTE, G.E., GUIDA, G., MOIOLI, P., RIDOLFI, S. and GARCIA, C.R., 2007. he use of a European coinage alloy to compare the detection limits of mobile XRF systems. A feasibility study. X-Ray Spectrometry 36: 167-172. GUERRA, M.F., 2004. he circulation of South American precious metals in Brazil at the end of the 17th century. Journal of Archaeological Science 31: 1225-1236. KARYDAS, A.G., 2007. Application of a portable XRF spectrometer for the non-invasive analysis of museum metal artifacts. Annali di Chimica 97(7): 419-432. KUMP, P., NÈCEMER, P. and RUPNIK, P., 2005. Development of the quantiication procedures for in situ XRF analysis. Report IAEA-TECDOC-1456. Vienna. MILAZZO, M., 2004. Radiation applications in art and archaeometry X-ray luorescence applications to archaeometry. Possibility of obtaining non-destructive quantitative analyses. Nuclear Instruments and Methods in Physics Research B 213: 683-692. NEIVA, A.C., MELO, H.G., BENDEZÚ, H.R.P., RIZZUTTO, M.A., TABACNIKS, M.H., ADDED, N., APPOLONI, C.R., PARREIRA, P.S. and CUNHA LIMA, S., 2005. Caracterização espectroscópica das ligas e dos produtos de corrosão de peças metálicas do Museu de Arqueologia e Etnologia da USP, in Proceedings of 2o Congresso Latino-Americano de Restauração de Metais, Rio de Janeiro, 25 to 28 July of 2005. Rio de Janeiro: Museu de Astronomia e Ciências Ains (MAST), 143-157. PORTUGAL, N.B., 2000. D. João I, Regência e reinado, in R.M.L. Vieira (ed.), Moedas Portuguesas da Época dos Descobrimentos na Coleção do Museu Histórico Nacional 1383-1583. Rio de Janeiro: Museu Histórico Nacional Press, 29-51. VIEIRA, R.M.L., GUERRA, M.F., SCORZELLI, R.B., SOUZA AZEVEDO, I., DUTTINE, M. and BRITO PEREIRA, C.E., 2007. Estudo preliminar de algumas moedas holandesas da coleção do Museu Histórico Nacional do Rio de Janeiro. Revista Brasileira de Arqueometria, Restauração e Conservação 1(6): 296-300. Acknowledgements We are grateful to LIN/CENA/USP for the loan of the 238Pu source, to Centro Brasileiro de Pesquisas Físicas for the inancial support, and to the Museu Histórico Nacional for providing the coins. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 313-318 Gold jewellery of devotional images in Campeche State, Mexico Bijoux en or des images de dévotion dans l’état de Campeche, Mexique Eva Brito* and Feliciano Chaire** Abstract: he objective of this work is to understand the cultural signiicance of gold, speciically in the use of jewels for the decoration of devotional sculptures, for a population of Mayan origin that currently lives in the State of Campeche, Mexico. We study three cases of cult images of great reputation among Catholics: one from Campeche City and two others from rural areas. he method mainly consisted in recovering data from historic documents, living testimonies and oral traditions, and carrying out an analysis of religious festivities. he cultural value given to these objects by people makes it necessary to recognize them as Communal Cultural Heritage. Only through joint action and cooperation between Catholic communities, the Mexican Government, Church and society in general, we will be able to guarantee their security and preservation. Résumé : L’objectif de l’étude est de comprendre la signiication culturelle de l’or pour la population d’origine Maya qui vit actuellement dans l’état de Campeche, au Mexique, particulièrement en ce qui concerne l’utilisation de bijoux pour la décoration de sculptures de dévotion. Trois images de culte, qui jouissent d’excellente réputation parmi les catholiques, ont été étudiées : une située dans la ville de Campeche et les autres deux dans des villes rurales. La méthode comprend l’étude de documents anciens, la récolte de témoignages et traditions orales en direct et l’analyse des fêtes religieuses. La valeur culturelle que les gens donnent à ces objets exige qu’ils soient reconnus comme un héritage de la communauté culturelle. Seule une action conjointe et la coopération entre les communautés catholiques, le gouvernement mexicain, l’église et la société en général, nous garantira leur sécurité et leur préservation. Keywords: gold jewels, devotional images, Mayan ethnic groups, popular religiosity, México. Mots-clés : bijoux en or, images de dévotion, groupes ethniques Maya, religiosité populaire, Mexique. 1. INTRODUCTION he ancient Mayan civilization lourished in south-eastern Mexico, in an area including the current states of Campeche, Yucatán, and Quintana Roo (Yucatan Peninsula), Chiapas and Tabasco, extending to Nicaragua in Central America. It was characterized by a polytheistic religion in which gods were classiied according to diferent categories, such as: the creation of the Universe; stars and planets (e.g. sun, moon, Venus); natural phenomena (e.g. rain, wind, thunder); animals (e.g. jaguar, snake); plants (maize, for example); good and evil; and death. Both historical documentary sources and archaeological evidence prove that gold was considered a very important gift to these deities. With the Spanish conquest of America during the 16th century, an unknown god and new saints came to these * Instituto Nacional de Antropología e Historia – Centro INAH Estado de México, Avenida José Vicente Villada 103, 105 y 107, Toluca, Estado de México, C.P. 50000. (evalebrito@yahoo.com.mx) ** Archaeologist (Escuela Nacional de Antropología e Historia), independent collaborator – Calle José Loco Adam 11, Fraccionamiento Villamar, Campeche, Cam., México, C.P. 24020. (fchaire@live.com.mx) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 319-325 320 lands and a process of syncretism of Mayan and Catholic beliefs was initiated (López, 1994). his phenomenon is visible until today in the living manifestations of regional popular piety. In Campeche State, we can still witness the thousand years old relationship between humans and the sacred, using the precious metal par excellence as a mediator. It is worth mentioning that today, Campeche State has a total population of 754,730 inhabitants and that 13.2% of it has the native language as the primary one. Addressing this topic, the objective of the present work is to understand the cultural signiicance of gold, speciically in the use of jewels for the decoration of devotional sculptures, for the population of Mayan ethnic origin that currently lives in this part of Mexico. We speciically studied the cases of three of the cult images that have the best reputation among Catholics: the Black Christ of the Parish of Saint Roman the Martyr, in Campeche City (the capital of the state); the Immaculate Conception of the Virgin Mary, in the village of Pomuch; and the Lord of Health, in the town of Hecelchakán. he local communities contributed money to buy expensive crowns, necklaces, bracelets, cruciixes, and other golden objects to adorn these sculptures. Finally, we raise some points of discussion and a conclusion focused on the cultural importance of this type of objects, and the necessity to recognize them as Communal Cultural Heritage. 2. METHOD he irst phase of our study included documentary research in order to collect data on the importance of gold for the ancient Mayan society, consulting the works of J.E. hompson (1982), M. de la Garza (1990) and M.O. Marion (1994), three prestigious researchers of this area. he archaeological evidence pertaining to this subject was taken from the results published by C. Coggins (1994) and M. Schmidt (1994). We also reviewed chronicles of the history of the Church in colonial times, written by Diego López de Cogolludo (17th century, reprinted 1996), which helped us to select the three cases of cult images for this study. he ieldwork was the longest and most important part of this investigation, which began after we decided to examine one case from Campeche City and two others from towns located on the side of the road that connects the states of Campeche and Yucatan. It consisted, irstly, in the observation and description of the main features of the sculptures, as well as of the jewellery collection of each of them. hen, we registered the calendar of religious holidays and attended to those considered the most signiicant. During the festiArcheoSciences, revue d’archéométrie, 33, 2009, p. 319-325 Eva BRITO, Feliciano CHAIRE vities, we collected information about the traditions transmitted in oral form, and we also obtained testimonies about the meaning of gold for communal members. We then confronted all the data we collected in order to be able to draw conclusions regarding the current role played by gold jewellery for popular religiosity in the region. Finally, we evaluated this role based on the Unesco deinition of culture and cultural heritage. 3. AURUM: THE GIFT PAR EXCELLENCE FOR MAYAN DEITIES he religious ideas of the ancient Mayans represented the guiding axis of their lives. hey conceived a Universe inhabited by invisible deities characterized by a dual personality: fussy and dangerous, but also kind and gentle. herefore, it was possible for human beings to awake their commiseration, to avoid their attacks and receive their blessings, satisfying their demands with rituals, sacriices and presents (Garza, 1990; hompson, 1982). he main objective of these rituals was to ensure the arrival of the rains in order to guarantee the agricultural cycle, especially of maize, which was the basis of their food and was also considered a holy plant. Representations of the rain god Chac, which we can still see on the facades of the pre-Columbian ceremonial centres along the old Camino Real, express concern over this matter (Marion, 1994). One way to please the sacred beings was ofering them gold gifts, which were thrown into the cenotes, the natural water deposits located in the Yucatan Peninsula. he clearest example of this is the Sacred Cenote of Chichen Itza (an ancient city that reached its apogee between the years 800 and 1100 AD), which was used as a great receptacle for this type of artefacts (Coggins, 1994; Schmidt, 1994). However, the essence of this tradition is not a thing of the past, as current examples are observed in Campeche State. he irst one is of a place named Hool, where the devotees ask for favours to the Virgin of Candelaria by throwing golden jewels in a lake beside the village. he second example is that of the small town of Chuiná, in which, during Easter, pilgrims from diferent locations launch gold jewels into the Laguna de Términos, formed by the salt water of the Gulf of Mexico. At the same time, they pray to Our Lady of Sorrows, which is best known precisely as the Virgin of Chuiná. his practice shows the syncretism of the Mayan and Catholic religions, which undoubtedly agree on the concept of gold as an ideal present for the deities. Gold jewellery of devotional images in Campeche State, Mexico 4. THE JEWELS OF THE LORD 321 he town and port of San Francisco de Campeche was established in the year 1540 by the Spanish Captain Francisco de Montejo; in 1776, it received the title of City, and in 1857 it became the capital of Campeche State (Brito, 2007). Oral tradition holds that during the second half of the 16th century a small chapel was built by local people to implore God to cease the attack of a plague of locusts. hey also commissioned an image of Christ to a wealthy merchant and a frequent traveller, named Juan Cano de Coca Gaitan. After he bought the artwork, and during the return trip, a irst miracle happened: the ship had a fast and successful travel on a rough sea, and inally a safe arrival to Campeche on September 14 of 1565. It was since then that the seamen adopted him as their Patron Saint; until today, this representation is the most venerated statue of the Diocese of Campeche, having the title of Patron Saint of all the residents of Campeche (Cahuich and Aguayo 1998) (Fig. 1). he Christ is an anonymous sculpture of unknown origin, probably manufactured with maize pastry, dark brown, and nearly two meters tall. he religious celebration takes place in September, and it has more than four hundred years of annual performance, becoming one of the oldest traditions not only in Campeche State, but in all Mexican territory. During the celebrations, in a similar way to the relationship between the ancient Mayan gods and the vital luid, the ishermen make a trip along the seashore with the Lord on a boat. his is the occasion when the statue proudly displays its jewellery (Fig. 2). he sculpture’s collection includes gold and silver objects, such as crowns, cruciixes, medals and nails, from the 19th and 20th centuries, donated by the diferent religious guilds of bakers, carpenters, masons, women and others (Fig. 3). he jewels are housed in the temple, in a safe-deposit box under the supervision of the parish priest, but have no inventory or photographic record. Unfortunately, some of them were lost without any trace. Figure 1: (See colour plate) he “Patron Saint of all the residents of Campeche” of the Diocese of Campeche (Photography: A. Rosado). Figure 1 : (Voir planche couleur) Le « Saint Patron de tous les résidents de Campeche » du Diocèse de Campeche (Photographie : A. Rosado). Figure 2: he “Patron Saint of all the residents of Campeche” of the Diocese of Campeche. (Photography: A. Rosado). Figure 2 : Le « Saint Patron de tous les résidents de Campeche » du Diocèse de Campeche. (Photographie : A. Rosado). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 319-325 322 Figure 3: (See colour plate) One of the gold and silver objects from the collection of the Diocese of Campeche. (Photography: A. Rosado). Figure 3 : (Voir planche couleur) Un des objets en or et argent de la collection du Diocèse de Campeche. (Photographie : A. Rosado). Eva BRITO, Feliciano CHAIRE he Patron Saint is the Immaculate Conception of the Virgin Mary, represented by a wooden carved and polychrome statue about 1.50 meters tall, with blue dress decorated with small silver stars. It is an anonymous work that seems to correspond to either the second half of the 17th century or the irst half of the 18th century. It represents a young woman of regular complexion, in standing position and attitude of prayer (Fig. 4). According to oral tradition, the chaste Lady appeared miraculously inside a cave next to an underground river; this aspect is again indicative of the relationship between divine beings and water (Ucan, undated). Nowadays she is ensconced in a niche at the centre of the main altar and is the protagonist of three annual events: the Holy Day of the Immaculate Conception of the Virgin Mary on December 8; the Day of the Assumption of the Virgin Mary on August 15; and a festival in April just after Easter, when village women carry the image in a procession (Fig. 5). All this information was provided by Mr. Rafael Pérez Novelo, trader of Pomuch. he story of her jewels began in 1981, when two sisters of the village, Noemí and Emma Carrillo Briseño, convinced At this location, we encountered a problem with the priest, who did not allow us to see the collection, much less analyze it, even though we had a written permission from the bishop. he solution we found was that another church member (a deacon) took photographs of the objects and gave us a copy. 5. A GOLDEN REWARD FOR CHASTITY Pumuch is a small village of Mayan origin, located on one side of the Camino Real, an old road that connects the states of Campeche and Yucatán. he word Pumuch comes from two Mayan terms: poc, meaning ‘burn’; and much, meaning ‘toad’; therefore, it can be translated as “Place where the toads sunbathe”. Today the population consists of 8,180 inhabitants, of whom 3,029 are reported as speakers of Mayan language; their economic base is agriculture (INEGI, 2005). With the Spanish conquest, the monks of the order of San Francis of Assisi arrived in the mid-16th century to the Yucatan Peninsula with the mission of evangelizing the natives. In Pomuch, they built a rustic wooden chapel with leaf roofs, which years later became a masonry building that was consecrated on January 9, 1636, as indicated by an inscription located on the facade of the temple, to the left of the front door. Currently, the Pomuch Church is under the ecclesiastical jurisdiction of the Parish of Hecelchakán, assigned to the Diocese of Campeche. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 319-325 Figure 4: he Immaculate Conception of the Virgin Mary from the Pomuch Church. (Photography: A. Rosado). Figure 4 : L’Immaculé Conception de la Vierge Marie de l’Église de Pomuch. (Photographie : A. Rosado). Gold jewellery of devotional images in Campeche State, Mexico 323 he Catholic community of Pomuch provided us with all kinds of facilities to study the sculpture, and allowed us to photograph it. In addition, they also gave us some photographs. Many people tried to help us by providing testimonies and mentioning elements of oral traditions, denoting a surprising openness to people who are not part of the village. 6. EXCHANGING GOLD FOR RELIEF Figure 5: (See colour plate) Easter procession at Pomuch. (Photography: C. Ucan). Figure 5 : (Voir planche couleur) Procession de Pâques à Pomuch. (Photographie : C. Ucan). the community that their Queen needed a crown. he Catholics accepted, and the two sisters organized the purchase and all members of the community worked together very hard to collect money. Finally they bought an 18-karat gold crown, weighing 4 kilograms, with stone inlays simulating rubies and brilliants. It cost 140,000 Mexican pesos (about 1,200 US dollars), which is a signiicant sum, considering the poverty of the peasant population. A few years later, three thieves climbed to the roof of the temple and tried to steal the jewel, but one of them fell and was successfully apprehended by the police, while the other two were able to lee. Due to the fact that the theft was not consummated, the captured ofender obtained his freedom. Today, the chaste Lady wears her crown permanently, together with the other gold pieces that make up its collection (rings, necklaces, medals, cruciixes, earrings and rosaries), all of them donated by local devotees. Other metal jewels (not made of gold) are kept locked in a safe-deposit box in the temple, whose responsibility is with the sacristan. here is no inventory or photographic record, but people airm that there is no missing jewellery. he word Helelchakán comes from the Mayan words helel and chakán, meaning ‘rest’ and ‘savannah’, respectively. herefore, it can be translated as “Savannah for resting”. his is a town located on the old Camino Real; it was founded by Mayan Indians before the Spanish arrival. In 1579, the Franciscan monks built in this place a temple and a convent, which worked as a base for the evangelization of neighbouring sites (López de Cogolludo 1996). Nowadays, it has the title of Parish of Hecelchakán, assigned to the Diocese of Campeche. he Patron Saint is San Francis of Assisi, but the most venerated image is a Black Christ, known as the Lord of Health because of its reputation for curing illnesses (Fig. 6). his is a sculpture of carved wood, dark brown, about 2 meters tall. It is an anonymous work of unknown origin, but because of its similarity to the Christ of Saint Roman, it is considered an artwork of the 16th century. he religious festival in its honour is performed in April, when the statue is taken on a tour around the town, and the guilds and civil authorities organize bullights, rides, handicraft selling, cooking of regional foods, ireworks, and folk dances. Figure 6: (See colour plate) “Lord of Health”, Black Christ of the Diocese of Campeche. (Photography: E. Brito). Figure 6 : (Voir planche couleur) Le « Seigneur de la Santé », Christ Noir du Diocèse de Campech. (Photographie : E. Brito). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 319-325 Eva BRITO, Feliciano CHAIRE 324 he Lord has several cruciixes and a crown of gold (part of it), and nails of silver with golden cover, all of them donated by the parishioners asking for health and relief or thanking for it. he jewels are kept by one of the ministers of the Church at his own home. here is no inventory or photographic record, but members of the Catholic community say they have not sufered any loss or theft of jewellery. At this location, we encountered a similar problem to the one at Saint Roman, because the parish priest and some members of the community did not authorize the minister to show us the collection that is stored in his house, even though we had a written permission of the bishop, nor did we have the possibility to photograph it. he data that we obtained was in oral form, from other members of the Catholic community and priests, and the result of our own personal observations and photographic record during the religious events. 7. DISCUSSION AND CONCLUSION In addition to the historical signiicance that gold had for the pre-Columbian civilization, it still maintains to this day an important meaning for the current Mayan population in Campeche State, where it is still possible to observe practices of popular religiosity that recover past cultural elements. Neighbours of the towns of Hool and Chuiná throw jewels into the lake and lagoon similarly to how their ancestors did it in the cenotes. However, the ieldwork of this study allowed us to conclude that nowadays, people prefer ofering more personal items to their statues. For this reason, they donate jewels such as earrings, necklaces, rings, cruciixes and crowns, which come into physical contact with the bodies of their sacred images. If the contact is closer, believers feel they are closer to God and the saints, and thus, they materialize their need to be comforted, healed, forgiven, blessed, but most of all, to be heard praying for the eternal salvation of their souls. In this way, devotees prove their faith to themselves and to the community, repeating the cycle every time that the image is made to carry and show the presents, especially during holy celebrations. Added to this, as gold has historically represented one of the ideal gifts for the gods, the belief exists that, with a jewel of this material, a saint will be more pleased than with objects of other types of material. If the jewel is iner, the mortal is believed to be closer to heaven. he fact that we were not allowed to personally examine or photograph the jewellery collections of two of the representations we studied, rather than relecting a safety ArcheoSciences, revue d’archéométrie, 33, 2009, p. 319-325 concern, indicates that the objects have already been transferred to a sacred level. Moreover, these objects become mediators between God or the represented saints, and the donors, whether individuals or groups (as the religious guilds). Because of this, only their owners and custodians are entitled access to the jewellery. With this experience, we conclude that these gold pieces go beyond their purely monetary value, and become an essential part of the religious beliefs and rituals. In this regard, the deinition given by the United Nations Educational, Scientiic and Cultural Organization (Unesco) for cultural heritage of the people states that it includes: […] the body of values which give meaning to life. It includes both tangible and intangible works through which the creativity of that people inds expression: languages, rites, beliefs […] (Unesco, 1982). hus, according to this deinition, the gold jewels of devotional images need to be recognized as Communal Cultural Heritage. his recognition will generate rights for the owners of the materials, but also an obligation to ensure their safety and preservation. Nevertheless, this commitment must also be assumed by the Mexican government, the Church, religious tourism, and society in general. Only then we will be able to preserve the culture that is producing these assets, in this case, the culture of the population of Mayan origin that currently lives in the State of Campeche, whose wealth is a gift of gold that Mexico ofers to the entire world. Aknowledgements Alfonso de Maria y Campos Castelló, Director of Instituto Nacional de Antropología e Historia; Bishop Dr. Ramón Castro Castro and Priest Armando Rosado Cel, of Diocese of Campeche; Lic. Carlos Ucan Yam, Mrs. Emma and Mimí Carrillo Briseño, and Mr. Rafael Pérez Novelo, of the Village of Pomuch. References BRITO, E.L., 2007. La Historia de la Catedral (México). Gobierno Constitucional del Estado de Campeche. CAHUICH, G. and AGUAYO, M., 1998. La Feria de San Román. Historia de una mentalidad 1505-1997 (México). Secretaría de Cultura y Deporte, Secretaría de Turismo del Gobierno del Estado de Campeche, Instituto de Cultura de Campeche. COGGINS, C.Ch., 1994. El cenote sagrado, Arqueología Mexicana 2(7): 47-49. Gold jewellery of devotional images in Campeche State, Mexico GARZA, M. de La, 1990. El hombre en el pensamiento religioso náhuatl y maya (México). Universidad Nacional Autónoma de México. INEGI, 2005. II Conteo de Población y Vivienda (México). Instituto Nacional de Estadística y Geografía. LÓPEZ, A., 1994. El conejo en la cara de la luna (México). Consejo Nacional par la Cultura y las Artes, Instituto Nacional Indigenista. LÓPEZ DE COGOLLUDO, D., 1996. Historia de Yucatán 1688. Tomo II (Campeche), Heroico Ayuntamiento de Campeche. MARION, M.O., 1994. Fiestas de los pueblos indígenas. Identidad y ritualidad entre los mayas (México), Secretaría de Desarrollo Social, Instituto Nacional Indigenista. 325 SCHMIDT, M., 1994. Chichén Itzá, Arqueología Mexicana 2(7): 38-45. THOMPSON, J.E., 1982. Historia y religión de los mayas (México), Siglo XXI Editores, S.A. UCAN, C.E., UNDATED. Leyendas Purísima Concepción de la Virgen (Campeche). UNESCO. World Conference of Cultural Policies, Mexico City, 26 July -6 August, 1982. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 319-325 A:    AuthentificAtion : ApplicAtion des connAissAnces du spéciAliste Forgeries and public collections Les contrefaçons et les collections publiques Susan La Niece* Abstract: Large public collections, however prestigious, inevitably have some forgeries in their storerooms (and occasionally on display) and, although this aspect may seem reprehensible in a publicly funded collection, it is almost inevitable, given the means by which collections are built up by museums. All collectors are vulnerable to forgers who customise their products to tempt the buyer with something that will ill an important gap in their collection, and museums are just as likely to fall for this as private collectors. Forgeries may also enter museum collections as gifts. Museums not infrequently receive bequests from generous patrons, sometimes comprising large numbers of objects, and amongst them a forgery may be included, unrecognised, and may remain unquestioned for many years before coming to the attention of a curator or visiting scholar. his paper presents some case studies of objects in the British Museum for which scientiic analysis has demonstrated that either the materials or the technology used in their making are not compatible with their supposed date of manufacture. Résumé : Les collections publiques, bien que prestigieuses, contiennent inévitablement quelques objets modernes dont certains sont exposés. Bien que cela puisse sembler critiquable dans le cas de collections publiques, ceci est inévitable au vu de la façon dont les collections des musées se sont formées au cours des décennies. Tout collectionneur d’art se trouve vulnérable face aux faussaires qui adaptent leurs « produits » pour correspondre à l’attente des acquéreurs ; les collectionneurs privés cherchant à combler une lacune importante dans leurs collections, tout comme les musées. Les faux ou les copies modernes, peuvent également être entrés dans les collections publiques par voie de donation. Il est en efet courant que les musées publics reçoivent des donations de personnalités importantes, collections qui sont parfois conséquentes et possédent un ou plusieurs faux. L’authenticité de ces objets ne sera pas questionnée, parfois pendant plusieurs années, jusqu’au moment où, par chance, dans le cadre d’un projet de recherche spéciique, ils soient proprement analysés par des méthodes scientiiques. Cet article présente quelques cas d’études d’objets appartenant à la collection du British Museum, pour lesquels l’étude scientiique a pu prouver que les techniques de fabrications utilisées ne pouvaient pas être compatibles avec leurs dates de fabrication supposées. Keywords: Analysis, forgery, gold, pastiche, restoration, SEM. Mots-clés : analyse, faux, or, pastiche, restauration, MEB. 1. INTRODUCTION he basic rule of free-market economics – that increased demand will lead to an increase in supply – holds equally true for the supply of antiquities. he supply of genuine antiquities is of course inite, and so the demand has to be met by new ‘antiquities’. In 19th century Europe, collecting antiquities became particularly fashionable amongst gentlemen of means (though collecting goes back much earlier), and the output of forgeries of collectibles, which could encompass anything from Renaissance jewellery to instruments of torture of the Spanish Inquisition, turned into an expanding industry (Jones, 1990). All collectors are vulnerable to forgers who customise their product to tempt the buyer with something that will ill an important gap in their collections, and museums are just as likely as the private collector to fall into the trap. he so-called Tiara of Saitaphernes, who was a historical 3rd century * Scientiic Research, he British Museum – Great Russell Street, London WC1B 3DG, UK. (slaniece@thebritishmuseum.ac.uk) rec. Aug. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 329-333 330 BC Scythian king, was purchased by the Musée du Louvre in 1896 (Collignon, 1899; de Pradenne, 1932). In retrospect, it can be observed that the desire to acquire this unique and undoubtedly ine example of the goldsmith’s art overcame caution. It soon became known that it had been made by a Russian goldsmith and that the scenes depicted on it were inspired by illustrations from contemporary books on Greek vases, as well as by the tales of the Iliad and Odyssey of Homer and scenes from Scythian life, mixing together styles of diferent dates. Unlike private collectors, who almost invariably acquire by purchase, museums may acquire their collections by other routes. he irst of these is direct from archaeological excavation, which usually provides the best possible provenance. However, it is not unknown for deception to occur even on excavations, whether as a practical joke, attention seeking or malice. A second route by which items enter public collections is by donation, either of individual items or of a complete collection (often on the death of a collector, to mitigate taxation). Such large groups of artefacts sometimes include forgeries amongst the genuine pieces, and the forgeries may remain unquestioned for many years before coming to the attention of a curator or visiting scholar. King George III’s large collection of coins and medals, deposited at the British Museum in 1825, included a gold Roman coin depicting ‘Brutus’ (CM B.11447). It was once thought to represent Brutus, the legendary founder of Britain, and was thus considered very desirable at the time, as a missing link in a chronological sequence of coins and medals representing the rulers of Britain. his coin is in fact a close copy in gold of an ancient Roman silver coin depicting Marcus Junius Brutus, the assassin of Julius Caesar. Any artefact of gold will tend to command a higher price than the equivalent item made of other materials, and in some respects it is easier to produce a convincing ‘antiquity’ from gold than, for example, from silver. Gold is a particularly diicult material to authenticate scientiically, although this situation is already changing with the development of new techniques (Eugster et al., 2009). he forensic approach to the detection of forgeries relies on inding anachronisms in the materials or techniques used to make them, and a comprehensive database of materials and techniques is therefore of prime importance (Craddock, 2009). he holy grail of a simple test for authenticity may be approaching, but in reality there is as yet no substitute for knowledge acquired by in-depth scientiic research into the history of technology, as the following case studies demonstrate. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 329-333 Susan LA NIECE 2. EARLY MEDIEVAL EUROPEAN JEWELLERY Early medieval European jewellery became particularly popular amongst late 19th/early 20th century collectors, attracted to the artefacts of the ancestors of modern European nations. By the early 20th century, major collections in Europe and the USA contained forgeries of jewellery purporting to be from this period, and also a number of ‘improved’ genuine artefacts, such as a genuine iron Merovingian buckle (BM1928,0720.1) which has on it a copper-alloy overlay with what was claimed to be the earliest representation of the Cruciixion (Smith, 1928; Baum, 1937). However, radiography revealed the original surface, with its silver-inlaid decoration, beneath the faked additions. When an object has a feature which makes it the earliest or indeed the only example of its type, it merits very close inspection. A group of more than 30 items of gold work, purporting to originate from northern Italy, from the tomb of the Lombardic King Agilulf (died AD 615) and from that of his queen, heodelinda (died AD 628), was in fact manufactured about 1929-30. It is no coincidence that it was contemporaneous with the rise to power of Mussolini, as the Treasure was designed deliberately to appeal to a sense of nationalism. Individual items are based on genuine pieces. For example, the buckle is an extravagant combination of features from a number of buckles illustrated in the literature, and the dagger handle is remarkably like those on swords carried by the four emperors in a stone sculpture at San Marco in Venice. he inscription on the dagger bears the name of heodelinda and the iconography on the other pieces derives from the famous Lombardic helmet fragment depicting her husband King Agilulf, preserved in the Bargello Museum in Florence. Doubts grew about the Treasure on stylistic grounds and it failed to ind a buyer, and was thus withdrawn from the market. Dafydd Kidd, then a curator at the British Museum, was interested to see what a scientiic examination of these forgeries might reveal of the techniques being employed by the forgers, and in 1988 he purchased two gold mounts for a study. he scientiic examination of these two pieces ascertained that the gold was of unusually high purity, and that the solder of the major structural joint of each item contained the element cadmium, an element not found in ancient solders (Meeks and Craddock, 1991). A buckle from the Treasure has red and white glass inlay imitating cloisonné work (Fig. 1). he white glass is opaciied with arsenic oxide, which was not used for glass making until the 18th century (Craddock, 2009: 213, 224-5). here are numerous other anachronisms in the technology, although the iron blade of the dagger may well be ancient. Forgeries and public collections 331 Figure 1: (See colour plate) Items from the Lombard treasure. Figure 1 : (Voir planche couleur) Objets du trésor des Lombards. 3. AN ELECTROFORMED ‘PRE-COLUMBIAN GOLD’ ORNAMENT An extensive survey of the collection of pre-Columbian gold at the British Museum, utilizing a range of scientiic techniques, discovered that a fragmentary miniature model of a spouted vessel, in the style of the Chimu culture of pre-Columbian Peru, was not what it appeared to be (BM1947Am21.1) (Fig. 2). he artefact was donated to the British Museum in 1947, and from a visual examination was assumed to be of cast gold alloy; indeed, it was until recently registered as being made of gold. However, X-ray luorescence analysis and examination by scanning electron microscopy (SEM) showed that the metal is not gold, but pure copper which has been gilded. Furthermore, the columnar microstructure of a small polished cross-section taken from a broken edge indicated that it was neither cast nor worked; rather, it is an electroform, which was subsequently electroplated with a thin layer of gold. Electroforming is a modern method of depositing metal from solution onto the surface of a model by means of an electric current, a technique for making metal items which does not occur until the mid-19th century. 4. EGYPTIAN NECKLACE WITH FLY PENDANTS An Egyptian-style necklace of unknown provenance was donated to the British Museum in 1939 and was accepted Figure 2: (See colour plate) Fragmentary Chimu-style object (height c. 40 mm) with (right) a polished section through the metal revealing the columnar microstructure indicative of electroforming. (SEM BSE image). Figure 2 : (Voir planche couleur) Objet fragmentaire de type Chimu (hauteur c. 40 mm) avec (à droite) une section polie à travers le métal indiquant la microstructure colonnaire typique de l’électroformage. (image BSE au MEB). until recently as dating to the New Kingdom (EA65279) (Andrews, 1996). It consists of 38 cast, solid-gold ly pendant amulets and spherical almandine garnet beads of 4-6 mm in diameter (Fig. 3). he garnet beads are considered peripheral to the question of authenticity, as there is no certainty that the beads were found together with the gold amulets: the stringing is modern and no garnet beads of the size and quality of these could be found amongst the extensive collections of Egyptian beads at the British Museum. A scientiic investigation of the piece established that the alloy composition would not be out of place for the alloy ranges found amongst Egyptian New Kingdom ArcheoSciences, revue d’archéométrie, 33, 2009, p. 329-333 332 Susan LA NIECE family irm in Rome dealt in antiquities as well as making and restoring jewellery in the antique style so fashionable at the time. A study of the necklace using SEM established that one half of the necklace corresponds to the expectations for genuine Hellenistic Greek goldsmithing, but the wire used on the other half is manufactured by drawing, not by the classical strip-twist method, and is 99% pure gold (Meeks, 2007). It is therefore likely that this half was made by the Castellani irm. Presumably the original necklace was extensively damaged and had to be restored in the Castellani workshops to pass it of as a near perfect antiquity. 5. CONCLUSION Figure 3: (See colour plate) Garnet and gold necklace with ly amulets, purporting to be New Kingdom period, Egypt. Figure 3 : (Voir planche couleur) Collier de grenats et or avec des amulettes en forme de mouche, attribué au Nouvel Empire, Égypte. gold work, with circa 75% gold, 20% silver and 5% copper, but it could equally be described as a modern 18 carat alloy. he ‘wear’ on the gold is unusually coarse, but it is the manufacturing technique used to make the amulets that is most unusual. Large, cast amulets of Egyptian deities do exist, but sets of small identical pendants such as those on this necklace are usually made of foil and are hollow. In addition, those which are undoubtedly of ancient manufacture are mostly threaded via holes through the body or head of the ly, whereas these have a suspension ring at the top. Lilyquist’s study of Egyptian ly amulets identiied some examples of modern manufacture, though none of them was cast (Lilyquist, 2003). It is possible that well-provenanced examples of cast solid-gold ly pendants will be excavated in the future and vindicate this piece, but until then its authenticity remains very much in doubt. he necklace was a bequest, presumably in good faith, and there may have been no intention to deceive by either the donor or even by the maker. It is perfectly possible that it was made as a wearable piece of modern jewellery in the Egyptian style, but over the years, after changing hands, its origins have been lost. If this were the case, it would not of course be deined as a forgery, because forgery implies a deliberate attempt to deceive. A Hellenistic gold strap necklace with pendants acquired by the British Museum in 1872 from Alessandro Castellani does seem to have been made deliberately to deceive (BM1872,0604.651). he Castellani ArcheoSciences, revue d’archéométrie, 33, 2009, p. 329-333 To quote Mark Jones (1990), the curator of the exhibition ‘Fake? he Art of Deception’ at the British Museum and now Director of the Victoria and Albert Museum, London, it is “[…] not that the less well informed may sometimes make mistakes, but even the most …rigorously organised of institutions can and will occasionally be wrong. And this is not simply because knowledge and experience can never be complete, but because perception itself is determined by the structure of expectations that underpins it.” Curatorial judgements of potential purchases cannot be truly objective; all of us are inluenced by the times we live in. Some of the famous forgeries of the 19th century may seem obvious to us now, for example, with faces depicted according to the conventions of beauty fashionable in the 19th century. A forensic scientiic approach, on the other hand, can be used to determine whether the materials and techniques are consistent with the date attributed to the piece. his of course requires extensive knowledge of what materials and techniques were current during the cultural period concerned, and such knowledge can only be acquired by the scientiic study of large numbers of genuine artefacts. It is an advantage of large collections that they are likely to include many genuine pieces of well attested provenance which can be used for comparison. It is a truism that it is impossible to prove that an object is genuine, but scientiic analysis is the key to determining whether a piece is restored, a pastiche, enhanced or an outright forgery. References ANDREWS, C.A.R., 1996. Ancient Egyptian Jewellery. London: British Museum Publications. BAUM, J., 1937. La Sculpture Figurale en Europe à l’Epoque Mérovingienne. Paris, Éditions d’art et d’histoire. Forgeries and public collections COLLIGNON, M., 1899. Tiare en or oferte par la ville d’Olbia au roi Säitapharnès. Monuments Piot 6: 5-57. CRADDOCK, P.T., 2009. Scientiic Investigation of Copies, Fakes and Forgeries. Oxford, Butterworth Heinemann. EUGSTER, O., PERNICKA, E., BRAUNS, M., SHUKOLYUKOV, A., OLIVE, V. and ROELLIN, S., 2009. Helium, uranium and thorium analyses of ancient and modern gold objects: estimates of their time of manufacturing. ArcheoSciences 33: 53-61. JONES, M., 1990. Fake? he art of deception. Exhibition catalogue. London, British Museum Press. LILYQUIST, C., 2003. he tomb of three foreign wives of Tuthmosis III. New York, Metropolitan Museum of Art: 299-300. 333 MEEKS, N.D., 2007. Un collier grec en or ou le pastiche poussé à sa perfection, in F. Gaultier, C. Metzger (eds.), Les bijoux de la collection campana. De l’antique au pastiche. Rencontres de l’Ėcole du Louvre 21. Paris, École du Louvre, 127-144. MEEKS, N.D and CRADDOCK, P.T., 1991. Detection of cadmium in gold/silver alloys. Archaeometry 33. DE PRADENNE, A.V., 1932. Les Fraudes en Archéologie Préhistorique. Paris, Émile Nourry, 519-573. SMITH, R.A., 1928. he cruciixion on a Frankish buckle. British Museum Quarterly 3: 50-51. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 329-333 Micro rayures et signes d’usure : authentification d’orfèvrerie archéologique he analysis of micro-scratches and trace-wear in the authentication of archaeological gold Edilberto Formigli* Résumé : Les traces laissées par les instruments anciens et modernes de travail ont, depuis plusieurs années, fait l’objet d’études avec succès. Aux diférentes méthodes archéométriques utilisées pour l’identiication de faux, ainsi que dans le domaine d’authentiication d’orfèvrerie archéologique, nous proposons d’ajouter une technique basée sur l’observation de l’objet par microscopie optique ou électronique. Cette méthode repose sur l’étude de la surface de la tôle, pour identiier les stratiications des traces correspondant soit aux anciennes opérations du travail de l’orfèvre, soit au contact et au frottement au cours du temps de cette surface avec des matériaux extérieurs, qui viennent se superposer aux traces initiales. En reconnaissant la séquence chronologique des superpositions des marques, il est parfois possible de reconstruire l’histoire de l’objet et d’avancer une hypothèse sur son ancienneté. Les traces d’usure visibles sur la surface des objets d’orfèvrerie fournissent également des informations fondamentales sur la durée de vie de ces objets et sur leur utilisation. Abstract: he traces left by ancient instruments and by modern ones have been successfully studied for many years. In the present paper, we propose to add to the diferent archaeometric methods employed for the identiication of fakes a supplementary technique involving the observation of the object by optical or electron microscopy. his technique is based on the study of metallic surfaces in order to identify the stratiications of the marks related to the ancient stages of construction related to the work of the original goldsmiths and distinguish them from those which were added later to the initial marks. he latter are related to the contact and friction of the surface with external materials in the course of time. By recognising the chronological sequence of the marks’ superposition, it is at times possible to reconstruct the history of the object and to put forth hypotheses regarding its age. he visible wear marks on the surface of goldwork also provide fundamental information about the lifetime and use of the objects. Mots-clés : orfèvrerie, archéologique, faux, authentiication, micro-rayures, traces d’usure. Keywords: jewellery, archaeology, fakes, authentication, micro-scratches, use-wear. 1. INTRODUCTION L’étude scientiique au moyen de sections métallographiques qui mettent en évidence la structure cristalline interne de l’or, révèle le traitement subit par le métal pendant la production de l’objet. Il est alors possible de reconnaître, par exemple, la situation inaltérée de fusion, le degré et l’orientation des phénomènes de compression subis pendant le martelage, ou alors de recuit par chaufe (Scott, 1991). Les microanalyses chimiques en surface, ou encore mieux sur une section, au-delà de fournir des indications sur la composition de l’alliage d’or, peuvent permettre l’identiication de diférentes techniques de soudure (Guerra, 2006 ; Guerra, 2007 ; Ferro et al., 2008). *Str. Com della Miniera 4, Italy-53016 Murlo (Siena). (edilbertoformigli@gmail.com) (www.anteamurlo.it) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 335-339 336 Enin, l’examen radiographique peut aussi révéler des situations singulières à l’intérieur de l’objet et contribuer ainsi à reconstruire ses diférentes phases de production (Formigli, 1992 ; Guerra, 2006). Néanmoins, la simple observation de la surface, ce qui ne nécessite ni de prélèvement ni d’autre intervention invasive, peut à elle seule fournir tout une série d’informations utiles (Fig. 1). L’observation sous loupe binoculaire à divers grossissements et au microscope électronique à balayage (Formigli et Heilmeyer, 1993 ; Ferro, 2006 ; Meeks, 2007) de la surface des pièces en or, peut être une aide primordiale à la reconstitution de leur histoire et par conséquent les informations récoltées peuvent aussi être utilisées pour diagnostiquer leur authenticité. En particulier, l’étude microscopique des surfaces plus exposées au contact des éléments extérieurs contribue à la compréhension de la séquence des évènements mécaniques qui se suivent dans le temps, laissant des traces, même minimes et invisibles à l’œil nu. Nous ne considérons pas ici les questions de la détérioration de la surface des alliages anciens en or causée par les phénomènes d’enrichissement de surface ni la corrosion sélective, phénomènes déjà étudiés et publiés par diférents auteurs (Scott, 1983), mais seulement les évènements purement mécaniques qui laissent des traces sur les surfaces. Il s’agit ainsi de reconnaître en premier lieu les traces initiales du travail artisanal de l’or, ensuite celles fortuites et involontaires qui se superposent comme des cicatrices plus ou moins profondes sur la surface pendant la vie de l’objet et, enin, celles moins fortuites, mais pour nous riches d’in- Figure 1 : (See colour plate) Observation de la surface de la Phiale d’Achyris sous loupe binoculaire. Figure 1: (Voir planche couleur) Observation of the surface of the Phiale of Achrys under a binocular. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 335-339 Edilberto FORMIGLI formations, qui sont dues au port et utilisation prolongés de l’objet selon sa fonction, par exemple de parure (collier, bracelet, bague, etc.), ou ustensiles ou instruments rituels (ibules, récipients, etc.). 2. LES MICRO-RAYURES ET LES IMPRESSIONS DE LA PHASE DE PRODUCTION Déjà pendant les premières phases de production, les surfaces d’or subissent des micro-rayures : préparation de la lame, traitement par compression et micro-polissage de la surface. La technique ancienne pour réduire un in lingot initial à une épaisseur extrêmement ine était le martelage alterné de recuit. Pendant la dernière phase de préparation des feuilles très ines, une technique qui consiste à étirer l’objet entre des surfaces de peau, dont les traces en négatif peuvent encore parfois être relevées sur certains bijoux anciens, était aussi utilisée. Les traces du martelage d’origine sont rarement visibles à cause de leurs grandes dimensions par rapport aux dimensions des parties des lames utilisées dans les bijoux et à cause des opérations successives de travail des lames. Parmi ces opérations, notons deux très répandues dans l’antiquité : le polissage et le brunissage. Plus claires et individualisées, les traces du marteau sont reconnaissables sur les lames de plus grande épaisseur (environ 1-3 mm) qui ont subi une modiication successive pour obtenir une forme recourbée, comme par exemple celle d’une patère. Ces marques sont parfois reconnaissables au niveau macroscopique, avec une simple loupe binoculaire. Il est alors possible d’observer non seulement les coups de marteau qui, du fait de leur surface non totalement lisse, ont laissé leurs caractéristiques imprimées en négatif sur les objets, mais aussi de l’autre côté de la lame les marques du support du métal (enclume) sur laquelle la lame a été martelée (Formigli, 1985). L’instrument moderne utilisé pour aplatir des lames est constitué de deux cylindres en acier voisins, dont les distances réglables déterminent l’épaisseur de la lame que l’on a forcé à passer à travers. Comme la surface des cylindres ne peut pas être parfaitement lisse, l’impression de leurs défauts, même minimes, est identiiable sur les lames fausses agrémentées d’un fond de ines rayures en relief régulièrement parallèles. L’utilisation du microscope électronique à balayage avec un gros grossissement peut être indispensable pour l’identiication de ces traces, même dans le cas de cylindres extrêmement lisses. Les traces du travail initial d’aplatissement des lames soit pour les faux soit pour les objets originaux, sont celles de Micro rayures et signes d’usure : authentiication d’orfèvrerie archéologique 337 fond, auxquels se superposent toutes les autres traces qui se succèdent. 3. LES RAYURES OCCASIONNELLES Pendant la vie de l’objet, aux traces de travail viennent s’additionner, par superposition, d’autres rayures plus au moins profondes dont les causes sont variées : la micro-abrasion pendant l’enfouissement des matériaux environnants déplacés par des phénomènes dus à l’environnement, les évènements successifs de récupération de la terre, les opérations de polissage pendant la restauration et, enin, les manipulations (normalement plus nombreuses que prévu) par les individus qui ont pour diverses raisons eu en main les objets (propriétaires, savants, photographes, experts, etc.) chacun laissant toujours, involontairement, quelques traces. Chaque fois que, pendant sa vie, l’objet en or est touché, placé sur une substance ou frotté contre un matériau, les traces de ces évènements restent marquées sur sa surface. Ces marques apparaissent comme des rayures plus ou moins larges, longues et profondes, simples ou multiples ; des rayures peuvent avoir un fond pointu, plat ou ondulé. Ainsi, en résumé, toutes les rayures se caractérisent par un grand nombre de traits individuels distinctifs. Quand deux rayures identiques sont observées alors leur cause et origine sont les mêmes. Toutes ces traces observées sous loupe binoculaire, ou encore mieux, au microscope électronique à balayage, forment un réticule de lignes qui se croisent et se superposent. Chaque rayure, qui à un grossissement suisant apparaît comme une ligne presque droite, possède une certaine profondeur et une certaine largeur ainsi qu’une certaine orientation. Entre deux ou plusieurs rayures qui se croisent (et qui évidemment ne peuvent pas être exactement contemporaines), il est possible de reconnaître la séquence chronologique se basant sur l’empreinte présente dans le point de croisement (Fig. 2). Si les marques d’une rayure, causées par le transport de matériaux par l’outil ou matériau écorchant, couvrent l’espace vide, c’est-à-dire la disparition de la rayure croisée, cela veut dire qu’elle succède à l’autre. Si une troisième rayure “couvre” la deuxième, cela signiie que celle-ci succède à la deuxième, et ainsi de suite. Si deux ou plusieurs rayures sont orientées dans la même direction, il est probable qu’elles soient réalisées simultanément. En général, plus longue est la vie et le temps d’utilisation de l’objet, plus haut est le facteur chaos, c’est-à-dire la variété des trois paramètres et la concentration des tracés. Figure 2 : (See colour plate) Superposition de micro-rayures sur la Phiale d’Achyris : image au microscope électronique à balayage. Figure 2: (Voir planche couleur) Superposition of the micro-scratches on the Phiale of Achrys: images under the scanning electron microscope. Un autre critère valide pour évaluer l’état de l’objet est l’étude micro morphologique des rayures. Les tracés anciens tendent pendant les longues périodes à être réabsorbés par la matrice. Par analogie avec les tissus humains, nous pouvons imaginer assister à un phénomène de « cicatrisation » d’une blessure pour laquelle les lacérations plus externes s’étirent et les reliefs s’adoucissent. Ce phénomène est causé par une recristallisation qui ne se révèle qu’après une période très longue. Des tracés anciens qui ont subi une telle transformation, sont identiiables par microscopie électronique par son opacité ; les tracés récents, par opposition, possèdent de bordures afûtées, identiiables par un contraste plus net. En efet, une conirmation est fournie par le fait que tracés reconnus comme anciens ne se superposent jamais à ceux que nous avons, basés sur notre hypothèse, supposés modernes. L’ensemble du complexe réseau de micro-rayures présent sur la fameuse Phiale d’Achyris (Fig. 1 et 2), étudiée d’une façon approfondie à diférents grossissements sur ses diverses zones et résultant en une pièce authentique par ce type de recherche, montre la présence d’un bruit de fond tant orienté (traces de travail) que chaotique de type ancien auquel se superposent abondamment des traces de type récent, concentrés de façon particulière dans les zones plus extérieures de la phiale, accessibles à la manipulation. 4. LES TRACES D’USURE PAR UTILISATION Selon l’utilisation faite d’un objet précieux en or, se forment au long du temps des polissages et aplatissements dans les zones de frottement de ses diférentes parties. Nous ArcheoSciences, revue d’archéométrie, 33, 2009, p. 335-339 338 pouvons citer, par exemple, les accroches avec les anneaux d’une chaîne ou encore des frottements sur la surface la plus exposée au contact avec des matériaux externes, par exemple sur un pendentif qui, pendant son utilisation, frotte sur la peau ou sur le tissu des vêtements. L’observation précise de telles traces fournit des informations sur le mode d’utilisation de l’objet, sur la période de temps d’utilisation, sur la présence de réparations ou de remplacements anciens de parties d’objets ou encore sur la présence de restaurations modernes. En déinitive, il s’agit aussi d’informations qui considèrent l’histoire de l’objet mais également de son état. Un Kardiophylax (bulle pectorale) provenant de Cuma montre distinctement toute une série de détails d’usure par utilisation (Fig. 3). Si la direction de l’élargissement est vers l’extérieur d’une cavité alors l’objet se dit porté en longueur, adhérent à la poitrine aussi au moyen d’un il qui passait derrière le revers, et que le premier contact avec la lame était fait avec l’anneau d’un il métallique rond lequel était doublé sur la partie extérieure. L’usure par utilisation d’une zone en relief où passait la décoration correspond à un objet porté sous les vêtements. Généralement, toutes les traces d’usure confortent l’authenticité de la pièce. Pour identiier les faux modernes, l’étude des traces d’usure est évidement très importante. Même pour des faussaires expérimentés, il n’est pas aisé de reproduire d’une façon crédible les traces d’usure par utilisation, conséquence d’un long port d’un objet. Des cas de fausses traces d’usure par utilisations apportées par des faussaires ont été observés, traces facilement identiiables car localisées dans des endroits impossibles, comme par exemple l’aplatissement d’un il en or en forme de crochet non sur la partie interne mais sur la partie externe. De la même façon, l’outil utilisé, dans ce cas une lime, a laissé des traces possédant des proils aiguisés qui ne sont pas conformes à la typologie de l’usure par utilisation qui produit toujours des surfaces très lisses. Un objet ancien peut par une raison ou une autre ne plus avoir été utilisé quand, par exemple, a été produit volontairement à but funéraire, et ainsi malgré le fait d’être ancien il ne portera pas de traces d’usure. Cette information est déjà de grande importance pour l’archéologue qui étudie l’objet. En réalité, il s’agit de cas très rares car souvent un examen approfondi des pièces mène à exclure cette éventualité. 5. CONCLUSIONS Pour obtenir des informations qui apportent des renseignements sur l’âge d’un objet d’orfèvrerie et sur l’histoire des événements expérimentés par l’objet, il est essentiel d’obArcheoSciences, revue d’archéométrie, 33, 2009, p. 335-339 Edilberto FORMIGLI Figure 3 : (See colour plate) Traces d’usure par utilisation du Kardiophilax de Cuma. Figure 3: (Voir planche couleur) Use-wear traces produced by the utilisation of the Kardiophilax from Cuma. server les marques qui se sont superposées sur sa surface au cours du temps. Ces marques peuvent être provoquées par les actions suivantes : 1) Travail artisanal initial 2) Rayures occasionnées pendant la vie de l’objet 3) Traces d’usure par utilisation 4) Micro-abrasion par les matériaux environnants dans la terre d’enfouissement 3) Evènements pendant la fouille archéologique 4) Traces laissées par la restauration 5) Traces de manipulations successives Les caractéristiques qui peuvent indiquer un âge ancien pour un objet et ainsi conirmer son authenticité sont les suivantes : 1) Superposition de marques selon un ordre logique et chronologique 2) Caractère aléatoire des marques 3) Grand nombre de traces superposées 3) Présence de traces plausibles d’utilisation Micro rayures et signes d’usure : authentiication d’orfèvrerie archéologique Bibliographie FERRO, D. FORMIGLI, E., PACINI, A. and TOSSINI, D., 2008. La saldatura nell’oreiceria antica, Consiglio Nazionale delle Ricerche, Roma. FORMIGLI, E., 1992. Indagini archeometriche sull’autenticità della ibula prenestina, Mitteilungen, des deutschen archaeologischen Instituts 99: 330-343. FORMIGLI, E., 1985. Tecniche dell’oreiceria etrusca e romana, originali e falsiicazioni. Sansoni Editore, Firenze. FORMIGLI, E. and HEILMEYER W.D., 1993. Einige Faelschungen antiken Goldschmucks im 19. Jahrhundert. Archäologischer Anzeiger 3: 299-332. GUERRA, M.F., 2006. Etruscan gold Jewellery Pastiches of the Campana’s Collection revealed by scientific Analysis, in 339 M. Cavallini, G.E. Gigante (ed.), De Re Metallica, L’Erma di Bretschneider, Roma, 103-128. GUERRA, M.F., 2007. Examen et analyse élémentaire de quelques bijoux de la collection Campana, in Les bijoux de la collection Campana. F. Gaultier, C. Metzger (ed.), Paris, École du Louvre, 145-177. MEEKS, N., 2007. Un collier grec en or ou le pastiche puossé à sa perfection, in Les bijoux de la collection Campana, F. Gaultier, C. Metzger (ed.), Paris, École du Louvre, p. 127-144. SCOTT, D., 1983. he Deterioration of Gold Alloys and some Aspects of their Conservation, Studies in Conservation 28: 194203. SCOTT, D., 1991. Metallography and Microstructure of Ancient and Historic Metals, Marina del Rey (Cal.), he J. Paul Getty Trust, Los Angeles. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 335-339 A new approach to the authentication of goldwork using combined Scanning Electron Microscopy and External-beam PIXE Une nouvelle approche pour l’authentification des objets en or en couplant la microscopie électronique à balayage et le PIXE en faisceau extrait Olivier Bobin* and Hervé Guegan** Abstract: he study of goldwork encompasses several large ields of investigations. Scientiic analyses can respond to questions concerning production techniques, provenance and preservation. However, what about authentication? his type of works of art generally has very high prices, leading to an increase in the number of forgeries. Today, fakes can imitate artworks produced by most civilisations and with diferent origins. Moreover, most of the objects we analyse cannot be submitted to destructive sampling. hat is the reason why we employ external-beam PIXE for the analysis of such objects, a method able to ofer reliable and non-intrusive scientiic investigations. Our methodology consists of two steps, which, in their combination, provide information about the chronological compatibility of the objects with their presumed period. First, an optical and electronic microscope investigation allows us to identify tool marks, weathering features, to reveal traditional or fraudulent surface treatments, and to characterise supericial deposits. his irst approach is able to identify bad quality fakes. However, these analyses are generally insuicient to establish the authentication of the objects. hat is why we also perform external-beam PIXE analyses on goldwork. hus, the qualitative and quantitative studies of trace elements are able to provide insights into metallurgical methods and the ineness of alloys. his paper describes our methodology and presents diferent examples of its application, including the discussion of speciic problems encountered in the analysis of this type of objects. Résumé : Les objets en or sont étudiés par des approches très diverses, et parmi celles-ci les analyses scientiiques répondent généralement à des problématiques concernant la recherche des techniques de fabrication ou l’étude de provenance et de conservation. Mais, qu’en est-il de l’authentiication ? Ce type d’objets d’art étant en efet d’une valeur marchande élevée, ils sont de plus en plus souvent sujets à des contrefaçons, qui imitent aujourd’hui la plupart des styles et des civilisations. Les investigations scientiiques doivent être absolument non destructive et la technique PIXE en faisceau extrait est alors idéale pour ce type d’expertise. Nous avons ainsi développé une méthodologie qui nous fournira un faisceau d’indices qui sera compatible ou pas avec l’ancienneté présumée de l’objet. En premier lieu, l’examen par microscopie optique et électronique permet d’identiier les traces d’outils, les igures de corrosion, de révéler si l’objet a subi des traitements de surface anciens ou modernes et de caractériser la nature des dépôts supericiels. Ce premier niveau d’investigation permet d’écarter les faux lagrants ou de mauvaise qualité. Toutefois, cette étude est généralement insuisante pour établir l’ancienneté d’un objet en or. C’est pourquoi, nous réalisons en complément une étude par PIXE en faisceau extrait qui détermine la composition en éléments traces, eux-mêmes révélateurs des méthodes métallurgiques et de la pureté des alliages. Ce texte décrit cette méthodologie et présente diférents exemples d’application en précisant les problèmes auxquels nous avons été confrontés sur les diférents types d’objets rencontrés. Keywords: gold, analysis, Scanning Electron Microscope, Particle Induced X-ray Emission, external beam, trace elements, authentication, fakes. Mots-clés : or, analyse, Microscope Électronique à Balayage, PIXE faisceau extrait, éléments traces, authentiication, faux. * CIRAM, Chem’Innov, 16 avenue Pey Berland 33607 Pessac cedex, France. (olivier.bobin@ciram-art.com) ** ARCANE – CENBG (UMR 5797, CNRS – IN2P3 / Université Bordeaux 1) – rue du Solarium BP120, 33 175 Gradignan, France. (arcane@ cenbg.in2p3.fr) rec. Aug. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 341-347 342 1. INTRODUCTION he analysis of goldwork interests archaeologists, art historians, or museum curators. he aims of such investigations are to study the style, the iconography, and the techniques of manufacture, to establish a hypothetical period and provenance of these objects, and to develop a strategy of conservation. To these purposes, scientists can contribute useful information regarding the composition of the alloy, the analysis of deposits, and the weathering degree. Most of the studies on gold artefacts aim to deine the origins, trade and provenance of the ores based on the trace elements composition (Guerra and Calligaro, 2003; 2004; Bugoi et al., 2003; Demortier et al., 1999; Ontalba Salamanca et al., 1998). Generally, few methodological studies exist about the authentication of gold (Craddock, 2009; Guerra, 2008). First of all, gold objects are very resistant to long time corrosion. hus, we cannot establish our conclusions on the weathering degree of the gold alloy or the nature of the corrosion products, as for copper alloys (Craddock, 2009; Robbiola and Portier, 2006). hermoluminescence tests, performed on casting cores, can provide interesting chronological information for bronze objects, for example (Craddock, 2009). However, goldwork does not typically contain residual cast cores. In addition, gold forgeries have become progressively lawless, as the falsiiers became aware of the scientiic tools used in the authentication of works of art. It is quite easy to use ancient gold shreds to re-create a new and fake artefact, following an ancient style, and with a typical ancient trace elements composition. In this case, a single scientiic approach, which only uses trace elements to distinguish fakes from original ancient goldwork, is no longer suicient to discriminate all the forgeries. In order to improve the authentication of gold objects, this paper presents a combined approach, including as the irst step an investigation of the surface by scanning electron microscopy (SEM-EDX) to observe tool marks, supericial deposits, the microstructure of the alloy, and eventual weathering features. As the second step, a quantitative analysis of the alloy by external-beam PIXE is employed, in order to deine its trace element composition. 2. METHODS he study of the surface and the chemical composition of the deposits were carried out with an energy-dispersive X-ray spectrometer (EDS), using a 20 keV electron beam, coupled with a scanning electron microscope (SEM) Jeol JSM 840. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 341-347 Olivier BOBIN, Hervé GUÉGAN hese analyses were performed on the surface without pretreatment or metallization. We employed both back-scattered electron (BSE) and secondary electron (SE) images. External beam PIXE (Particle Induced X-Ray Emission) is a non-destructive method of analysis which allows determining the concentration of major, minor and trace elements. For this purpose, we used a 3MeV H+ ion beam produced by the singletron accelerator of CENBG (Centre d’Etudes Nucléaires de Bordeaux Gradignan). Due to the limits of the accelerator, this energy optimizes the ionisation cross-section for interesting elements, such as Sn, Sb, Cd, Ru, and Rh. In front of the dedicated external beam-line, artefacts can be analysed without sampling or limitations in terms of size. Two detectors are used in the experimental setup, and allow the measurement of emitted X-rays under the interaction with the incident ion beam, which constitute a spot size of 1 mm in diameter on the sample. 3. RESULTS SEM-EDS investigations he micrographs of the surface provide information on the manufacturing techniques employed. We also searched for tools marks, such as those as a result of hammering, burnishing or polishing. Finally, we also investigated the decoration methods and analysed the supericial deposits. he small objects (less than 5 cm) can be placed in the specimen chamber of the SEM. his allowed us to directly determine their composition (major and minor elements) and observe their microstructure. For the bigger objects, we worked on surface replications (cellulose acetate ilm), which reveal the ingerprint of the surface and remove supericial deposits (Funga and Sanipelli, 1982). his irst step allows identifying the noticeable fakes, which show fundamental anomalies, such as modern tool marks or traces of acid attack. Figure 1 shows a surface replication of a gold statuette presumed to originate from the New Kingdow of Egypt (15801085 BC). On the surface, we can observe very thin and unidirectional ridges, which cannot correspond to restoration or cleaning phases. hey correspond to burnishing tool marks. his polishing technique was traditionally performed with a burnisher, a type of little hammer with a perfectly polished head (Ogden, 1993; Arminjon and Bilimof, 1998; Scott, 1991). Nowadays, the burnishing is carried out with a mechanical burnisher using thousands of rotating micro-balls. he thinness and the regularity of the marks observed on the surface of this object are clearly a result of this modern technique. A new approach to the authentication of goldwork using combined Scanning Electron Microscopy… 343 association of iron and chloride, and the microstructure of these deposits indicate that they correspond to residues of an acidic solution, such as ferric chloride (Perego, 2005). he contents of iron and chloride are too high to correspond to natural clay or ‘earth’. his indicates that the object underwent a surface treatment to modify the surface aspect of the object and to simulate mineral deposits and/or presence of copper corrosion products. We can see that SEM surface analyses represent a powerful tool for discriminating the obvious fakes, which show formal traces of modernity. hese investigations are necessary, but not suicient, to establish an object’s authenticity, because it is easier to attest that an object is a forgery than to establish its authenticity (Robbiola and Portier, 2006). External-beam PIXE Figure 1: SEM SE image of a surface replication from a gold statuette. We observe thin and unidirectional ridges, which correspond to the use of modern tools and techniques. Figure 1 : Image MEB de la reproduction de la surface d’une statuette en or. Les stries ines et unidirectionnelles correspondent à l’utilisation d’outils modernes. Following the SEM-EDS study, we also detected silver chloride micro-particles, which are not mixed with the mineral deposits. In addition, we did not detect any silver sulphide micro-particles, which are usually also found in naturally corroded gold-silver alloys (Gusmano et al., 2004). As a consequence, it can be inferred that the silver chloride micro-particles originate from a deliberate chemical attack, carried out in order to simulate the natural weathering of the material. Aqua regia treatment leads to a partial dissolving of the metal and to the formation of numerous silver chloride micro-particles. Finally, we observed supericial cupules, which are also characteristic of an acid attack. his result is consistent with the large quantity of chlorine contained in the deposits. he presence, on the entire object, of modern tool marks and characteristics of chemical attack is inconsistent with the presumed period of this piece. It is thus easily understandable that, in this case, a complementary PIXE analysis is not necessary. he second example concerns a gold earring from China, presumed to be from the Shang Dynasty (11th-10th centuries BC). he SEM observation of the supericial deposits shows an amorphous to non-crystalline material, which is composed of iron, chloride, calcium and silicated phases. he he external-beam PIXE developed at CENBG is one of the ive beam-lines of the AIFIRA platform. Four of them are dedicated to macro-, micro-, and nano-analysis by nuclear techniques. AIFIRA is a brand new facility built in 2006 using the 40 years old experience of the CENBG in Ion Beam Analysis. While some other external-beam PIXE facilities exist in the world, for the present study we have developed an experimental setup dedicated to the analysis of gold items. For this purpose, two types of information are interesting: the composition of the gold alloy and the trace elements. However, it is not possible to obtain both at the same time, with the same detector. In our experimental setup, we simultaneously used two Si(Li) detectors with diferent shields in order to optimize the diferent types of investigated information. We will describe below the obtained spectra and the errors in interpretation which could occur if we did not pay attention. he quantitative data corresponding to Figures 2 and 3 are not included in Table 1. he irst spectrum reported is obtained from a gold necklace presumed to date from 1500 BC, with the detector equipped with a 315 µm carbon ilter before the silicon crystal (Fig. 2). his setup was used to stop the backscattered proton beam and the low energy X-rays (especially M-lines from Au and L-lines from Ag, Sn, Cd, etc.). his spectrum allowed the determination of the Au/Ag/Cu composition (98.1/0.7/0.9%), and the detection of trace elements Fe (620 ppm) and Ni (114 ppm), because of their large X-ray emission cross-section. In addition, we had to move the detector away from the target to prevent pile-up efects, due to the fact that the counting rate was so high as compared to the frequency of the multi-channel analyser. Nevertheless, we observed on the spectrum four ArcheoSciences, revue d’archéométrie, 33, 2009, p. 341-347 344 Olivier BOBIN, Hervé GUÉGAN Figure 2: PIXE spectrum of a gold necklace collected with a 315 µm carbon ilter. We identiied the composition of the alloy as 98.1% Au, 0.7% Ag, and 0.9% Cu, and we detected Fe and Ni as trace elements. Figure 2 : Spectre PIXE d’un collier en or dont l’acquisition est réalisée avec un iltre de 315µm de carbone. La composition de l’alliage a été déinie (Au 98.1, Ag 0.7, Cu 0.9) et le Fe et le Ni ont été détectés au niveau d’éléments trace. Figure 3: PIXE spectrum of a gold necklace collected with a 75 µm copper ilter. his coniguration allows detecting the presence of tin (1762 ppm) and antimony (123 ppm) in the alloy as trace elements. Figure 3 : Spectre PIXE d’un collier en or dont l’acquisition est réalisée avec un iltre de 75 µm de cuivre. Cette coniguration permet de déterminer l’étain (1762 ppm) et l’antimoine (123 ppm), éléments présents dans l’alliage à l’état de trace. little sum peaks due to this phenomenon, which prevent the detection of important elements in the process of authentication, like Pd and Cd. Furthermore, the little solid angle in this coniguration is not suitable for the detection of other trace elements, like Sn and Sb, due to their low X-ray emission cross-section, and the only interesting trace element which could be detected is Pb, due to its Lγ1 line at 14.762 KeV with a limit of detection (LOD) of 300 ppm. Detection of other trace elements, such as Hg or Pt, is not possible due to the high level of Au L-rays. At this stage of the study, we were only able to determine the composition of the alloy. Even if this information is important, it is not suicient for providing reliable chronological information about the object. he second experimental setup was thus necessary in order to obtain data on other trace elements. In this coniguration, an absorbing copper foil (75 µm thick) is placed in front of the detector. he reason we used copper is because its edge of absorption will neutralize especially the energy of the gold L-rays. his setup prevents the pile-up and allows us to move the detector closer in order to improve the LOD down to 60 ppm for the elements Ru, Rh, Pd, Cd, Sn, and Sb. On the resulting spectrum (Fig. 3), we can observe the energy range where the L-lines of gold are very low in intensity, the lack of pile-up in the region under consideration, and the presence of trace elements Sn (1762 ppm) and Sb (123 ppm). On the same spectrum, we can also observe the peak of xenon, which is present in the air on the path of the incident proton beams. his result indicates the fact that the reining of this gold alloy is quite poor and consistent with ancient processes. Nevertheless, an old alloy could also be used by a forger to manufacture a modern item. hus, we have to corroborate these analyses with an investigation of the surface of the alloy by SEM in order to identify eventual corrosion microstructures and surface treatments. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 341-347 Combined methods: SEM and PIXE Example 1 he observation of the surface of a European gold ring presumed to date from the 5th century AD indicates that the decorations have been performed with a hand-tool (Arminjon and Bilimof, 1998). he surface of the metal is smooth with irregular microstructures. he edges of the engraved designs are not raised. However, we did not observe clear weathering features, such as micro-pitting or corrosion of grain boundaries (Fig. 4). From these results, it is diicult to provide formal information about the chronological compatibility of this object with its presumed antiquity. 345 A new approach to the authentication of goldwork using combined Scanning Electron Microscopy… antimony (Bugoi et al., 2003; Craddock, 2009; Demortier et al., 1999; Guerra, 2008; Guerra and Calligaro, 2003; 2004; Ontalba Salamanca et al., 1998; Ott and Schindler, 2001). he absence of a signiicant set of trace elements in this alloy is abnormal. his aspect suggests that the metallurgical methods employed in the making of this object do not correspond to ancient traditional ones. Its composition indicates that modern extraction and reining (thermal or electrolytic) were used (19th or 20th century AD). Example 2 Figure 4: SEM SE image of a gold ring. he surface and the edges of the engraved designs are smooth and not raised. Figure 4 : Image MEB d’un anneau d’or. La surface et les bords des dessins gravés sont lisses et non saillants. In such a case, PIXE analysis of the trace elements composition of the alloy is necessary. he analysis showed a very pure gold (99.6 to 99.9 wt% Au), with silver, copper, iron, nickel, calcium and chromium as trace elements (Table 1). he last four elements probably originate from the contribution of a mineral deposit. hus, silver and copper would be the only constituting trace elements of this metal. he ancient gold alloys contained many trace elements, such as lead, tin, platinum, palladium, zinc, bismuth, and Object Gold ring (presumed from 5th century AD) Gold dagger (presumed from 4th-3rd centuries BC) Gold belt (presumed from 6th - 3rd centuries BC) Point 1 2 3 1 2 3 1 2 3 4 wt % Au 99.9 99.9 99.6 86.1 84.9 83.6 97.8 98.1 97.8 96.0 wt % or ppm Ag Cu 1000 270 716 193 815 580 8.0% 5.0% 9.0% 6.0% 8.6% 6.0% 0.99% 6780 1.1% 4610 0.96% 6630 3.2% 7730 he analysis of a Middle East gold dagger presumed to date from the 4th-3rd century BC required a combination of the SEM and PIXE approaches to verify its authenticity. he material was a gold-silver-copper alloy. he concentration of the gold was in the range of 84 to 86%. It corresponds to 20.5 carat gold. he three analysed areas contain iron, nickel and tin as trace elements (Table 1). he presence of tin and nickel should indicate an ancient metallurgical method. However, their concentrations are relatively low, and we did not detect any other trace elements, such as lead, antimony or platinum, for example. Yet, we also did not observe any formal evidence of modernity, such as the presence of aluminium by SEM (more than 0.1 wt%) or of cadmium by PIXE, for example (Scott, 1991; Craddock, 2009). From the PIXE results, it was diicult to draw a deinite conclusion regarding the authenticity of this dagger. he SEM micrographs of the gold indicate it has a smooth surface with irregular features and numerous pits (Fig. 5). It presents a rounded microstructure, and the proiles of depression are not sharp. We also observed an important micro-porosity. he microstructure of the surfaces appears to correspond to prolonged and natural weathering pro- Fe 105 100 980 7723 735 16000 1140 1310 1440 520 Cr nd nd 230 nd nd nd nd nd nd nd Ca nd nd 3400 nd nd nd 3614 1450 6630 nd Ni 93 125 110 152 238 216 94 105 85 130 Zn nd nd nd nd nd nd nd nd nd nd ppm Ru nd nd nd nd nd nd nd nd nd nd Rh nd nd nd nd nd nd nd nd nd nd Cd nd nd nd nd nd nd nd nd nd nd Sn nd nd nd 100 200 230 250 2700 1430 485 Sb nd nd nd nd nd nd 30 134 60 60 Pb nd nd nd nd nd nd nd nd nd nd Table 1: Compositions of the alloys of diferent objects. he major and minor elements are expressed in weight percent and trace elements in ppm. Even if the limits of detection are not presented, they are evaluated for each element and each analysed point. he non-detected elements or those whose for which the detected levels are less than or equivalent to the detection limit are indicated as ‘nd’. Tableau 1 : Composition des alliages des diférents objets. Les éléments majeurs et mineurs sont exprimés en % et les éléments traces en ppm. Malgré l’absence de limites de détection, ces valeurs ont été déterminées pour chaque élément et pour chaque analyse. Les éléments non détectés ou dont la concentration est égale ou inférieure à la limite de détection sont indiqués « nd » ArcheoSciences, revue d’archéométrie, 33, 2009, p. 341-347 346 Olivier BOBIN, Hervé GUÉGAN Furthermore, the alloy also contains signiicant concentrations of tin, antimony and nickel as trace elements (Table 1). heir presence is in good agreement with ancient extraction techniques that were not able to remove these elements. hus, once again, the combined study of both the surface microstructure and the composition of the alloy and the solders indicate that the object corresponds to ancient and traditional goldwork practices. 4. DISCUSSION AND CONCLUSION Figure 5: SEM SE image of a gold dagger. he surface of the metal has a smooth surface with irregular features and numerous pits. We can also observe a rounded microstructure. Figure 5 : Image MEB d’un couteau en or. La surface du métal est lisse avec des traits irréguliers et nombreux petits trous. Une microstructure arrondie peut aussi être observée. cesses afecting gold alloys (Ontalba Salamanca et al., 1998; Ott and Schindler, 2001; Gusmano et al., 2004; Craddock, 2009). hus, the results obtained from the combined methodological approach are in good agreement with the presumed period of the object. Example 3 he SEM examination of an East Central Asian gold belt presumed to date from the 6th-3rd centuries BC indicates that supericial deposits are composed of a mix of silicated mineral phases, such as clay, and calcium carbonate. hey also contain numerous microparticles of silver salts, which should result from the natural weathering of the gold alloy. he object has a smooth surface with irregular features, a rounded microstructure and a supericial micro-porosity. We also did not detect any modern tool marks or indications of a chemical treatment. he solders are composed of a goldsilver-copper alloy, which is consistent with ancient techniques, and we did not detect the presence of any modern element, such as zinc or cadmium. We detected iron and calcium as trace elements, but they most likely correspond to the contribution of supericial deposits. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 341-347 his paper emphasizes that the authentication of gold objects requires the use of complementary approaches. A combination of SEM and PIXE analyses is very powerful for this purpose. he use of a single method only provides partial information regarding the antiquity of the object. SEM investigates the surface of the object, and is able to identify fashioning and polishing tool marks, weathering features, traces of acid attacks, and the composition of the burial deposits. hese investigations are limited to a surface study. hus, in most cases, the observations of the microstructure of the gold are insuicient for obtaining formal evidence about the authenticity of an object, except for the cases of obvious fakes. PIXE determines the composition of the alloy and characterises the concentration of trace elements, providing information about gold working techniques. his method is non-intrusive and ofers powerful results, which allow estimating the chronological compatibility of the objects with their presumed antiquity. However, using this method exclusively does not allow us to discriminate forgeries made with ancient gold shreds. As a matter of fact, the forgers take advantage of the new scientiic knowledge available. If we want to be ahead of them, we have to combine several methods in order to investigate diferent characteristics of the objects and to improve our results in diferent ways. hus, X-ray radiography or CT scanning can contribute complementary information about the homogeneity of the objects and the manufacturing techniques used. When sampling is possible, other types of analytical methods, such as ICP-MS (Guerra and Calligaro, 2004) can be useful to improve the limits of detection of lead and platinum, for example, and to neutralize the efect of speciic treatments, such as supericial enrichment of gold by burnishing or la mise en couleur de l’or (Grimwade, 1999; Jacobson, 2000; Bachmann et al., 1999). A new approach to the authentication of goldwork using combined Scanning Electron Microscopy… References ARMINJON, C. and BILIMOFF, M., 1998. L’art du métal, Vocabulaire technique. Paris, Momum Éditions du patrimoine. BACHMANN, H.G., BURKHARDT, A., DEHN, R. and STERN, W.B., 1999. New aspects of Celtic gold coinage production in Europe. Gold Bulletin 32(1): 24-29. BUGOI, R., COJOCARU, V., CONSTANTINESCU, B., CONSTANTIN, F., GRAMBOLE, D. and HERRMANN, F., 2003. Micro-PIXE study of gold archaeological objects. Journal of Radioanalytical and Nuclear Chemistry 257(2): 375-383. CRADDOCK, P., 2009. Scientiic investigation of copies, fakes and forgeries. Oxford, Butterworth-Heinemann. DEMORTIER, G., MORCIAUX, Y. and DOZOT, D., 1999. PIXE, XRF and GRT for the global investigation of ancient gold artefacts. Nuclear Instruments and Methods in Physics Research B 150: 640-644. FUNGA, P.C. and SANIPELLI, G.G., 1982. Surface studies of feldspar dissolution using surface replication combined with electron microscopic and spectroscopic techniques. Geochimica et Cosmochimica Acta 46(4): 503-512. GRIMWADE, M., 1999. he surface enrichment of carat gold alloys – Depletion gilding, in D. Schneller (ed.), Proceedings of the 13th Santa Fe symposium on jewellery manufacturing. Albuquerque, New Mexico: Met-Chem Research, 16-23. GUERRA, M.F., 2008. An overview on the ancient goldsmith’s skill and the circulation of gold in the past: the role of X-ray based techniques. X-Ray Spectrometry 37: 317-327. GUERRA, M.F. and CALLIGARO, T., 2003. Gold cultural heritage objects: a review of studies of provenance and manufacturing 347 technologies. Measurement Science and Technology 14: 15271537. GUERRA, M.F. and CALLIGARO, T., 2004. Gold traces to trace gold. Journal of Archaeological Science 31(9): 1199-1208. GUSMANO, G., MONTANARI, R., KACIULIS, S., MONTESPERELLI, G. and DENK, R., 2004. Gold corrosion: red stains on a gold Austrian Ducat. Applied Physics A 79: 205-211. JACOBSON, D.M., 2000. Corinthian bronze and the gold of the alchemists. Gold Bulletin 33(2): 60-66. OGDEN, J., 1993. Aesthetic and technical considerations regarding the colour and texture of ancient goldwork, in S. La Niece and P. Craddock (eds.), Metal plating and patination. London, Butterworth, 185-147. ONTALBA SALAMANCA, M.A., DEMORTIER, G., FERNANDEZ GOMEZ, F., COQUAY, P., RUVALCADA-SIL, J.L. and RESPALDIZA, M.A., 1998. PIXE and SEM studies of Tartesic gold artefacts. Nuclear Instruments and Methods in Physics Research B 136138: 851-857. OTT, D. and SCHINDLER, U., 2001. Metallography of gold and gold alloys. Gold Technology 33: 6-11. PEREGO, F., 2005. Dictionnaire des matériaux du peintre. Paris, Belin. ROBBIOLA, L. and PORTIER, R., 2006. A global approach to the authentication of ancient bronzes based on the characterization of the alloy-patina-environment system. Journal of Cultural Heritage 7: 1-12. SCOTT, D.A., 1991. Metallography and microstructure of ancient and historic metals. Los Angeles, he Getty Conservation Institute. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 341-347 Copy or authentic? Analysis of a Phoenician gold ring from the National Museum of Archaeology, Valletta, Malta Copie ou authentique ? Analyse d’une bague phénicienne en or du Musée national d’archéologie, La Vallette, Malte Daniel Vella*, James Licari**, Nicholas Vella***, Sharon Sultana**** and Vanessa Ciantar**** Abstract: he present work is a study of a Phoenician inger ring from the collection of the National Museum of Archaeology, Malta. he item was irst described by the museum’s director hemistocles Zammit in 1925, and is believed to date to the 6th century BC. he ring consists of two stirrup-shaped hoops, which can be worn separately or itted together and worn as a single piece. Inscribed on the bezel surface is a seafaring vessel. Zammit described the artefact as manufactured in pure gold, quoting its mass as 9.65 g. he ring in the collection its Zammit’s description but difers signiicantly in weight. he aim of this paper is throw light on the authenticity of this ring using documentary sources and non-invasive scientiic techniques of analysis. Optical and electron microscopy allowed a thorough description of the manufacturing technique, while the material analysis was conducted via energy dispersive spectrometry. As a result of these analyses, it was concluded that the ring is not authentic, but is most probably a copy, possibly commissioned by Zammit himself. Résumé : Ce travail décrit l’étude d’une bague phénicienne appartenant à la collection du musée National d’Archéologie de Malte. Cet objet a été décrit pour la première fois en 1925 par le Directeur du musée, hemistocles Zammit, et est attribuée au VIe siècle av. J-C. La bague se compose de deux anneaux en forme d’étrier pouvant être portés séparément ou pouvant être assemblés pour être portés ensemble. Le chaton est inscrit d’un vaisseau. Zammit décrit l’objet comme étant fabriqué en or pur et indique un poids de 9,65 g. La bague de la collection correspond à cette description mais difère considérablement en ce qui concerne son poids. Le but de cet article est de jeter un jour nouveau sur l’authenticité de la bague à partir des sources documentaires et des analyses scientiiques non-destructives. Les microscopies atomique et électronique permettent une description approfondie des techniques de fabrication alors que l’analyse des matériaux a été réalisée par spectrométrie à énergie dispersive. Nous avons conclu que la bague n’est pas authentique, mais probablement une copie, peut-être commandée par Zammit. Keywords: Gimmal ring, iconography, Phoenician, gold, authentication. Mots-clés : Bague gimmal, iconographie, phénicien, or, authentiication. * Department of Metallurgy and Materials, Faculty of Engineering – University of Malta, MSD 2080. (daniel.vella@um.edu.mt) ** Objects Conservation Laboratory, Conservation Division, Heritage Malta, Bighi, Kalkara KKR 9030, Malta. (james.licar@gov.mt) *** Department of Classics and Archaeology, Faculty of Arts, University of Malta, MSD 2080, Malta. (nicholas.vella@um.edu.mt) **** National Museum of Archaeology, Heritage Malta, Republic Street, Valletta, Malta. (sharon.sultana@gov.mt), (vanessa.ciantar@gov.mt) rec. Aug. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 349-355 350 Daniel VELLA et al. 1. INTRODUCTION Historical background he National Museum of Archaeology (NMA) in Malta holds a small collection of gold artefacts, of which a ‘gimmal’ inger-ring (NMA Object ID: 866) merits particular attention. he artefact was irst described by the museum’s director Sir hemistocles Zammit in a short article published in 1925, and is thought to date to the Phoenician-Punic period (Zammit, 1925; Bonanno, 2005: 31). he ring consists of two stirrup-shaped hoops, which can be worn separately or itted together and worn as a single piece (cf. Smith, 1803). he hoops it to form an elliptical bezel of circa 20 x 10 mm, divided horizontally by twelve indentations on each side. When the complete ring is assembled, a rowing vessel appears inscribed on the bezel surface (Fig. 1). Although this inger-ring is a unique piece and no similar examples are mentioned in the literature dealing with ancient jewellery that we have consulted (e.g. Culican, 1986; Higgins, 1961; Pisano, 1974; 1987), rings with oval bezels with engraved devices set within a rim are in fact known from Punic sites in the western Mediterranean, and are believed have been produced until the 5th century BC (Boardman, 2003: 12; Quillard, 1987). Even the representation of the galley, redrawn in order to rectify some errors in Zammit’s line drawing (compare Fig. 2e with Fig. 2d), would ind parallels in ship representations from the 7th-6th centuries BC (Morrison and Williams, 1968). Figure 1: Photograph of the NMA ring, showing bezel surface under raking light conditions. Figure 1 : Photographie de la bague du NMA, montrant le chaton sous lumière rasante. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 349-355 Figure 2: (a, b) Scanned digitized positive images of two glass plate negatives believed to have been used in Zammit’s 1925 publication; (c) Zammit’s published photograph (1925: Fig. 1b) of the ring, clearly retouched to emphasize the engraved lines; (d) a mirror image of Zammit’s published line drawing (1925: Fig. 3) of the details on the bezel; (e) reconstruction drawing of the features visible on the bezel photographed in (a). Figure 2 : (a,b) Images numérisées de deux négatifs sur plaques de verre dits avoir été utilisées par Zammit dans la publication de 1925 ; (c) photographies de la bague publiées par Zammit (1925 : ig 1b) ; (d) image en miroir du dessin publié par Zammit (1925 : ig. 3) montrant des détails du chaton ; (e) dessin reconstituant ce qui est visible sur le chaton photographié en (a). Copy or authentic? Analysis of a Phoenician gold ring from the National Museum of Archaeology… In his paper, Zammit described the artefact as manufactured in ‘pure gold’, quoting the mass of the ring at 9.650 g (Zammit, 1925: 266). he ring was shown to Zammit by an undisclosed owner. From Zammit’s publication we are led to believe that the owner was reluctant to supply details about the discovery, but did indicate that the ring was passed on to him by a farmer, who had found it himself (Zammit, 1925). In a subsequent note, recently retraced in Zammit’s ield notes, the identity of the farmer who found the ring is disclosed, and Zammit speciies that the ring was discovered in a rock-cut tomb at Ghajn Klieb in Rabat, Malta (Zammit, 1926: 1923). He also adds that this was the same tomb from which he had unearthed a gold amulet in 1906 (Vella et al., 2001: 12; Sagona, 2002: 794-795). Nevertheless, it is not known when the ring was removed from the site. he current ring, which is part of the NMA collection, its the physical description of Zammit, but difers substantially in weight. he current ring weighs 4.418 g, about half the mass of Zammit’s ring. A thorough search in the NMA archives for notes describing the acquisition of the NMA ring for the museum collection was inconclusive. he recent discovery of ive gelatine dry plate negatives of the ring published by Zammit has raised hopes that further documentation may still exist. Two of the gelatine plate negatives portraying the ring (Figs. 2a and 2b) are believed to be the same photographs included in Zammit’s 1925 paper. he photographs were however edited, perhaps to enhance speciic features appearing on the ring bezel (Fig. 2c). he glass plate negatives were used to help authenticate the NMA ring. he presents study is meant to answer two important research questions: is the NMA ring the same object described by Zammit in 1925? If the NMA ring is not authentic, what further information can be obtained from a non-invasive examination of the artefact? 2. METHODOLOGY Given the uniqueness of the artefact, a non-invasive examination was undertaken. he manufacturing technique was observed both visually and through optical and electron microscopy. he material of the ring was characterized by elemental X-ray analysis using an energy dispersive spectrometer attached to the electron microscope. Gelatine dry plate negatives he gelatine dry plate negatives were digitally recorded by scanning, and the digitized negatives converted into positi- 351 ves using Adobe Photoshop CS Version 8. Two of the more informative images (the same ones used for Zammit’s 1925 publication) are presented in Figure 2 (a, b). he representation of the vessel obtained from the gelatine plates was visually compared to the one currently visible on the NMA ring (Fig. 1) for authentication purposes. Optical Microscopy he NMA ring was observed at low magniication using a Nikon SMZ 2T stereomicroscope at 1x and 3x magniication. Optical ibres (Fibre Optic Source GLI-156P) were arranged in such a manner that light relects of the surface of the metal at 45°. his angle reduces difuse relection of the metal to a minimum and permits a good observation of the surface. Where necessary, a polarizing plate was placed between the light source and the ring in order to minimize surface relections. Digital photographs were recorded directly through the microscope via a Leica PFC290 digital camera. he digital images were processed using the Leica IM 500 software, Version 5. Electron Microscopy Observation at higher magniication was performed using an Oxford Link 1430 Scanning Electron Microscope (SEM). Electron imaging was carried out in both secondary emission and backscatter modes. he secondary emission detector creates topographic images of the surface and was used to visualize manufacturing marks and other decorative features. he backscatter (BS) detector is designed for elemental contrast. In BS mode, the higher atomic weight elements appear white or light grey, and contrast with the lower atomic weight elements, which appear dark grey or black. he elemental X-ray analysis of the NMA ring surface was carried out using the energy dispersive spectrometer attached to the electron microscope. 3. RESULTS AND DISCUSSION Gelatine dry plate negatives he digitized positives, corresponding to Zammit’s photographs (Figs. 2a and 2b) were compared to the image of the NMA ring (Fig. 1). A number of diferences are evident: the galley mast projects into the edge decoration motif in Zammit’s photograph, while a gap exists between the mast and the edge decoration in the current ring. he stern post projects between the 3rd and 4th serration from ArcheoSciences, revue d’archéométrie, 33, 2009, p. 349-355 352 the right in Zammit’s photograph; it projects between the 4th and 5th serration in the NMA ring. he thickness of the irst oar, far right of the galley, constricts at its middle in the NMA ring. his is not the case in Zammit’s photographs. A surface loss occurs adjacent to the irst oar on the right of the NMA photograph. Again, this is absent in Zammit’s record. his photographic evidence suggests that the original ring was used as a model onto which the NMA ring was cast. he loss of ine decorative detail and the introduction of new surface defects would seem to conirm this hypothesis. Manufacturing technique and surface decoration Daniel VELLA et al. evenly spaced and measure between 5 and 10 µm in width (Fig. 3b). he even spacing of the iling marks and their small width suggest that they were made by a modern implement. Filing marks were also observed on the depressed surface forming the hull of the galley. he average width of these marks is 5 µm. he curved geometry of the surface indicates that a ile with a circular cross-section was used to shape this area. A coarser ile was used to inish the lat underside of the two hoops. he average width of these marks is 60 µm. It is probable that the goldsmith was not concerned about the quality of the surface inish here, knowing that these roughened surfaces would be hidden when the ring is worn or, indeed, when exhibited as a single piece. It appears that the ring hoops were cast as single pieces and decorated to produce the inal object (pers. comm. J. Aquilina, 2009). Tool marks were examined and recorded in an attempt to identify the technique used to decorate the ring and to shed light on its authenticity. Tool marks were categorized into two types, i.e. decoration marks (for example, a decorative motif along the edge of the bezel, oars and bow screens), and manufacturing marks (for instance, iling and shaping marks). Decoration marks he circular decoration around the bezel was produced by chasing. Chasing is carried out by tapping the metal surface with a small lat chisel having a rounded tip and building up a continuous ridge from a series of individual indentations (Craddock, 2009: 173). No metal is removed from the object during this process, but the dislodged material is pushed up and protrudes from the surface. he indentation marks observed on the NMA ring are circular, with an average diameter of 250 µm. he oars and bow screen structure were also chased into shape. he width of the marks forming the sea galley also measure around 250 µm; it appears that the same tool was used to produce all the decorations on the bezel. In another area of the edge decoration, the indentation marks are missing, although a palimpsest of the original decoration is still evident. If the NMA ring were indeed cast of an original ring, the smith would have had to emphasize the decoration motif by chasing or engraving over. We believe that this particular area of the ring was overlooked. Manufacture and shaping marks he serrated edges were inished using a very ine ile. he ile marks are barely visible under the stereomicroscope, but are well deined under the SEM, even at low magniication (Fig. 3a). At higher magniication, the ile marks appear ArcheoSciences, revue d’archéométrie, 33, 2009, p. 349-355 Figure 3: (a) Electron micrograph of serrated edges showing horizontal iling marks and (b) magniied image of the surface of one of the serrations. he width of the grooves is approximately 5-10 microns. Figure 3 : (a) Image au MEB des bords en dents de scie montrant les marques horizontales et (b) image ampliiée de la surface d’une des dents. La largeur des rainures est d’environ 5-10 microns. Copy or authentic? Analysis of a Phoenician gold ring from the National Museum of Archaeology… he underside of the bezel piece with the sail decoration exhibits a number of parallel grooves running along the horizontal length. he irregular width of the grooves and their disconnected nature excludes their deliberate manufacture. he marks probably resulted from a defective mould surface. Spherical shaped cavities also occur on this surface and are prominent on the serrations (Fig. 4). he entrapment of air bubbles between the molten metal and the surface of the mould during casting can explain these odd features, and their presence conirms the hypothesis of the NMA ring being a cast artefact. he fact that these defects are located on the back of the bezel might explain why the goldsmith spared their removal. 353 SEM backscattered images of the lat underside of the hoop exhibiting the bow screens decoration presented white areas with some dark grey along the outer edge of the ring. his diference in grey contrast suggests the presence of two diferent materials. X-ray point analyses were performed on the separate areas. White areas consisted of elemental gold (major element) accompanied by some silver and copper. Elemental silver (major element) and copper formed the grey areas. Sulphur was also detected over silver rich areas and indicates the presence of a silver tarnish. he absence of a gold signal on silver rich areas supports the hypothesis of a silver cast ring that was plated with gold. Material Analysis Close examination of the NMA ring under the stereomicroscope revealed a number of inconsistencies. A dark grey discoloration was observed in some areas along the ring edges, suggesting the presence of gold plate rather than a solid gold artefact. Indeed, if a gold plate were applied, the edges of the ring would be most at risk of material wear through handling and cleaning. A green coloured deposit was observed between the serrated edges of the ring. he green colour of this material suggests the presence of copper corrosion products, an unusual inding on gold items. Figure 4: (See colour plate) Photomicrograph of the underside of the bezel exhibiting the oar decoration. Note the horizontal grooves and the spherical cavities. Figure 4 : (Voir planche couleur) Image MEB du dessous du chaton montrant la décoration en forme de rames. Remarquons les rainures horizontales et les cavités sphériques. Figure 5: (a) Back scatter electron micrograph showing the three adjacent exposed serrations; (b) corresponding EDS X-ray spectrum of the exposed surface. he exposed material is composed of silver (main element) and copper. Figure 5 : (a) Image en électrons rétrodifusés de trois dents de scie adjacents ; (b) spectre de rayons X correspondant à la surface exposée. Le matériau analysé est constitué d’argent (élément principal) et de cuivre. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 349-355 354 Unequivocal evidence for a silver cast ring is presented in Figure 5a. hree of the serrations on the hoop exhibiting the bow screens were apparently iled down, exposing the underlying silver. his was performed after the ring was plated. Perhaps the fabricated copy required some adjustment for the hoops to it properly. he exposed material consists of silver and copper (Fig. 5b). he weight percentage of copper was determined semi-quantitatively at 10.2 ± 1.3 wt%. Elemental X-ray analyses performed on the green material revealed the presence of copper, carbon, oxygen, and sodium. he occurrence of copper corrosion products was therefore conirmed. Sodium is probably present as carbonate; this compound could have been used as a mild abrasive agent in polishing products. It is possible that in the early years following the introduction of the ring into the museum collection, the artefact was cleaned using abrasive polishing products. Indeed, the ring surface exhibits ine scratch marks that suggest it was polished, perhaps in preparation for exhibition. he determination of the thickness and composition of the gold layer requires destructive analyses of the ring. his, of course, was not possible. If the ring was plated in the mid-1920s, then this was very probably carried out by an electrolytic process – electroplating was introduced in the mid-19th century, about seventy years prior to the presumed manufacture date of the NMA ring (Hunt, 1973). 4. CONCLUSION he results presented above conirm that the NMA ring is most probably a copy of the original ring seen and published by Zammit in 1925. he salient evidence supporting our conclusion is summarized below: – the clear diferences between the representations seen on the glass plate negative images (Zammit’s ring) and the representations on the NMA ring; – the evenly spaced iling marks observed on the serrated edges and bezel surface of the NMA ring; – the presence of copper corrosion products on the NMA ring. he SEM-EDS data established that the NMA ring was in fact composed of silver that was plated with gold. he silver alloy was found to contain about 90 wt% silver and 10 wt% copper. he NMA ring is thus not an authentic artefact. Nonetheless, it represents a very important piece of evidence in itself. Together with Zammit’s 1925 publication and the recently discovered dry gelatine glass plate negatives, it is one of the three surviving records for the existence of this ArcheoSciences, revue d’archéométrie, 33, 2009, p. 349-355 Daniel VELLA et al. important archaeological ind. Not only is the NMA ring considered important for its historical value, but it is now the only remaining copy of an original ring, very probably Phoenician. he likelihood that the NMA ring was commissioned by Zammit to serve as a copy of the original ring holds much ground. his suggestion, however, can only be conirmed once more evidence comes to light. Finally, the weight discrepancy between the NMA ring and Zammit’s published gold ring can be explained. For simplicity, we consider the gold plate on the NMA ring to have a negligible mass and Zammit’s published ring to be made entirely of pure gold. he density of gold is 19.32 g/ cm3 (Lide, 2004, Part 4: 59), which is almost twice that of the NMA silver, at 10.31g/cm3, calculated from the semiquantitative compositional data and the densities of the pure elements (Callister, 2000: 71-73). If indeed the NMA ring was modelled on the original ring, the volumes of the two rings should be very similar. Hence, a pure gold ring with this volume should weigh about 1.874 times the weight of the NMA ring, i.e. 8.280 g. his value comes close to the original mass measured by Zammit. he remaining discrepancy in mass can be explained by factoring in material losses resulting from the manufacture of the NMA ring, especially in shaping the ring serrations by iling. Acknowledgments his work would not have been possible without the help and assistance received from various institutions and individuals: Heritage Malta for inancially supporting this work; Dr Maria F. Guerra for her assistance with the interpretation of the analytical data; Lindsey Buttigieg for contributing to the recovery of the gelatine glass plate negatives; Maurizio Fenech from the Department of Metallurgy and Materials Engineering, University of Malta, for allowing use of the light microscopes; silversmith Joseph Aquilina for his contribution towards elucidating the manufacturing technique of the NMA ring; Alexia Grech for her help withthe 3D Laser scanning of the ring; Mario Galea for the fruitful discussions and continuous encouragement throughout this work. One of us (NCV) would also like to acknowledge the advice received from Sir J. Boardman, Dr A. Blackman, Prof. A.J. Frendo, Dr K. Lapatin, Dr E. Marcus, and Dr A.J. Parker whilst carrying out research on the iconography of this ring as Visiting Scholar at the J. Paul Getty Research Institute in Spring 2007. Editorial constraints did not allow the complete publication of the results in this paper. Copy or authentic? Analysis of a Phoenician gold ring from the National Museum of Archaeology… References BOARDMAN, J., 2003. Classical Phoenician Scarabs: A Catalogue and Study (Studies in Gems and Jewellery II/BAR International Series 1190). Oxford, Archaeopress. BONANNO, A., 2005. Malta, Phoenician, Punic and Roman. Malta, Midsea Books Ltd. CALLISTER, D.W., 2000. Materials Science and Engineering: An Introduction. 5th edition. New York, John Wiley and Sons Inc. CRADDOCK, P., 2009. Scientiic Investigation of Copies, Fakes and Forgeries. Oxford, Butterworth-Heinemann, Elsevier Ltd. CULICAN, W., 1986. Opera Selecta: From Tyre to Tartessos. Göteborg, Paul Åströms. HIGGINS, R.A., 1961. Greek and Roman Jewellery. London, Methuen. HUNT, L.B., 1973. he Early History of Gold Plating, a Tangled Tale of Disputed Priorities. Gold Bulletin 6: 16-27. LIDE, D.R. (ED.), 2004. Handbook of Chemistry and Physics. 85th edition. Boca Raton, CRC Press Taylor and Francis Group, LLC. MORRISON, J.S. and WILLIAMS, R.T., 1968. Greek Oared Ships 900-322 BC. Cambridge, Cambridge University Press. PISANO, G., 1974. I gioielli fenici di harros nel Museo Nazionale di Cagliari. Roma, CNR and Centro di Studio per la Civiltà Fenicia e Punica. 355 PISANO, G., 1987. Jewellery, in R.D. Barnett, C. Mendelson (eds.), harros: A Catalogue of Material in the British Museum from Phoenician and other Tombs at harros, Sardinia. London, British Museum Press, 78-95. QUILLARD, B., 1987. Bijoux Carthaginiois II: Porte-amulettes, scceaux-pendentifs, pendants, boucles, anneaux et bagues. Louvain-la-Neuve, Institut Supérieur d’Archéologie et d’Histoire de l’Art. SAGONA, C., 2002. he archaeology of Punic Malta, Ancient near Eastern Studies Supplement, Series 9, Leaven, Peeters. SMITH, R., 1803. Remarks on a gimmal ring. Archaeologia 14: 7-13. VELLA, N.C., BORDA, K., BRIFFA, J.M., FENECH, K., GRINDE, K., SCERRI, E., SISK, M., STÖGER, H. and ZAMMIT, M.E., 2001. Ghajn Klieb, Rabat (Malta): A preliminary report of an archaeological survey. Malta Archaeological Review 4: 10-16. ZAMMIT, T., 1925. On a Phoenician ring from Malta. Antiquaries Journal 2: 266-267. ZAMMIT , T., 1923-1926. Archaeological Field Notes VIII. Manuscript held at the Archives of the National Museum of Archaeology, Malta (Heritage Malta). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 349-355 Modern and ancient gold jewellery attributed to the Etruscans: a science-based study Orfèvrerie moderne et ancienne attribuée aux Étrusques : une étude scientifique Anna Rastrelli*, Marcello Miccio*, Lore Troalen**, Marcos Martinón-Torres***, Maria Filomena Guerra****, Salvatore Siano*****, Andrej Sumbera****** and Maria Luisa Vitobello******* Abstract: A ibula with a lion and a pair of cluster earrings from the collection of the Florence National Museum of Archaeology, attributed to the Etruscans and said to originate from Populonia but suspected to be modern productions, were studied in the framework of the European project AUTHENTICO. hey were compared to genuine objects from this same collection and from the collection of the National Museums Scotland. he study was carried out in situ in Florence with a stereomicroscope equipped with a digital camera, a handheld XRF spectrometer, and the portable LIPS prototype developed by AUTHENTICO. Some objects were studied with SEM and X-radiography. he aim of the study was to show that portable techniques of examination and analysis provide information on gold jewellery that allow the diferentiation of modern and ancient productions. Résumé : La ibule au lion et les boucles d’oreille à motifs loraux de la collection du musée national d’Archéologie de Florence, attribuées aux Étrusques, provenant de Populonia, mais suspectées d’être des productions modernes, ont été étudiées dans le cadre du projet européen AUTHENTICO et comparées à des objets originaux de cette même collection et de la collection des Musées Nationaux d’Écosse. L’étude a été réalisée in situ avec une loupe binoculaire équipée d’un appareil photo numérique, un analyseur de luorescence X portable miniaturisé et le LIPS portable développé dans le cadre du projet AUTHENTICO. Certains objets ont été sélectionnés pour analyse au MEB et par radiographie. Le but de ce travail est de montrer que les informations obtenues grâce aux techniques portables d’examen et d’analyse permettent de diférentier les productions modernes des anciennes. Keywords: Etruscan, fakes, goldwork, XRF, LIPS, SEM, stereomicroscope. Mots-clés : Étrusque, faux, orfèvrerie, FX, LIPS, MEB, binoculaire. * Soprintendenza Archeologica per la Toscana, Firenze-Centro di Restauro – Via della Pergola, 65, 50121 Floerence, Italy. (annarastrelli@yahoo.it) ** National Museums Scotland, Department of Conservation & Analytical Research – 242 West Granton Road, Edinburgh EH5 1JA. (l.troalen@nms. ac.uk) *** UCL Institute of Archaeology – 31-34 Gordon Square, London WC1H 0PY, United Kingdom. (m.martinon-torres@ucl.ac.uk) **** Laboratoire du Centre de Recherche et de Recherche et de Restauration des Musées de France, UMR171 CNRS – 14, quai François-Mitterrand, 75001 Paris, France (maria.guerra@culture.gouv.fr) ***** Istituto di Fisica Applicata, CNR – via Madonna del Piano (Ed C), 50019 Sesto Fiorentino (FI), Italy. (s.siano@ifac.cnr.it) ****** DIGART SME – Conservation & Restoration – Vlnitá 56, 14700 Praha 4, Czech Republic. (a.sumbera@volny.cz) ******* EJTN GEIE, European Jewellery Technology Network – 124 rue du Commerce, Brussels, Belgium. (info@ejtn.org) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 357-364 358 1. INTRODUCTION he high skill of the Etruscan goldsmiths is largely revealed by the incomparable quality, delicacy, and complexity of the iligree and granulation patterns of their jewellery, as well as by the thinness of the gold foils employed and the excellence of the joining techniques. Among the most representative objects of the 6th century BC ine Etruscan gold work, the bauletto (basket) earrings can only be fully appreciated with a magniication lens. Accurate observation at low and high magniication of Etruscan gold items produced around the 6th century BC shows the following characteristics (Guerra, 2006; Guerra et al., 2007): plain and hollow granules may reach several millimetres but, in the case of dust granulation, their diameter may be as small as 120 µm; diferent wires – beaded, twisted, spooled, strip-twisted, etc. – may reach diameters of about 150 µm; and the gold foils used in the small pendant decorations are often about 100 µm thick. he dimensions of the added decorative elements can be observed under the stereomicroscope in Figure 1 for one bauletto earring from the collection of the National Museums Scotland (NMS). In Etruscan gold objects, joining is usually carried out by copper difusion (Mello et al., 1983); the technique is described by Pliny, heophilus and Cellini (Wolters, 1981). However, hard-soldering was also used for the application of large elements. he use of both techniques can be found in a single object, when, for example, decoration patterns are made with elements of diferent dimensions, such as the reels from Vulci (reference I.8 in Gaultier and Metzger, 2006), dated to the beginning of the 4th century BC (Guerra et al., 2007). In spite of the technical diiculties inherent to the reproduction of Etruscan work (Rudoe, 1991), partly owing to the lack of knowledge regarding Etruscan art, at least in the 19th century, copies and imitations, as well as invented items of Etruscan inspiration, have been produced since the 18th century for the antiquities market (Guerra, 2008a). he Pasinati cista and the Praeneste gold ibula, both attributed to Francesco Martinetti (Gordon, 1982; Williams, 1992), are known Etruscan fakes. he sarcophagus with the inscription of the gold ibula from Chiusi (Borrelli, 1992) purchased by the British Museum (Williams, 1992), the bronze chariot said to originate from Orvieto, and the colossal Etruscan warriors of Orvieto, purchased by the Metropolitan Museum of Art (Duchêne, 2006) are but three more examples of famous Etruscan fakes created by restorers who worked in the Campana workshop (19th century, Rome). he Campana collection included nearly 15,000 objects according to Borrelli (1992), some genuine, other fantasist and unscrupulous fakes. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 357-364 Anna RASTRELLI et al. Many gold objects held in museum collections were repaired in the past. Others are either pastiches produced using heteroclite ancient parts, or fully modern inventions (Guerra, 2006; Guerra et al., 2007; Guerra, 2008b). his is the case of some of the objects presented in this paper: the ibula with a lion and a pair of cluster earrings from the collections of the Florence National Museum of Archaeology (FNMA). Purchased in 1911 from Maruzzi and assumed to originate from Populonia, they can be conidently revealed as modern fakes. he aim of this work – carried out in the framework of the European project AUTHENTICO – is to provide analytical information on these objects, to compare their production technology and the composition of their alloys with genuine objects and pastiches from the FNMA, and to show how portable equipments can be used to provide criteria for the identiication of modern jewellery. A small number of objects from the NMS were also analysed for comparison. he study of the objects from the FNMA was carried out in the museum. hey were examined under a SMZ1000 Nikon stereomicroscope equipped with a digital camera, and by X-radiography. he analysis of the alloys was carried out with a handheld X-ray luorescence (XRF) spectrometer, Innov-X Systems Model Alpha 8000 LZX, with a silver tube, and operating at 40 kV, 29 µA. he area of analysis was of about 6mm2. Precision and accuracy were tested against a set of gold standard alloys (Table 1). he majority of the results provided are averages of 2-3 measurements and they have been normalised to 100 wt%. he objects studied at the NMS were examined using an Olympus stereomicroscope (Olympus SZX12 x 7-90) equipped with a digital camera (Olympus DP70), a CamScan Scanning Electron Microscope and a 300 kV Pantak X-radiography system. hey were also analysed by X-ray Fluorescence with an Oxford Instruments ED 2000 system, with a Rhodium target X-ray tube at 46 kV. Details on this equipment can be found in Troalen et al. (2009). A few objects were selected for analysis with the portable low-cost laser induced plasma spectroscopy instrument (LIPS) developed in the scope of the European project AUTHENTICO, based on a compact Q-switched Nd:YAG laser (further details in Agresti et al., 2009). In this case, measurements were carried out with a spot of about 70 µm diameter. Based on the diferent results obtained for the gold alloy standards of Table 1, the ablation could be limited to 100-200 µm laser pulses, which prevented major deposition of Cu and Ag oxides around the ablation hole, and allowed observing the saturation of the elemental depth proile towards the composition of the bulk. he total ablation Modern and ancient gold jewellery attributed to the Etruscans: a science-based study 359 depth was of about 60 µm. he calibration for quantitative elemental analyses was carried out with a set of gold alloy standards. a 2. RESULTS he bow-shaped ibula with a lion and the pair of cluster earrings were purchased by the FNMA in 1911 as Etruscan jewellery from Populonia. he objects and their mounting, revealed by X-radiography, are shown in Figure 2. he ibula was produced in two parts: (1) the lion obtained by joining two embossed metal sheets decorated with granules; (2) the shank obtained by decorating a gold sheet with a line of ive animals in granulation, outlined by two rows of granules. he 1 mm diameter spring is coiled one and a half times and soldered to the ibula under the lion. he cluster earrings consist of a horseshoe-shaped central element with a wide oval single piece clasp on the back. he central element is decorated with knurled and wavy wires, ropes, lowers with ive petals and gold dust granulation, and it is adorned with hollow globules with a diameter ranging from 220 to 3000 µm. he observation of the ibula under the stereomicroscope shows that the granules of 300-500 µm diameter forming the granulation patterns are joined by hard-soldering. Figure 3a reveals some solder left over on the metal sheet. In spite of being obtained by cutting small pieces of wires in an Etruscan tradition (Nestler and Formigli, 1994), as illustrated in Figure 3b, the granules are far too irregular to be considered as ancient work. Comparison of the granulation patterns from an ancient earring (Fig. 1) and the modern ibula (Figs. 3a and 3b) clearly shows this aspect. Formigli’s (1985) assumption of a modern fabrication and our analytical results are conirmed by the observation of the spring, for which the use of a modern drawn wire, characterised by its mechanical seams (Oddy, 1977; Ogden, 1991; Swadling et al., 1991), is obvious (Fig. 3c). We also note that the section of the spring terminal is reduced by cutting with a chisel instead of being hammered, as expected for an ancient fabrication. he cluster earrings show a quality of work very similar to the one of the lion ibula. Figure 4a shows details of the earring: the use of hard-soldering can be identiied by observation only and the thick gold sheets with marks of ile polishing on the borders that contrast with the delicacy of the lion’s head and the lower of the genuine bauletto earring are shown in Figure 4b. he iligree patterns are also of very dissimilar quality in modern and Etruscan productions (Ogden, 2004), such as b Figure 1: Details of a bauletto earring (reference K2004.40.1) from the NMS collection showing (a) iligree and (b) granulation patterns. Figure 1 : Détails des motifs en (a) iligrane et (b) granulation de la boucle d’oreille à baule (référence K2004.40.1) de la collection du NMS. in several objects belonging to the Campana collection held in the Louvre museum (Bagault et al., 2006). In order to illustrate these aspects, Figure 5 shows SEM-SEI images of earring A.1905.914.2 from the collections of the NMS. his earring consists of modern and ancient beads. he quality of the strip-twisted wires of the ancient granulated beads on Figure 5a, certainly joined by copper difusion, contrasts with the quality of the beaded and block-twisted wires of the modern beads in Figure 5b, which were joined with thick hard-solder. If we exclude some particular cases (such as the use of modern 18 carat gold alloys, Guerra et al., 2007), the composition of the modern and ancient alloys are in most cases ArcheoSciences, revue d’archéométrie, 33, 2009, p. 357-364 360 XRF Oxford rolled gold heoretical Value Standards NMS XRF Oxford GCS5 gold heoretical Value XRF Authentico A1 heoretical Value XRF Authentico A2 heoretical Value Standards AUTHENTICO XRF Authentico E1 heoretical Value XRF Authentico E2 heoretical Value XRF Lyon Allemand A heoretical Value XRF Lyon Allemand 6917 heoretical Value XRF Standards Lyon Allemand C2RMF 6907 heoretical Value XRF Lyon Allemand 6905 heoretical Value XRF Lyon Allemand 6906 heoretical Value Anna RASTRELLI et al. Au % 79.4 80.0 80.0 79.4 92.1 92.0 71.3 70.0 47.0 50.0 51.9 55.0 91.6 92.0 74.2 75.0 75.8 75.2 74.4 75.0 73.5 75.0 Ag % 10.6 10.0 17.1 17.9 6.7 6.0 25.1 27.0 53.0 50.0 47.7 44.0 4.1 4.0 18.9 18.0 5.0 5.7 6.8 6.0 14.0 12.5 Cu % 10.0 10.0 2.9 2.7 1.2 2.0 3.6 3.0 0.5 1.0 4.4 4.0 7.0 7.0 19.2 19.1 18.8 19.0 12.5 12.5 Table 1: Results obtained by stationary and handheld XRF for the gold alloy standards. Tableau 1 : Compositions obtenues par luorescence X ixe et portable miniaturisée pour les standards d’or. a b Figure 2: Photography and X-radiography of the FNMA (a) lion ibula (reference 85037: 14.05 g; 9.4 cm long) and (b) one element of the pair of cluster earrings (reference 85036: 27.70 g; 6.1 cm height and 3.5 cm large). Figure 2 : Photographie et radiographie à rayons X de (a) la ibule au lion (référence 85037 : 14,05 g ; 9,4 cm de longueur) et (b) d’une des boucles d’oreille à motifs loraux de la paire du FNMA (référence 85036 : 27,70 g ; 6,1 cm de hauteur et 3.5 cm de largeur). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 357-364 Modern and ancient gold jewellery attributed to the Etruscans: a science-based study a 361 b c Figure 3: (See colour plate) he lion ibula: (a) granulation pattern, (b) detail of the granulation pattern, and (c) seams characteristic of the use of a modern drawn-plate to produce the spring. Figure 3 : (Voir planche couleur) La ibule au lion : (a) motifs en granulation, (b) détail du motif en granulation et (c) lignes caractéristiques de l’utilisation d’une ilière pour fabriquer le ressort. a b Figure 4: Details of the (a) modern cluster earring with indication of the ile marks; (b) ancient bauletto earring reference 15718 FNMA (Currie’s collection, 1836). Figure 4 : Détails de la boucle d’oreille à (a) motifs loraux modernes avec localisation des traces de lime et (b) baule étrusque référence 15718 FNMA (collection Currie 1836). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 357-364 362 Anna RASTRELLI et al. a b Figure 5: SEM-SEI micrographs of earring A.1905.914.2 from the NMS, showing the diference in quality between the (a) ancient and (b) modern gold work (scale: 500 µm). Figure 5 : Images MEB-SE de la boucle d’oreille A.1905.914.2 du NMS, montrant la diférence de qualité du travail d’orfèvrerie (a) ancien et (b) moderne (échelle: 500 µm). barely diferent (Guerra, 2006; 2008b). In the present work, we analysed the lion ibula and the cluster earrings by portable XRF and LIPS. A few Etruscan bauletto earrings from the collections of FNAM and NMS, and a genuine pendant and a pastiche from NMS were also analysed for comparison. he results obtained are shown in Table 2. Considering the fact that LIPS analysis is carried out on a small region of the object while XRF analysis is carried out on a large zone, which may include the region of interest (plate, granule, etc.) together with the solders, we can say that the results provided in Table 2 for the two techniques applied to the FNMA jewellery are in good agreement. he largest diferences are observed for the bauletto earring 15718, which includes many decorative parts. he results obtained for the lion ibula and the cluster earrings show the use of diferent alloys to produce their diferent parts. hese alloys consistently have Ag values lower than 10%, and Cu values lower than 3.5%. Within the relative compositional variation, we can tentatively identify an alloy containing about 7-9% Ag and 2-3% Cu, another containing about 5% Ag and about 1% Cu, and a third one containing about 2% Ag with variable contents of Cu. We also note the use of diferent alloys for the production of the two elements of the pair of cluster earrings. he results obtained for the lion ibula and the cluster earrings are signiicantly diferent from the composition of the genuine objects analysed in the present work. hey are also diferent from the results obtained for the modern parts of several published ArcheoSciences, revue d’archéométrie, 33, 2009, p. 357-364 Etruscan pastiches (Guerra, 2006; Guerra et al., 2007), such as cartouche inger-rings (74-79% Au, 15-21% Ag, and 3-6 Cu%, Guerra, 2006). he high concentrations of silver and copper detected in the granulation applied to the lion’s head are explained by the presence of leftover solder. Of potential interest is the notably high iron content detected on the earrings’ surface (with one reading reaching a Fe concentration of 11%): this is possibly due to a surface coating with iron oxides, which might be explained by the goldsmith’s intention of giving to the modern productions the aspect of objects originating from archaeological excavations. Earring A.1905.914.2, suspected to be a modern mounting of ancient and modern parts, shows a very homogeneous composition, with an average of 10.4% Ag and 2.6% Cu. his composition is distinct from the other published genuine objects, such as the Vulci reels, made with an alloy containing 97.5% Au, 1.5% Ag, and 1% Cu, and the ancient parts of the hinged bracelets made with an alloy containing 94% Au, 4% Ag and 2% Cu, while the alloy of the modern parts contains 98 % Au, 1.3% Ag and 0.7% Cu (Guerra et al., 2007). However, the granulated bead is without a doubt ancient. hese results seem to indicate that the other beads could be partially ancient, but restored before being assembled in a modern mounting. he higher silver contents of this alloy are explained by the use of a silver-rich solder. Modern and ancient gold jewellery attributed to the Etruscans: a science-based study Cu % Ag % FNAM XRF Fibula 85037 lion’s body granulation neck* spring shank Cluster earring 85036/1 back plate three globules top side sphere top central sphere top 1 2 Cluster earring 85036/2 front lower back hook three globules top side sphere top central sphere top front lower Bauletto earring 15718 main barrel Bauletto earring 15751 main barrel side plaque Bauletto earring 7746 main barrel side plaque Au % Fe % 0.6 2.2 0.4 3.4 2.7 17.4 2.2 8.7 96.3 77.0 100.0 88.4 2.8 3.5 0.5 1.1 2.7 2.0 9.7 2.0 2.3 8.8 95.2 85.6 95.8 94.5 77.5 ≤0.5 1.2 1.7 2.2 11.0 2.0 1.1 1.3 1.0 1.2 1.9 6.7 ≤0.2 7.4 4.9 5.4 5.7 89.1 98.2 88.1 93.4 92.7 88.8 2.3 0.7 3.2 0.7 0.7 3.6 3.1 4.4 91.5 1.0 2.7 2.1 17.8 16.6 78.5 80.7 2.0 2.9 FNAM LIPS 4.5 4.5 92.9 92.2 0.2 1.4 98.4 2.8 2.5 94.7 2.0 1.9 NMS XRF Bauletto earring A.1956.381 Front plate 1.9 Lateral plate 1.8 Bauletto earring A.1956.381 A 2.3 Front plate 1 2 1.9 Lateral plate 1.7 Earring A.1905.914.2 Granulated bead 2.7 2.7 Small bead 1 2.5 2 3 2.5 4 2.7 3.5 2.9 94.5 95.2 1.9 2.1 96.2 96.2 1.9 3.0 1.8 95.8 95.1 96.5 11.3 11.7 10.7 7.7 10.7 86.0 85.6 86.8 89.8 86.6 Fibula 85037 lion’s body Cluster earring 85036/1 back plate Bauletto earring 15718 clasp big granule 0.7 *granules and solder together Table 2: Compositions obtained for the diferent objects from FNMA and NMS collections analysed in this work by handheld XRF, stationary XRF, and portable LIPS Tableau 2 : Compositions obtenues pour les diférents objets appartenant aux collections du FNMA et du NMS, analysés pour ce travail par luorescence X avec système portable miniaturisée et ixe et par LIPS portable. It is at this stage impossible to establish convincing criteria for the diferentiation of modern and ancient productions by measuring only the elemental composition of the objects. Only the results published by Cesareo and Von Hase (1976) on the compositions of the reproductions of gold Etruscan objects made by Castellani in the 19th century showed the use of diferent gold alloys, containing 2.5-6.6% Ag and 363 0.6-2.2% Cu. Some of these alloys are close to those obtained for the lion ibula and the cluster earrings. In any case, it has to be acknowledged that, given the variable compositions documented, it is not possible to discriminate modern from ancient gold items based on Au/Ag/Cu ratios only. 3. CONCLUSION he analytical information obtained in the framework of the European project AUTHENTICO using portable equipment for the study of the lion ibula and the pair of cluster earrings from Populonia, attributed to the Etruscans and belonging to the collections of the FNMA, provided convincing evidence that they are fakes. Examination of these objects showed the use of thick gold sheets inished by ile polishing, drawn wires, and big granules joined by hardsoldering. he delicacy of the Etruscan iligree and granulation patterns, the thinness of the gold foils, and the excellent joining techniques are not matched in the modern pieces. he elemental analysis of Etruscan genuine gold items and pastiches from FNMA and NMS revealed the variety of the ancient and modern alloys, showing that composition alone does not provide suicient criteria for distinguishing between ancient and modern productions. Etruscan genuine objects, pastiches with modern and ancient parts, and 19th century reproductions of Etruscan objects published by different authors were also taken into account for this comparison. Further comparison of the technical aspects of ancient and modern goldsmithing, as well as further analysis of the alloys used in the manufacture of modern and ancient gold items, should allow the establishment of an useful database for authentication assessment. In the present work, we note how both XRF with stationary and portable systems and LIPS can be successfully used in this type of study, even taking into account the error introduced by the analysis of non-lat surfaces and the diferent degrees of surface damage. References AGRESTI J., MENCAGLIA A.A. and SIANO, S., 2009. Development and application of a portable LIPS system for characterising copper alloy artefacts. Analytical and Bioanalytical Chemistry 395(7): 1618-2642. BAGAULT, D., BOREL, T. and GUERRA M.F., 2006. Les bracelets articulés de la collection Campana, in F. Gaultier, C. Metzger (eds.), Trésors antiques: bijoux de la collection Campana. Milan, 5 Continents; Paris, Musée du Louvre, 114-117. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 357-364 364 BORRELLI, L.V., 1992. Faux, pastiches, imitations, in M. Pallotino (ed.), Les Étrusques et l’Europe. Paris, Réunion des musées nationaux, 432-439. CESAREO, R. and VON HASE, F.W., 1976. Analisi di Ori Etruschi del VII sec. a.C. con uno strumento portatile che impiega la tecnica della luorescenza X eccitata da radioisotope. Atti dei Convegni Lincei 11: 259-296. DUCHÊNE, H., 2006. La guerre du faux n’aura pas lieu. Dossiers d’Archéologie 312: 2-7. FORMIGLI, E., 1985. Techniche dell’oreiceria etrusca e romana: originali e falsiicazioni. Firenz, Sansoni editore Nuova S.p.A. GAULTIER, F. and METZGER, C., 2006. La collection de bijoux: la composition, in F. Gaultier, C. Metzger (eds.), Trésors antiques: bijoux de la collection Campana. Milan, 5 Continents, Paris, Musée du Louvre, 192 p. GORDON A.E., 1982. On la cosiddetta fibula Prenestina by Margherita Guarducci. he Classical Journal 78(1): 64-70. GUERRA, M.F., 2006. Etruscan gold jewellery pastiches of the Campana’s collection revealed by scientiic analysis, in M. Cavallini, G.E. Gigante (eds.), De Re Metallica: dalla produzione antica alla copia moderna, Studia Archaeologica 150. Roma, L’Erma Di Bretschneider, 103-128. GUERRA, M.F., 2008a. Archaeometry and museums: Fifty years of curiosity and wonder. Archaeometry 50(6): 951-967. GUERRA, M.F., 2008b. Etruscan gold jewellery: genuine, restored or pastiche?, in S. Rovira Llorens, M. García-Heras, M. Gener Moret, I. Montero Ruiz (eds.), Actas del VII Congreso Ibérico de Arqueometría. Madrid, Quadro-CSIC, 479-489. GUERRA, M.F., BAGAULT, D., BOREL, T., DI MANTOVA, A., ESQUÈS, C. and PLÉ, E., 2007. Examen et analyse élémentaire de bijoux étrusques de la collection Campana, in F. Gaultier, C. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 357-364 Anna RASTRELLI et al. Metzger (eds.), Les bijoux de la collection Campana: de l’antique au pastiche. Paris, École du Louvre, 145-177. MELLO, E., PARRINI, P. and FORMIGLI, E., 1983. Etruscan Filigree: Welding Techniques of Two Gold Bracelets from Vetulonia. American Journal of Archaeology 87(4): 548-551. NESTLER, G. and FORMIGLI, E., 1994. Granulazione Etrusca: Un’antica Tecnica Orafa. Siena: Nuova Immagine. ODDY, A., 1977. he Production of Gold Wire in Antiquity. Gold Bulletin 10(3): 79-87. OGDEN, J., 1991. Classical gold wire: some aspects of its manufacture and use. Society of Jewellery Historians 5: 94-105. OGDEN, J., 2004. Revivers of the lost Art: Alessandro Castellani and the quest for classical precision, in S.W. Soros and S. Walker (eds.), Castellani and Italian Archaeological Jewelry. New Haven, London: Published for the Bard Graduate Center for Studies in the Decorative Arts, Design, and Culture, New York, by Yale University Press, 181-200. RUDOE, J., 1991. Alessandro Castellani’s letters to Henry Layard. Society of Jewellery Historians 5: 107-119. SWADDLING, J., ODDY, A. and MEEKS, N., 1991. Etruscan and Other Early Gold Wire from Italy in Classical Gold Jewellery and the Classical Tradition. Society of Jewellery Historians 5: 7-21. TROALEN, L., GUERRA, M.F., TATE, J. and MANLEY, B., 2009. Technological Study of Gold Jewellery from the 17th and 18th Dynasties in Egypt. ArcheoSciences 33. WILLIAMS, D., 1992. he Brygos tomb reassembled and 19thcentury commerce in Capuan antiquities. American Journal of Archaeology 96(4): 617-636. WOLTERS, J., 1981. he ancient craft of granulation. Gold Bulletin 14(3): 119-129. Gold hracian appliques: authentic or fake? Appliqués thraces en or : authentiques ou faux? Ivelin Kuleff*, Totko Stoyanov** and Milena Tonkova*** Abstract: A few years ago, 25 gold appliqués with a total weight of 255 g and dated to a period between the 5th and the 3rd centuries BC were ofered to the National History Museum in Soia (Bulgaria). Some Bulgarian archaeologists – experts in the ield of toreutics – expressed the opinion that the ofered inds are originals. At the same time, the results of an investigation carried out by the Bulgarian Police brought evidence to the contrary, and, on these bases, started a hearing of the case. Using a non-destructive method of investigation (ED-XRF), the concentrations of Ag, Au, and Cu were determined. A detailed observation of the surface of the inds with relective optical microscopy was carried out, and, at the same time, stylistic parallels of the ofered inds were sought by comparing them to the known similar ancient appliqués. On the bases of the chemical composition, technological observations, and the comparison with parallels of similar chronological inds reported in the literature, we established that the 25 golden appliqués ofered to the Museum by the treasure hunters are fake. Résumé : Il y a quelques années, le Musée National de Soia (Bulgarie) a acquis 25 appliqués en or, d’un poids total de 255 gr, datées du Ve-IIIe siècle av. J.-C. Des archéologues bulgares, experts en toreutique, ont identiié les objets comme étant des originaux. Cependant, des recherches menées par la Police bulgare ont débouché sur une remise en question de l’authenticité de ces pièces. Une étude non destructive (par ED-FX) a permis de déterminer les concentrations d’argent, d’or et de cuivre des appliqués en or. Leur surface a été observée minutieusement au microscope optique et une analyse stylistique et iconographique a été réalisée. Les résultats de l’analyse de la composition chimique du métal, les observations technologiques et la comparaison avec des parallèles archéologiques ont inalement permis de formuler la thèse que ces 25 appliques en or sont fausses. Keywords: authentication, Bulgaria, copper, ED-XRF, gold, silver. Mots-clés : argent, authentiication, Bulgarie, cuivre, ED-FX, or. 1. INTRODUCTION A few years ago, 25 gold appliqués with a total weight of 254.83 g and dated to a period between the 5th and the 3rd centuries BC were ofered to the National History Museum in Soia (Bulgaria). Some Bulgarian archaeologists – experts in the ield of toreutics – expressed the opinion that the ofered inds are originals. At the same time, the results of an investigation carried out by the Bulgarian Police brought evidence to the contrary, and, on these bases, started a hea- ring of the case. he authors of the present study were involved in this project as experts aiming to determine if the ofered appliqués are original or fake. Using energy dispersive X-ray fluorescence analysis (ED-XRF) for the determination of the basic chemical composition of the objects, relective optical microscopy for a detailed observation of the surface of the inds, and research for stylistic parallels to the ofered inds in the existing literature, we propose our opinion regarding the originality of the appliqués. his paper presents the results of this investigation. * Faculty of Chemistry University of Soia, Bulgaria. (kulef@chem.uni-soia.bg) ** Faculty of History University of Soia, Bulgaria. (totko@mail.bg) *** National Archaeological Institute with Museum, Bulgarian Academy of Sciences, Soia, Bulgaria. (milenatonkova@hotmail.com) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 365-373 366 Ivelin KULEFF, Totko STOYANOV, Milena TONKOVA 2. EXPERIMENTAL Method of analysis Materials he chemical composition of the investigated 25 gold inds was determined using ED-XRF (Shimatzu EDX720) at the Laboratory of Conservation and Restoration of the National Archaeological Institute with Museum of the Bulgarian Academy of Sciences (NAIM). he X-ray lines used for analysis and other parameters of the instrument are presented in Kulef et al. (2009). he analytical results obtained are presented in Table 2. he description of the investigated objects is provided in Table 1. In Figures 1 and 2, some images of the objects in question are presented. he place where the investigated gold appliqués were found is unknown, because they were ofered to the National History Museum in Soia (Bulgaria) by treasure-hunters. It was assumed that the objects were found somewhere in north-eastern Bulgaria. Figure 1. Images of the investigated objects and some optical observations. 1a = head of ram (5 objects); 3 = struggle of animals (3 objects) (rectangle form); 6 ÷10 = ‘copies’ of some of the igures on the bronze mould from Gartshinovo (6 objects); 11, 12d = grifon attacking goat (5 objects); 13c = image of grifon (4 objects); Garchinovo mould – image of the bronze matrix from the village of Gartshinovo. Images under the microscope of the surface of some of the objects, showing the texture. Figure 1 : Photographies des objets étudiés et des observations optiques. 1a = tête de bélier (5 objets) ; 3 = combat d’animaux (3 objets) (forme rectangulaire) ; 6-10 = empreintes de certaines igures du moule en bronze de Gartshinovo (6 objets) ; 11, 12a = grifons attaquant un bouc (5 objets) ; 13c = représentation d’un grifon (4 objets) ; moule de Gartshinovo – photographique du moule en bronze du village de Gartshinovo. Prises de vue microscopiques de la surface de certains des objets illustrant la texture. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 365-373 367 Gold hracian appliques: authentic of fake? code 1-a-151* 1-b-155 1-c-156 1-d-161 1-e-173 2 - 2177 3 - 2117 4 - 2066 5 - 2653 6-a-246 6-b-435 7 - 500 8 - 360 9 - 2742 10 - 2310 11-1214 12-a-596 12-b-654 12-c-679 12-d-761 13-a-831 13-b-930 13-c-955 13-d-1076 14 - 1385 description An appliqué in a shape of a ram’s head with holes (Fig. 1) An appliqué in a shape of a ram’s head with holes An appliqué in a shape of a ram’s head with holes An appliqué in a shape of a ram’s head with holes An appliqué in a shape of a ram’s head with holes An appliqué - a head of the Gorgon-Medusa, with holes (Fig. 2) Rectangular appliqué – two grifons attacking a horse, with holes (Fig.1) Rectangular appliqué – a lion attacking a grifon, with holes (Fig. 2) An appliqué with a bust representation of goddess An appliqué in a shape of a lion (?) walking to the left, with holes (Fig. 1) An appliqué in a shape of a lion walking to the right, with holes (Fig. 1) An appliqué in a shape of a kneeling deer, with holes (Fig. 1) An appliqué in a shape of a boar walking to the left, with holes (Fig. 1) An appliqué in a shape of a eagle, with holes (Fig. 1) An appliqué in a shape of a lion walking to the right, with holes (Fig. 1) Trapezoid appliqué – winged lion attacking a goat, with holes (Fig. 1) Triangular appliqué – winged lion attacking a goat, with holes Triangular appliqué – winged lion attacking a goat, with holes Triangular appliqué – winged lion attacking a goat, with holes Triangular appliqué – winged lion attacking a goat, with holes (Fig. 1) Rectangular appliqué – a grifon to the left, with holes Rectangular appliqué – a grifon to the right, with holes Rectangular appliqué – a grifon to the left, with holes (Fig. 1) Rectangular appliqué – a grifon to the right, with holes An appliqué with a representation of two grifons attacking a goat, with holes (Fig. 2) Table 1: Description of the investigated objects. * = he numbers represent the weight of the investigated appliqués, expressed roughly in centigrams. Tableau 1 : Description des objets étudiés. * = Les nombres représentent le poids des appliqués exprimé en centigrammes. Typological classification A preliminary analysis carried out on the representations on the appliqués in question leads to their division in two general groups. he irst one (Table 1, Nos. 1-3, 11-14 – in all, 18 pieces) includes artefacts having parallels in inds from the 5th-4th century BC rich burials in Scythia. Some ive subgroups could be distinguished as well: 1) ive appliqués with representations of a ram’s head (Fig. 1: 1a; see Artamonov 1970: 36-39, Abb. 35, 48; Galanina and Grach 1986: Abb. 80); 2) three appliqués representing animals struggling (Fig. 1: 3; Fig. 2: 4, 14; see Artamonov 1970: 36-39, Abb. 35, 48; Galanina and Grach 1986: Abb. 118; Treister, 2001: Fig. 63); 3) subgroup of 5 appliqués representing a winged lion attacking a goat (Fig. 1: 11, 12d; see Artamonov, 1970: Taf. 122; Galanina and Grach, 1986: Abb. 106); 4) subgroup consisting of 4 appliqués representing a grifon (Artamonov, 1970: Abb. 93; Galanina and Grach, 1986: Abb. 195; also the gold appliqués from a rich grave at Kralevo, NE Bulgaria – Echt 2004: Kat. No 224g); and 5) an appliqué representing a head of Medusa (Fig. 2: 2; see Artamonov, 1970: 36-39, Abb. 35, Taf. 103; Galanina and Grach, 1986: Abb. 218, 259). he 6 pieces of the second group (Table 1, Nos. 6-10) are representations of animals which are to be qualiied as clumsy imitations of some of the animals that are part of the famous bronze mould from Gartshinovo, Shumen district, NE Bulgaria (Fig. 1: centre 6-10; see Damyanov, 1998; Treister, 2001: 161-168, Figs. 1-2; Venedikov and Gerasimov, 1979: 94-96, 370, No 152). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 365-373 368 Ivelin KULEFF, Totko STOYANOV, Milena TONKOVA Figure 2: (See colour plate) Images of the investigated objects and some optical observations. 2 = Medusa (1 object) – image under the microscope showing traces of casting (photograph by P. Penkova); 4 = image under the microscope showing a very smooth surface with lack of traces of use (photograph by P. Penkova); 14 = traces indicating the use of a rolling-mill for the preparation of thin foil (photograph by P. Penkova). Figure 2 : (Voir planche couleur) Photographies des objets étudiés et des observations optiques. 2 = Méduse (1 objet) – prise de vue microscopique montrant les traces de moulage (photo de P. Penkova) ; 4 = prise de vue microscopique montrant une surface polie sans traces d’utilisation (photo de P. Penkova) ; 14 = traces d’utilisation de cylindres pour la réalisation de ines feuilles. 3. RESULTS AND DISCUSSION Chemical composition All the investigated objects are prepared from an alloy with very high gold content – 89% to 99.4%, corresponding to a range between 21.4 and 23.8 carats gold (see Table 2). On the basis of the analytical data, the investigated objects could be classiied in two groups: ArcheoSciences, revue d’archéométrie, 33, 2009, p. 365-373 1) objects with basic chemical content of Au and Ag without copper – 6 objects (4; 6-a; 6-b; 7; 8; 10). hese objects are prepared practically from pure gold (23.8 carats); 2) objects with basic chemical content of Au, Ag and Cu – 19 objects (1-a; 1-b; 1-; 1-d; 1-e; 2; 3; 5; 9; 11; 12-a; 12-b; 12-c; 12-d; 13-a; 13-b; 13-c; 13-d; 14). According to their copper content, the objects belonging to the second group could be additionally divided in 3 subgroups: 2.1) objects with copper concentration of less than 2.7%, lower than that of Ag – 9 objects: 1-a; 1-b; 1-c; 1-d; 1-e; 2; 3; 9; 11. he values of the ratio Ag/Cu are between 1.2 and 4.9; 369 Gold hracian appliques: authentic of fake? n° object gold [%] silver [%] M±SD RSD M±SD copper [%] RSD M±SD RSD Ag/Cu 1 1-a-151 97.78±0.06 0.06 1.44±0.04 2.8 0.81±0.07 8.6 2.53 2 1-b-155 97.66±0.08 0.08 1.61±0.09 5.6 0.72±0.05 6.9 2.24 3 1-c-156 97.54±0.23 0.24 2.02±0.04 2.0 0.55±0.10 18.2 4.93 4 1-d-161 97.26±0.11 0.11 2.10±0.06 2.9 0.64±0.05 7.8 3.28 5 1-e-173 97.77±0.04 0.04 1.43±0.02 1.4 0.80±0.03 3.8 1.79 6 2-2177 98.06±0.06 0.06 1.11±0.05 4.5 0.85±0.03 3.5 1.31 7 3-2117 98.67±0.28 0.29 1.03±0.03 2.9 0.30±0.30 100 3.43 8 4-2066 96.91±0.08 0.08 3.09±0.08 2.6 < 0.001 - (3090) 9 5-2653 96.93±0.62 0.64 0.64±0.37 57.8 2.43±0.26 10.7 0.26 10 6-a-246 99.24±0.01 0.01 0.76±0.01 1.3 < 0.001 - (760) 11 6-b-435 99.41±0.03 0.03 0.59±0.04 6.8 < 0.001 - (590) 12 7-500 99.22±0.02 0.02 0.78±0.02 2.6 < 0.001 - (780) 13 8-360 99.18±0.03 0.03 0.82±0.03 3.7 < 0.001 - (820) 14 9-2742 99.20±0.21 0.21 0.61±0.02 3.3 0.19±0.19 100 3.21 (840) 15 10-2310 99.16±0.04 0.04 0.84±0.04 4.8 < 0.001 - 16 11-1214 93.76±0.69 0.74 3.19±0.22 6.9 2.64±0.16 6.1 1.21 17 12-a-596 89.00±0.93 1.04 2.45±0.05 2.0 8.55±0.89 10.4 0.29 18 12-b-654 90.76±1.47 1.62 2.57±0.05 1.9 6.67±1.48 22.2 0.39 19 12-c-679 89.47±0.43 0.48 2.46±0.07 2.8 8.07±0.42 5.2 0.30 20 12-d-761 90.47±0.57 0.63 2.46±0.03 1.2 7.06±0.57 8.1 0.35 21 13-a-831 92.65±0.30 0.32 2.59±0.06 2.3 4.76±0.24 5.0 0.54 22 13-b-930 91.86±0.54 0.59 2.52±0.02 0.8 5.61±0.52 9.3 0.45 23 13-c-955 92.26±0.71 0.77 2.57±0.04 1.6 5.17±0.67 13.0 0.51 24 13-d-1076 92.35±0.55 0.60 2.64±0.09 3.4 5.01±0.64 12.8 0.53 25 14-1385 89.75±0.48 0.53 2.33±0.03 1.3 7.92±0.52 6.6 0.29 Table 2: he concentrations of gold, silver and copper in the investigated objects. M = mean value; SD = standard deviation; RSD = relative standard deviation RSD = Tableau 2 : La concentration d’or, d’argent et de cuivre des objets étudiés. M 100 [%] M = valeur moyenne ; SD = écart-type ; RSD = écart-type relatif RSD = M 100 SD [%] SD 2.2) objects with copper concentration higher than 4.5%, higher than that of Ag – 9 objects: 12-a; 12-b; 12-c; 12-d; 13-a; 13-b; 13-c; 13-d; 14. he values of the ratio Ag/Cu are between 0.3 and 0.5; 2.3) one object (5-2653) with copper concentration of 2.4%, which is higher than that of Ag (0.64 %), but below 2.7 %, with an Ag/Cu ratio of 0.26. he distribution of the investigated objects according to the concentrations of Au and Cu is presented in Figure 3. In all investigated objects, the concentration of iron was below the detection limit (< 0.001). Iron was found practically exclusively in areas presenting inhomogeneities visible with the naked eye. he presence of tin or platinum was not detected in any object. he results of the analysis could be briely summarized as follows: the objects are produced using gold alloys with very high purity – from 21.4 to 23.8 carats. Objects 5 and 11 are produced with a material which is diferent from the material used for the production of the other objects. Objects 6a, 6b, 7, 8, and 10 were produced using an alloy containing only gold and silver. Objects 1 and 2 were probably produced using the same alloy used for the production of appliqués 6a, 6b, 7, and 8. Object 4 was produced using the so-called ‘gold for artiicial teeth’. Copper was added to the alloy used for object 4 and objects 12, 13, and 14. Probably object 12a was produced irst; subsequently, the content of copper in the alloy decreased, as shown in Figure 3, presenting the sequences of production. According to this hypothesis, the last object that was produced is 13a. he reduction ArcheoSciences, revue d’archéométrie, 33, 2009, p. 365-373 370 Ivelin KULEFF, Totko STOYANOV, Milena TONKOVA Figure 3: Diagram of concentration of gold vs copper in the investigated objects. Figure 3 : Diagramme de concentration d’or vs cuivre des objets analysés. of the concentration of copper is due to oxidizing of copper during the melting of the alloy. he high concentration of gold (21.4 to 23.8 carats) and relatively low concentration of silver (less than 3.5 %) shows that the investigated objects were probably not produced from native gold. his assumption was supported by the relatively high concentration of copper (higher that 3%) in some of the investigated objects (12, 13, and 14). herefore, it could be established with great probability that at least part of the investigated appliqués were produced with artiicially prepared alloys. It could also be demonstrated with certainty that the appliqués were not produced with electrum – the alloy used for most of the investigated breast plates from hrace (see Kulef et al., 2009). At the same time, the hypothesis that the objects were produced using natural gold with a very high purity is impossible to accept. Such a hypothesis is not supported by the data about natural gold from Transylvania (Romania) (see Cojocaru et al., 2003), Bulgaria (Kovachev et al., 2007a), and Republic of Macedonia (Kovachev et al., 2007b; Stefanova et al., 2007), which are the nearest gold sources to the probable place where the artefacts were discovered. herefore, according to the results of the analysis, the gold used for producing the appliqués could be determined as technologically worked and reined gold, to which, in some cases, copper was intentionally added. According to the analytical results presented in Table 2, it is possible to propose that for the production of part of the appliqués (objects 12, 13, and 14), copper was intentionally ArcheoSciences, revue d’archéométrie, 33, 2009, p. 365-373 Figure 4: Images under the microscope of the appliqués (photograph by P. Penkova). very smooth surface with lack of traces of use; a view of the appliqués. Figure 4 : Images sous loupe binoculaire des appliques (photographies par P. Penkova). surfaces très lisses sans signe de traces d’usure ; vue des appliqués. added to the gold of object 4. he composition of the gold alloy (97% Au and 3% Ag) corresponds exactly to the socalled ‘tooth-alloy’ used by the dentists. he concentration of silver (2.5%), which could provide plasticity to the alloy, was not suicient, and, in some cases, cracks can be seen on the plates. hese aspects represent evidence for a low level of professionalism and a lack of knowledge regarding work with such types of material. he ratio Ag/Cu for objects 12, 13, and 14 is between 0.25-0.55. his aspect represents an additional argument that the alloy used for their production was prepared intentionally. At the same time, there is no analytical data concerning the chemical composition of Scythian gold objects from the period between the 5th and 3rd centuries BC available in the literature. he lack of such types of analyses makes a decision about the authenticity of the investigated appliqués very diicult, because the proposed origin of these objects is Scythia. Gold hracian appliques: authentic of fake? 371 casting technique (see Fig. 2: 2 upper). his result is in contradiction with the known hammering technique used by the craftsmen of the Late Iron Age (for general observations, see Treister, 2001). he use of a casting technique is most probably the reason why plots of faceted surface structures can be observed under the microscope on the pieces of the ‘Gartshinovo group’ (Fig. 1: 6-8). his fact and the intentional (most probably) coarse, barbaric appearance of these appliqués are indications of forgery. Using optical microscopy in most of the investigated objects with parallels in Scythia (No 2, 3, 13a-d, 14), evidence of tracks on the surface were found, indicating the use of a rolling-mill for the preparation of thin foil (see Fig. 1: 3; 2: 14-left). his aspect represents direct evidence for the forgery of the ofered appliqués. At the same time, according to the results of the investigation by optical microscope, no evidence of wear, or scratches due to usage, has been found on the surface of the objects (see Fig. 2: 4, 14 right). his represents yet another evidence for the forgery of the ofered appliqués. Some of the beads of the rows on the borders of plaques Nos. 3, 4 13a-d, and 14 look oddly amorphous, and hemispheres are connected on some spots with strips resembling traces of casting (Fig. 1: 3; 2: 4, 14-right). his is quite illogical, considering that such ornaments were performed with a punch by the ancient goldsmiths. Even if we accept that low quality models have been followed in the making of the artefacts in question, many of the basic details of the represented beasts – feathers, paws, even entire limbs, as well as the beak at No 13 – show an Figure 5: Image under the microscope showing lack of traces of use on the surface of the appliqués (photograph by P. Penkova). Figure 5 : Images sous loupe binoculaire illustrant l’absence de traces d’usure sur la surface des appliques (photographies par P. Penkova). Technological and traceological observations, archaeological parallels and evaluation of the results he analysis of a group of breast plates from ancient hrace belonging to the same period has shown that all the plates are produced from natural gold – the concentration of the copper in the plates is below 1.1%, and the Ag/Cu ratio is between 5.9 and 71.3. hese artefacts show clear evidence of hammering as a manufacturing technique, and many scratches on both surfaces (Kulef et al., 2009). According to the investigations carried out on the surface of some of the objects under consideration here with optical microscopy, it appears that they were produced using a Figure 6: Image under the microscope showing the surface of the appliqué which presented traces of casting (photograph by P. Penkova). Figure 6 : Images sous loupe binoculaire illustrant la surface des appliques présentant des traces de coulé (photographies par P. Penkova). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 365-373 372 iconography and a style which are too unconvincing. In the specialized literature, both on hracian and Scythian metalwork, it has been established that the artisans always dedicated special attention to a clear, comprehensible representation of the aforementioned animal details, regardless of the personal ability of the craftsmen. hus, the appliqués under consideration here convey the impression of a wanted coarseness, which, if corroborated with the contents of the metal, represents in our view an indication of a modern forgery. he only subgroup which might not consist of fakes is the one containing the appliqués with a representation of a ram’s head (Table 1, 1a-d; Fig. 1: 1a). Surface morphology observed under the microscope has shown the typical pattern of the use of one or more punches. hus, these artefacts can be evaluated as roughly performed replicas of the artefacts known from Scythia; this does not exclude the possibility of a forgery of better quality. Using optical microscopy on some of the investigated objects, some evidence of tracks was found on the surface, indicating the use of a rolling-mill for the preparation of thin foil (see Fig. 2: 14). hat is one piece of direct evidence for the forgery of the ofered appliqués. At the same time, according to the results of the investigation of the surface of the objects under an optical microscope, no evidence of scratches related to use were found (see Fig. 2). his represents yet another evidence for the forgery of the ofered appliqués. 4. CONCLUSION here is suicient evidence that the collection of 25 gold objects ofered to the National History Museum was produced from diferent materials (see Table 2). he grouping of the objects on the basis of similarity in terms of chemical composition generally corresponds to the typological classiication of the objects. he high purity of the gold used for producing the appliqués is an evidence for the use of reined gold. his result is in contradiction with the data obtained from the analysis of hracian breast plates (see Kulef et al., 2009) and Scythian authentic gold jewellery. hese authentic gold inds were usually produced using natural gold, and in many cases electrum – a natural gold alloy with a high content of silver (20-50%). Finds with such a high purity of gold are normally very rare (according to some evaluations, less than 0.15% of all Scythian gold objects are made from the purest gold.) he objects belong to the so-called ‘Gartshinovo group’ (Nos. 6 ÷ 10), that are also produced from pure gold (98ArcheoSciences, revue d’archéométrie, 33, 2009, p. 365-373 Ivelin KULEFF, Totko STOYANOV, Milena TONKOVA 99%). Some diferent textures of the gold, as well as amorphous relief and a very rough manufacture, were identiied. he evidence that these objects were produced by casting rather than hammering, which was the common practice in the period under consideration, are arguments that they are actually forgeries. At the same time, we were surprised by the coincidence of the golden appliqués with the matrix from Gartshinovo. his coincidence brings very serious doubt to the authentications of these appliqués. For other appliqués (Nr. 2 – head of Medusa; Nrs. 3 and 4, 11 ÷ 14 – representing animals ighting) with some Scythian parallels, there are insuicient arguments towards their authentication. At the same time, there is also a lack of data concerning the chemical composition and technology employed in the making of such types of parallels. In order to reach a more adequate conclusion regarding the authenticity of the appliqués, additional investigations are necessary – for example, destructive chemical investigations for the determination of the micro-quantities of certain elements. In spite of this, by using XRF for the determination of the base elemental composition, optical microscopy for observing the scratches on the surface resulting from use, and researching the literature for parallels of the investigated objects, it is possible to obtain objective data for providing some conclusions regarding the authenticity of the gold artefacts. Acknowledgements he authors are indebted to eng. Plamen Bonev for the ED-XRF analysis and to Petja Penkova for the optical microscopy investigation and observations of traces on the surface of the gold artefacts, both from the Laboratory for Conservation of the National Institute of Archaeology with Museum of the Bulgarian Academy of Sciences. he authors would like to thank Mrs. Elka Penkova from the National History Museum in Soia for her support with this investigation. References ARTAMONOV, M., 1970. Goldschatz der Skythen in der Eremitage. Hanau, W. Dausien. COJOCARU, V., BADICA, T. and POPESCU, I.V., 2003. Natural gold composition studied by proton activation analysis. Romanian Reports in Physics 55: 460-463. DAMYANOV, M., 1998. he Matrix from Garchinovo: Problems of Origin and Dating. Arcaeologia Bulgarica 2(2): 28-39. ECHT, R. (ed.), 2004. Die hraker: Das goldene Reich des Orpheus: Bonn, Kunst- und Ausstellungshalle der Bundesrepublik Gold hracian appliques: authentic of fake? Deutschland; Soia, Ministerium f̈r Kultur der Republik Bulgarien; Mainz am Rhein, Philipp von Zabern. GALANINA, L. and GRACH, N., 1986. Scythische Kunst: Altertümmer der scythischen Welt, Mitte der 7. bis zum 3. Jahrhundert v. u Z. Leningrad, Aurora-Kunstverlag. KOVACHEV, V., MAVRUDCHIEV, B. and YOSSIFOV, A., 2007a. Late Cretaceous and Palaeogene golden sources and their connection with magmatism and deep structure, in Proceedings of the International Scientiic-Technological Conference “Gold – he metal of all times”, Varna, 7-9 June 2007, National Technical Union – Union of Minig Geology and Metallurgy, 34-47 (in Bulgarian). KOVACHEV, V., STEFANOVA, V., NEDIALKOV, R. and MLADENOV, V., 2007b. Eluvial-alluvial gold from the gold-copper occurrence Borov Dol (R. Macedonia). Part I: Geochemistry of stream 373 sediments and their relation to the source rocks and ores. Review of the Bulgarian Geological Society 6: 66-76. KULEFF, I., TONKOVA, M. and STOYANOV, T., 2009. Chemical composition of gold breast plates from ancient hrace (5th-4th century BC). Archaeologia Bulgarica 13(2): 11-20. STEFANOVA, V., KOVACHEV, V., MLADENOV, V. and STANIMIROVA, TZ., 2007. Eluvial-alluvial gold from the gold-copper occurrence Borov Dol (R. Macedonia). Part II: Mineralogy of gold and stream sediments. Review of the Bulgarian Geological Society 6: 77-91. TREISTER, M., 2001. Hammering Tecniques in Greek and Roman Jewellery and Toreutics. Colloquia Pontica, vol. 8. Leiden, Boston, Köln, Brill. VENEDIKOV, I. and GERASSIMOV, T., 1979. hracian Art Treasures. Soia, Bulgarski Houdozhnik Publishing House. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 365-373 A unique 10th century AD gold-plated brooch from south-east Russia: technical and stylistic authentication Une broche unique plaquée or du Xe siècle apr. J.-C. provenant du Sud-Est de la Russie : authentification technique et stylistique Natasha Eniosova* Abstract: he gold-plated brooch with iligree and granulation was recovered by hobby metal-detecting in the vicinity of the city of Bryansk. he aim of this paper is to present some initial results representing the outcome of the archaeological and technological examination of this unique piece of jewellery. he composition of the metal and the method of fabrication were investigated using ED-XRF and microscopic analyses. he technical examination revealed that the brooch belongs to the Scandinavian stylistic and technological circle. However, technically it difers in three signiicant details from the Northern art tradition of iligree and granulation. First, the upper embossed panel, made of high purity gold, is placed on a solid silver base and ixed with six gold rivets. Usually, it was constructed of a lat based plate and a domed relief upper plate made of the same metal. he second unusual technique whose application was identiied on this brooch is the use of a mercury amalgam solder for joining the iligree and granulation to the substrate. he last peculiar detail is the use of geometrical granulation associated with Slavic art. Such elements appear on a number of Gotlandic ornaments from the late Viking Age. he gold brooch was possibly produced in Gotland based on the Scandinavian tradition, but also displays features of Slavic inluence in its design. Résumé : La broche plaquée or décorée de iligranes et de granulation a été trouvée avec un détecteur de métal dans les environs de la ville de Bryansk. Le but de cet article est de présenter les premiers résultats obtenus par examen archéologique et technique de cette pièce unique d’orfèvrerie. La composition du métal et les techniques de fabrication ont été étudiées par ED-FX et au moyen d’analyses microscopiques. L’examen technique révèle que la broche appartient à l’aire de circulation stylistique et technologique Scandinave. Néanmoins, du point de vue technique, trois détails signiicatifs diférent de l’art du iligrane et de la granulation du Nord. Premièrement, le panneau supérieur en repoussé fabriqué en or très pur est placé sur une base en argent massif et ixé à l’aide de six rivets en or. En général, ce panneau était constitué d’une plaque plate et d’une plaque supérieure bombée, en relief, réalisée dans le même métal. La deuxième technique inhabituelle trouvée dans la fabrication de la broche est l’utilisation de l’amalgame de mercure pour souder les iligranes et les granules au substrat. Le dernier détail particulier est l’utilisation de granulation géométrique associée à l’art Slave. Ces éléments apparaissent sur certains ornements de Gotland, de l’Age des Viking récente. La broche en or a pu avoir été produite en Gotland, à partir de la tradition Scandinave, mais sous inluence Slave en ce qui concerne le motif. Keywords: Filigree, gold jewellery, granulation, Viking Age. Mots-clés : Filigrane, orfèvrerie, granulation, Âge Viking. *Department of Archaeology, Faculty of History – Moscow State University. Lomonosovsky prospect, 27-4, 119992, Moscow, Russia. (eniosova@inbox.ru) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 375-380 376 1. INTRODUCTION he elaborately decorated gold ornament was recovered as part of an alleged hoard by hobby metal-detecting in the vicinity of the city of Bryansk, in the region between the Dnepr and Desna Rivers (Fig. 1). According to the collec- Natasha ENIOSOVA tor’s information, this alleged group consists of the brooch and Byzantine gold coins 1. he provenance of the brooch is not recorded, but it is of the type found throughout the Viking sphere of inluence. Brooches are the most common ornaments of the Viking period. heir function was to hold up a shawl, cloak or jacket over the dress (Hägg, 1971). Figure 1: Location map showing the city of Bryansk in the south-eastern part of Russia. he present city of Bryansk has yielded no traces of settlement dating to the Viking period. According to the chronicle, it was known since the 12th century AD. Figure 1 : Carte situant la ville de Bryansk dans le sud-est de la Russie. La ville actuelle de Bryansk ne montre pas de traces d’habitation de la période Viking. Selon les chroniques, la ville est connue depuis le XIIe siècle. 1. Constantine VII Porphyrogenitus with Romanus II (945-959). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 375-380 A unique 10th century AD gold-plated brooch from the south-east Russia… More than 4,000 brooches have been found in an area spanning from Iceland and Ireland in the West to the Western Siberian plain in the East (Jansson, 1985). Almost 300 genuine Scandinavian brooches originate from the territory of modern Russia, Ukraine and Belorussia, united in the Old Rus’ Kingdom in the course of the 10th century AD. A signiicant percentage of the brooches belonging to the oval, round, trefoil, annular, penannular and equal-armed types are made of copper alloys. Others are made of a gold, silver and leadtin alloy. Brooches made of gold represent the rarest group of Viking Age jewellery. Twelve gold disc-shaped brooches with iligree and granulation are already known from the Viking World (Eilbracht, 1999; Eniosova, 2007). he aim of this paper is to present some initial results representing the outcome of the archaeological and technological examination of this unique piece of jewellery in light of the distribution and use of gold in Scandinavia and Rus’ during the period between the 10th and the mid-11th century AD. 2. METHOD OF EXAMINATION he irst aims of the examination were to determine the composition of the metal of the various components and to ind out how they were made 2. he identiication of the nature of the metal was based on ED-XRF analyses, using an ARTAX (BRUKER AXS) instrument and software, itted with a Mo target and semiconductor detector. Typical analytical conditions were a tube voltage of 50 kV and a current of 700 µA. Each spectrum was recorded for 180 s. he sample regions have been analysed with a 0.2 mm collimator. Quantitative data were obtained for gold, silver, copper and mercury. he calculation of the weight concentration of elements without standards has been carried out by a method of fundamental parameters. he next stage was to investigate the components of which the brooch was constructed and how they were joined together. Microscopic examination of small-scale details was carried out using a Stemi 2000 stereomicroscope with digital camera and AxioVision (Karl Zeiss) software for a precise measuring of the delicate grains and wires. 377 gree and granulation and a solid silver back with a silver pin, pin attachment and catch-holder (Fig. 2). he technical examination of the brooch revealed that a relief of the upper panel was made from hammered sheet of gold with a positive die to form the main elements of the relief, with an animal pattern, where a pair of beasts is opposed in the mirror symmetry. he upper sheet was folding over a solid thick (2 mm) silver base. he silver base was cast in the clay bivalves mould made, possibly, by the same die impression. Two notches were cut for the pin attachment and one for the catch-holder in the bottom part of the mould. Both parts were attached with six gold rivets. Without radiography, we could not reveal what type of surface (smooth or relief ) was present under the gold plate. It is unlikely that the silver base of the brooch was used as a positive die for making its gold covering, because the silvercopper alloy (Ag – 80%; Cu – 20 %; HV – 45) is much softer than brass, gunmetal or tin-bronze, and would be easily damaged during its use as a patrix (Scott, 1991; Meeks and Holms, 1985). he pin attachment and catch-holder also show a complicated use of the silver base as a die. Highly stylized animals are shown in the front-facing part of the brooch, with round eyes and a nose made with a twisted wire (D 0.4 mm) and granules (D 1.2 mm). his particular type of animal ornament corresponds to the Viking Jellinge art style, dating to the 10th century AD (Jansson, 1991). he necks, bodies, heap and shoulders were ‘drawn’ by two-strand bands of beaded wires and illed by the gra- 3. RESULTS AND DISCUSSION he equal-armed brooch is 68 mm long and 45 grams in weight. It consists of an embossed upper panel with ili2. he examination was not complete because the brooch originated from a private collection. Figure 2. Front and back of gold equal-armed brooch from the Bryansk area. Figure 2 : Anvers et revers d’une broche à bras symétriques de la région de Bryansk. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 375-380 378 Natasha ENIOSOVA nules (D 0.36-1.25 mm). Beaded wires of three diferent sizes (D 0.7-0.95 mm) were made holding the swages at a right angle. Interlacing bands composed of beaded wire are characteristic of the Scandinavian art of iligree and granulation. his tradition is also characterised by the use of dies with embossed relief and irregular granules. However, the upper panel of the brooch is decorated with triangles of regular granules; the row of the animal teeth on the edge is depicted by triangles made of 3 granules (Fig. 3). Geometric patterns of regular granules, and the subordinate role of the iligree occupying the margins and rear, using beaded wire, are characteristic of Slavic ornaments from the 9th-10th centuries AD (Duczko, 1985). ED-XRF analyses indicate a high purity of the metal and a relatively low content of silver and copper (Au – 91.3%; Ag – 2.51%; Cu – 2.64%; Hg – 3.55%). he silver base was cast of silver alloyed with copper (Ag – 77.62%; Cu – 22.38%) and possible minute traces of lead and gold. here are no visible traces of solder on the upper plate, but the microprobe analysis shows that the base plate clearly contains more copper than either the tops of grains or the wire. We can thus put forth the hypothesis of a conventional solder alloy – that is, extra copper has been added to a gold alloy in order to reduce its melting temperature. However, a small quantity of mercury was detected on the surface of the upper sheet (Fig. 4). It could derive from the use of an amalgam solder. Gold-mercury solder recipes have survived in the medieval Mappae Clavicula manuscript, possibly dated to the 10th century (Ogden, 1994). Soldering experiments by F. Mishukov in Moscow and A. Minzulin in Kiev clearly demonstrated that the joining of granules of gold alloy to a substrate can be performed using a gold/silver alloy and mercury (1:6). he metallographic examination of the joint area shows an annealed structure, indicating that the solder was not solidiied from the molten state. In this case, it would have normally shown a cast structure (Mishukov, 1962; Minzhulin, 1990). Objects of gold are comparatively rare in Scandinavia, as well as in the Western and Eastern European countries where the Northern inluence was considerable. Certain information has been published on the composition of Viking Age gold. he list of compositions includes almost seventy items of gold jewellery, ingots and waiste (foil, wire, globules). Most Viking Age gold ranges in purity from 54% to about 99% gold (Table 1). However, high purity gold (Au > 90%) is much more common for ingots of the period. We can infer that a wide range of gold alloys was in use in the Viking period, and that there is no evidence of a progressive debasement of the metal during this time. Precious metal was reined or diluted by silver and copper depending on factors such as the availability of gold, or a customer or goldsmith’s approach to debasement (Ogden, 1994). Relatively ine gold Figure 3: (See colour plate) Detail of brooch, showing varied wire proiles and triangles of granules. Figure 3 : (Voir planche couleur) Détail de la broche, montrant diférents proils de ils et de triangles en granulation. Figure 4: (See colour plate) Spectra of the gold plate showing a notable concentration of mercury in the metallic solder. Figure 4 : (Voir planche couleur) Spectre X de la plaque en or montrant la présence d’une faible teneur en mercure dans la soudure métallique. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 375-380 A unique 10th century AD gold-plated brooch from the south-east Russia… Purity range (%) Items of jewellery Items of waist Ingots 99-100 - 2 3 91-98 11 18 5 80-90 6 7 1 54-79 9 7 - Table 1: Fineness of Viking period gold objects from Sweden and Great Britain (Oddy and Meyer, 1986); Northern Germany – Hedeby (Pernicka, 2002); Russia – Gnezdovo (Eniosova, 2007). Tableau 1 : Titre d’objets en or de la période Viking de Suède et du Royaume Uni (Oddy and Meyer, 1986); Allemagne du Nord – Hedeby (Pernicka, 2002); Russie – Gnezdovo (Eniosova, 2007). was used on occasion, and certainly more frequently than for manufacture of Gotlandic bracteates in the 7th and 8th centuries AD, which are mostly made of gold containing between 45% and 75% Au (Oddy and Meyer, 1986). 4. CONCLUSION he gold-plated brooch from Russia appears to be without parallels. We can compare its overall shape with a large group of Scandinavian equal-armed brooches made of brass or silver and dated to a period between the late 6th century and the late 10th century AD (Aagård, 1984). Stylistically and technologically, the upper relief panel of the brooch has much more in common with the Scandinavian iligree art dated to the Late Viking Age. Some distant parallels are the rectangular silver brooches from the Erikstorp hoard in Sweden, the gold spur from Röd, Östfold, Norway, and the gold embossed panels with iligree and granulation on the shells of the box-shaped brooches from Gotland (Duczko, 1995; Thunmark-Nylén, 1998). Box-shaped brooches belong to the large group of Gotland ornaments combining the Germanic animal style with Slavic geometric elements. hese specimens look entirely Scandinavian; geometric patterns play a subordinate role on these pieces of jewellery, typically occupying the margins. Outside Scandinavia, ornaments bearing Northern and Slavic features have been found in Estonia, Pskov and Kiev, and in the regions of the towns of Novgorod, Smolensk and Vladimir (Duczko, 1983). hey should be understood as objects made according to the Scandinavian tradition, but also showing Slavic inluence in their design. Considered all together, the evidence seems to suggest the local manufacturing of these obvious hybrids in Gotland. We can thus assume that the luxury gold brooch from Russia was produced by Gotlandic craftsmen. One could also put forth the counter-argument that ‘common 379 Scandinavian’ brooches were not used by the islanders. Other types of ornaments have been found in the richly furnished Gotlandic graves. However, judging from the direct manufacturing evidence, Gotlanders could also have produced jewellery for customers from the Scandinavian cultural area outside the island (Jansson, 1995). Being part of a wealthy Scandinavian woman’s dress, a gold-plated brooch travelled far to the East. Despite the considerable remoteness of the Bryansk region from the main trade routes used by the Vikings to reach the Black Sea or the Volga Bulgharia State, this particular area provides evidence for contacts between Scandinavians and the local Slavs in the late 10th-early 11th century AD. Evidence of such contacts comprises genuine Northern weapons and amulets (Shinakov, 1998). Acknowledgement I wish to thank Dr Robert Mitoyan, Department of Geochemistry of Moscow State University, for his kind assistance with the ED-XRF analysis of the Viking Age gold objects from Russia. References AAGÅRD, G-B., 1984. Gleicharmige Spangen, in G. Arwidsson (ed.), Birka II:1. Systematische Analysen der Gräberfunde. Stockholm, KVHAA, 95-110. DUCZKO, W., 1983. Slaviskt och gotländskt smide i älda metaller, in I. Jansson (ed.), Gutar och vikingar. Stockholm, Staten historiska Museum, 347-356. DUCZKO, W., 1985. Filigree and granulation work of the Viking Period: an analysis of the material from Björkö. Stockholm, Almqvist & Wiksell. DUCZKO, W., 1995. Kungar, thegnar, Tegnebyar, juveler och silverskatter. Om danskt inlytande i Sverige under senvikingatid. TOR. 27(2): 625-662. EILBRACHT, H., 1999. Filigran-und Granulationkunst im Wikingischen Norden Untersuchungen zum Transfer frühmittelalterlicher Gold-und Silberschmiedetechniken zwischen dem Kontinent und Nordeuropa, in W. Janssen et al. (eds.), Köln and Bonn, Zeitschrift für Archäologie des Mittelalters 11: 58-65. ENIOSOVA, N., 2007. Viking Age Gold from Old Rus’, in U. Fransson, M. Svedin, S. Bergerbrant, F. Androshchuk (eds.), Cultural interaction between east and west. Archaeology, artefacts and human contacts in Northern Europe. Stockholm, Stockholm University, 175-180. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 375-380 380 HÄGG, I., 1971. Mantel och Kjortel; vikingatidens dräkt. Fornvännen 66: 144-145. JANSSON, I., 1985. Ovala spännbucklor. En studie av vikingatida standard smycken med utgångpunkt från Björkö-fynden. Aun 7. Uppsala: Institutionen für arkeologi, 11-13. JANSSON I., 1991. År 970 och vikingatidens kronology, in M. Iversen (ed.), Mammen. Grav, kunst og samfund i vikingetid. Aarhus, Jysk Arkæologisk Selskab, i kommission hos Aarhus Universitetsforlag, 267-284. JANSSON I., 1995. Dress pins of East Baltic type made on Gotland, in I. Jansson (ed.), Archaeology East and West of the Baltic. Stockholm, Department of Archaeology, University of Stockholm, 83-91. MEEKS, N.D. and HOLMES, R., 1985. he Sutton Hoo garnet jewellery : an examination of some gold backing foil and a study of their possible manufacturing techniques. Anglo-Saxon Studies in Archaeology and History 4: 143-157. MINZHULIN, A., 1990. Soldering technology. Soviet Archaeology 4 : 236-237 (in Russian). MISHUKOV, F., 1962. Nevidimy pripoj juvelirov drevnosti. Voprosy dekorativnogo iskusstva. Trudy Moskovskogo vysshego chudozestvenno-promyshlennogo uchiliza. Moscow, 78-81 (in Russian). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 375-380 Natasha ENIOSOVA ODDY, W.A. and MEYER, V.E.G., 1986. he Analysis of the gold inds from Helgö and their relationship to other early medieval gold, in A. Lundström, H. Clarke (eds.), Excavations at Helgö X. Coins, Iron and Gold. Stockholm, Almqvist & Wiksell, 153173. OGDEN, J., 1994. he Technology of Medieval Jewelry, in D.A. Scott, J. Podany, B.B. Considine (eds.), Ancient and Historic Metals. Conservation and Scientiic Research. Marina del Rey, CA, Getty Conservation Institute, 153-182. PERNICKA, E., 2002. Röntgenluoreszenzanalyse der Goldobjekte von Haithabu, in Armbruster B., 2002. Goldschmiede in Haithabu – Ein Beitrag zum frühmittelalterlichen Metallhandwerk, in K. Schietzel (ed.), Das archäologische Fundmaterial VII. Berichte über die Ausgrabungen in Haithabu. Bericht 34. Neumünster, Wachholtz, 199-200. SCOTT, D.A., 1991. Metallography and microstructure of ancient and historic metals. Marina del Rey, CA, Getty Conservation Institute. SHINAKOV, E., 1998. Severnye elementy v kul’ture Srednego Podesenya, in Yanin (ed.), Istoricheskaya archeologia. Traditsii i perspektivy. Moscow, 307-323 (in Russian). THUNMARK-NYLÉN, L., 1998. Die Wikingerzeit Gotlands II. Stockholm, KVHAA, 70-71. R  C RestauRation et conseRvation Observations and questions on gold artefacts from underwater excavations he view of an archaeologist-restorer working in situ Observations et interrogations sur des objets en or provenant de fouilles archéologiques sous-marines Le point de vue d’un archéologue-restaurateur travaillant in situ Olivier Berger* Abstract: In the present paper, we will study examples of settlement excavations that can sometimes reveal abundant gold material. hese underwater excavations took place in Alexandria and Aboukir Bay, Egypt, where we found numerous gold or gold-plated objects. Of these, some objects provide us with information about the trade routes used in those times, and others show us that reinement existed even during the Byzantine period. Some objects tell us fabulous stories, while others attest a high mastery of the available technology. hrough speciic examples, we will approach the scientiic, aesthetic, historical and technological aspects pertaining to these gold objects. While most of these gold objects, jewellery and coins, have been studied before, and the results published elsewhere, considering the number of excavated objects, it is typically the less spectacular ones or the ones that were too diicult to interpret that have been neglected. Being an archaeologist specialised in the restoration of metal, I am thus not a metallurgist, a technologist, or even a numismatist, nor a historian specialised in the production or trade of gold objects. However, on the boat, underwater, or in the land laboratory, many questions concerning these archaeological objects deserve attention. With the help of three case studies, we will see that some problems related to these objects have come up, and that a close collaboration with various gold artefact specialists appears necessary in order to try to solve these uncertainties. Résumé : Seront abordés des exemples de fouilles d’habitats qui parfois peuvent livrer un abondant matériel en or. Ces fouilles sous-marines, sont menées en Egypte, à Alexandrie et Aboukir. Lors de nos campagnes de fouilles, nous avons trouvé de nombreux objets en or ou recouverts d’or. Certains objets nous parlent des voies commerciales, certains nous démontrent un rainement d’une période dite décadente. D’autres objets nous racontent de fabuleuses histoires, d’autres encore sont des preuves d’une haute maîtrise technologique. Au travers de quelques découvertes, nous allons approcher l’aspect scientiique, esthétique, historique et technologique de ces objets d’or. Si la majorité des objets, monnaies ou bijoux, sont étudiés et publiés étant donné la masse de documents exhumés, il en est de moins spectaculaires et de plus délicats à interpréter qui sont délaissés. Etant un archéologue spécialisé dans la restauration du métal, je ne suis donc ni analyste, ni métallurgiste, ni technologue, ni numismate, ni non plus un historien spécialiste de la production, du commerce ou de la circulation des objets en or, et pourtant, que de questions se posent face à ces objets archéologiques trouvés lors de fouilles et restaurés in situ. A travers trois cas, sera entrevue une base d’étude où seule une collaboration pluridisciplinaire pourrait répondre aux problématiques posées. Keywords: FTIR, Egypt, goldsmith, gold-working, metal restoration, underwater archaeology. Mots-clés : FTIR, Égypte, orfèvre, travail de l’or, restauration du métal, archéologie sous-marine. *Conservateur-restaurateur au musée des Antiquités de Bâle – 5, St. Alban-Graben, 4051 Bâle, Suisse. (bergoliv@hotmail.com) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 383-387 384 Olivier BERGER 1. INTRODUCTION he methods employed by the ancient Egyptian goldsmiths have been covered in some detail elsewhere, although there is still a need for a far more comprehensive study (Ogden, 2000: 164). The European Institute of Underwater Archaeology (Institut européen d’archéologie sous-marine, IEASM) led by Franck Goddio, has been carrying out excavations in Egypt for more than ten years. We are working in the Nile Delta, and more precisely in Alexandria and Aboukir Bay, where, under the sea, we found the remnants of the sunken cities of Canopus and Heracleion. More than 15,000 artefacts were discovered at these sites, and, among this abundant material, we mainly found stone work. However, we also found objects made of organic material, ceramic, glass, and diferent metals. Some rare objects made of silver and those made of iron generally disappear in the sea water environment, but bronze, lead and gold objects are frequently discovered. What is unusual about these sites is that they were abandoned because of several natural cataclysms and not reoccupied during the last centuries. hat is why we had the chance to ind many intact artefacts, such as ritual dishes, which are rather infrequent in other excavations. he corpus of gold artefacts is mainly represented by jewellery and gold coins, originating mostly from the Ptolemaic and Byzantine periods. he jewellery is relatively standard, and it includes rings, earrings, beads, pendants, and pendants with some gem stones (amethyst, emerald, sapphire, garnets), glass, or pearls. Several rings are delicate and some are made of massive beaded wires. Some of them are without decoration or with just one gem, but some others are inely decorated and show scenes of everyday life. hrough speciic examples, we will approach the scientiic, aesthetic, historical and technological aspects pertaining to these gold objects. 2. METHODS § 1 Some rings show interesting stories, like a romantic Byzantine wedding ring originating from Canopus and dating from the 6th or 7th century AD (Fig. 1). he translation of the Greek inscription engraved inside the ring reads “Peace, which is mine, I give it to you. Amen”. his citation is from the Gospel of John (14, 27). here are, interestingly, two orthographic mistakes in this sentence. he face of the ring bears the inscription OMONOIA, which means ‘union’ (Lat. ‘concordia’), as was traditionally ArcheoSciences, revue d’archéométrie, 33, 2009, p. 383-387 Figure 1: Wedding ring and ring with a bezel in the shape of an oil lamp. Figure 1 : Bague de mariage et bague avec un chaton en forme de lampe à huile. written on wedding rings at that time. OMONOIA was an inscription chosen by Christians in reference to the irst word of verse 16 of chapter 25 in the book of Sirach, which deals with marital and family relationships: “he Union of brothers, the friendship of neighbours, and wife and husband living in perfect harmony” (Tardieu, 2006). he Byzantine period is nicely represented by a ine handicraft ring with a bezel in the shape of an oil lamp (Fig. 1). he details are so perfectly made that the top lid can still be moved. Found on the site of Canopus, this ring is dated to a period between the end of the 6th and the beginning of the 7th century AD. his type of ring seems unique, as we could not identify any parallels to it, and only its decorative waves and vegetal ornament can be compared to those of the ring found in the treasure of Pantalica, Sicily (Stolz, 2006). An important discovery was made at the site of Heracleion in 2003, when we found a gold plaque with inscriptions. Unfortunately, it was not in its original position but at the bottom of a canal. he object is 5 cm high, 10.8 cm wide and 0.8 cm thick. Written in Greek, this gold plaque was Observations and questions on gold artefacts from underwater excavations made to celebrate the construction of a monument (gymnasium) by Ptolemy III. In 1818, a 6 by 6 cm gold blade was found in Aboukir; it is also engraved in Greek, and it is currently in the British Museum (N° 1063). he plaque bears the dedication of a Temenos to Osiris in the name of Ptolemy III and Berenice II (Yoyotte, 2006; Clauss, 2006). Sometimes, gold objects can be of help in understanding the trade between Egypt and other countries. he Hemistater of Pumiyaton is an artefact originating from Cyprus, with a naked walking Heracles wearing only his traditional lion skin. It dates from the 4th century BC. An unusual aspect of the representation in this object is the fact that Heracles is beardless. he reason for this is that the Cypriot Herakless-Melquart is not heavily bearded, like the classical one. On the reverse, a Phoenician inscription with the name of the king Pumiyaton dates the issue to the seventh or eighth year of his reign (355/353 BC) (Yoyotte, 2006; Lichocka, 2006). An object that is very interesting from a technological point of view is a gold eye of Horus (Wedjat) found in Heracleion (Fig. 2). his bead, hollow inside, is made of two gold blades and a strip of gold sheet used for the sides (Stolz, 2006). We can put forth a hypothesis regarding the technique used to make the pupil, using a lat sphere, iligree and granulation for the decoration. his object is really small, measures 8 mm and weighs 0.4 g; it is less than 2 mm thick. It makes us wonder how the ancient craftsmen were able to weld all those small pieces together, and to produce such tiny details. Usually, this type of jewellery is embossed or moulded, and it is quite unusual to identify the use of iligree and granulation, indicating the intricate work involved in the production of such a small object. Figure 2: Gold eye of Horus (Wedjat). Figure 2 : Œil d’Horus (Oudjat). 385 3. METHODS § 2 On the boat, underwater or in the land laboratory, many questions concerning archaeological objects arise, and even a simple object can provide signiicant information, but also bring about numerous other questions. Case 1 – he most interesting questions that arise from our research concern a carved limestone deposit holding more than one kilogram of gold leaves. A hard black substance was found among the gold leaves. After microchemistry tests carried out at the Expert Center laboratories in Lausanne, we established that it was an organic substance, and after analyses with FTIR (Fourier Transform Infrared Spectrometry) and GC-MS (Gas Chromatography – Mass Spectrometry), we were able to determine that it was a natural resin like rosin with what appear to be traces of perfume rosin, like myrrh or aloe vera. he gold foil does not have a regular size, but displays a constant layer thickness of 12.3 ± 0.9 µm, almost comparable to chocolate paper. he examination of the gold composition by XRF (X-ray Fluorescence Spectrometry) and AAS (Atomic Absorption Spectrometry), carried out at the research laboratory of the Swiss National Museum, provided us with the composition of the metal. he gold contains silver, platinum and iron, but no copper (Table 1). Does this information allow us to infer that it was used in a temple as an ofering to the gods, with gold and perfume resin, or that it was some type of ofering deposited in the foundation wall during the construction of an important building, like a foundation deposit? Is it possible that it was hidden by a goldsmith, and that the metal was meant to be melted again, or that it was a sort of kit for gilding, with the resin used as glue? As mentioned in the literature (Higgins, 1961; Pliny the Elder, 1983), it was not unusual to ind this thickness for a gold leaf, but the question still remains concerning the kind of material that this type of gold leaf was meant to gild: stone, metal or wood? Case 2 – Gold is stable even underwater; however, in land excavations, the fragments of gold artefacts or even small objects are mixed with sediment. In such cases, gold starts to tarnish and we cannot easily distinguish it from other materials or sediments. On the other hand, gold is not altered in sea water; in Nile clay, in sand or other sediments, gold preserves its shiny metal relection. he objects are so shiny and dense that they can be seen immediately in the sediment or in the sifter discarded by the water dredge. hat is why, in the context of our work, we found a considerable amount of small pieces of gold and elements of goldsmithing, a less common phenomenon in land excavations. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 383-387 Olivier BERGER 386 Denomination Gold sample % Ni 0.0007 % Sn <0.0010 % Bi <0.0010 % Co <0.0010 % Pb 0.0006 % Zn <0.0010 % As <0.0010 % Fe 0.0176 % Sb <0.0010 % Cu 0.0025 % Pt 0.0202 % Ti <0.0020 % Ag 0.1741 *% Au 99.7843 *% Au: he gold composition has been calculated (100% – total composition of trace elements) Table 1: Composition of the gold leaves. Tableau 1 : Composition des feuilles d’or. We can focus on two small representative deposits of gold fragments, one containing 63 gold pieces, and the other one 42 fragments. Having a closer look at the fragments, we can observe that they represent small pieces of objects, mostly jewellery. We can thus recognise fragments of objects, palmette-shaped appliqués, beads of diferent sizes and shapes, chain links, ornaments, earrings, parts of necklaces, bracelets, annuli, granules, pendants, wires, etc. he question that arises is: did they break naturally, were they there to be melted, or were they intentionally broken to be ofered to the gods and then spread onto the land? In the area of one temple, we found many small factice oferings of lead coins, thousands of them. his inding conveys perhaps a similar phenomenon: instead of ofering jewellery, ancient people sometimes preferred to ofer a large amount of precious metal – but in small fragments. sculpture has a gold blade inlay for the eyes and is overlaid with extremely thin gold leaves on the body. he antique surface was polished without any intentional engraving. It seems that there was a resin or a drying oil between the metal and the gold leaves. In seawater, an organic material like resin can be well preserved. he analyses were performed to see if we could recognise what type of medium they used to attach the gold leaves. his method has been attested in Egypt since the third millennium BC (Raub, 1993). Unfortunately, the result was negative and we were not able Case 3 – Very often, we see that Egyptian archaeological bronze sculptures have been gilded and we can observe traces of gold or technological details employed in the gilding preparation. A crown found in Heracleion, from a life-size statue, has a roughened surface with a lot of cross hatchings made with gravers or burins; however, we could not ind any trace of gold on it. It is not rare in Egypt to ind bronze sculptures overlaid with gold, using the technique of roughening by stippling or chiselling the surface in order to facilitate the adhesion of the gold leaf to the copper alloy (Oddy et al., 1988). his strong and thick gilding requires diferent layers of gold leaves to cover the cross hatching. In some cases, a mineral preparation is carried out between the two metals (Oddy, 1981). On the other hand, some objects are perfectly polished and gilded, and in many cases we can still observe the gold leaves. he ire gilding, using mercury-gold amalgam, was unknown prior to the Ptolemaic period (Ogden, 2000); other methods were used instead. In the research laboratory of the Swiss National Museum, we analysed a Mummiform Osiris found in Heracleion. his ArcheoSciences, revue d’archéométrie, 33, 2009, p. 383-387 Figure 3: Deposit of gold fragments. Figure 3 : Dépôt de fragments d’or. Observations and questions on gold artefacts from underwater excavations to ind any organic component, only gold and products from the corrosion of the bronze. Can this be an indication that the gold leaves can stick directly to the metal without any preparation, just with heat and strongly burnished? As we know from literary sources, this technique seems to have been attested in Egypt since the second millennium BC (Oddy, 1982). However, specialists do not agree on this subject (Ogden, 2000). he ancient artists were able to cast a life-size sculpture in bronze, to polish its surface, to chisel all the details, and, after they spent a signiicant number of hours only to scratch, to hatch and engrave the nice skin of the sculpture, they would disigure it for the only purpose of having a terribly roughened surface. his method, entailing creation followed by destruction, must have had a higher purpose. However, if they were able to simply apply the gold leaf directly on the polished metal surface, with a very nice result and while using less gold, being faster and requiring less preparation, the question is why, in the same period, in the same place, on the same bronze support, for the same object, would they have used diferent methods for gilding? On an object like an Osiris, they used two diferent types of gold: one for gilding with very thin gold leaves, the other, using thick gold blades, for inlaying the eyes. hus, we can observe, on the same object, the use of gold blade and gold leaves without resin, and a roughened surface, carried out during the same period, at the same archaeological site, and even in the same temple. hen, the question of why so many diferent techniques were used remains open. 4. CONCLUSION As archaeologists, we do not ind many gold objects on our excavations; as restorers specialised in metal, we do not have to perform many interventions on stable gold objects, but, however, we come across many questions and there is so much information to obtain from these small precious objects. Sometimes, as restorers, we enjoy the privilege of having access to some technological details and useful information. We spend so many hours working on and with the objects that we come to know them very closely. For example, we can observe gold traces in the corrosion products which will disappear during the restoration treatment. However, if we have this privilege, we also have the responsibility to describe and to document our observations. Most of the time we do our best, but we are not specialized in archaeological gold objects and we sometimes overlook certain technological knowledge pertaining to them. We lack reference articles 387 and one cannot help but notice that we are in need of a more close collaboration with specialists and experts. Acknowledgements Hildebrand Erwin, Chemistry scientist, Swiss National Museum. Hunger Katja, Chemistry scientist, Swiss National Museum. Chartier-Raymond Maryvonne, Egyptologist. Samson Guylaine, Translation. Soares-Woerle Marie, Head of the research laboratory of the Swiss National Museum. WichgersDraganski Anka, Dr. in Archaeology. References HIGGINS, R.A., 1961. Greek and Roman Jewellery, London: Methuen. CLAUSS, M., 2006. Gold for the house of gods. Ptolemy III foundation plaque, in Franck Goddio, Manfred Clauss (eds.), Egypt’s Sunken Treasures, Munich and London, Prestel, 140. LICHOCKA, B., 2006. Cypriot hemistater, in Franck Goddio, Manfred Clauss (eds.), Egypt’s Sunken Treasures, Munich and London, Prestel, 351. ODDY, W.A., 1981. Gilding through the Ages, Gold Bulletin 14: 75-79. ODDY, W.A., 1982. Gold in Antiquity: aspects of gilding and of assaying, Journal of the Royal Society of Arts, 5315 (130): 730-743. ODDY, W.A., PEARCE, P. and GREEN, L., 1988. An unusual gilding technique on some Egyptian bronzes, in S.C. Watkins and C.E. Brown (eds.), Conservation of Ancient Egyptian Materials, Conference Proceedings, Bristol, 15-16 December 1988, London, United Kingdom Institute for Conservation, Archaeology Section, 35-39. OGDEN, J., 2000. Metals, in P.T. Nicholson, I. Shaw (eds.), Ancient Egyptian Materials and Technology, Cambridge, Cambridge University Press, 148-176. PLINE L’ANCIEN, [1983]. Histoire naturelle, livre XXXIII, chap. XIX, §61. RAUB, C., 1993. How to coat objects with gold, in C. Eluère (ed.), Outils et Ateliers d’Orfèvres des Temps anciens, Saint-Germainen-Laye, Antiquités Nationales, mémoire 2, 101-110. STOLZ, Y., 2006. Styles in early byzantine gold jewellery and the bead in the shape of an eye of hours or wedjat, in Franck Goddio, Manfred Clauss (eds.), Egypt’s Sunken Treasures, Munich and London, Prestel, 212 and 316. TARDIEU, M., 2006. Wedding ring, in Franck Goddio, Manfred Clauss (eds.), Egypt’s Sunken Treasures, Munich and London, Prestel, 300. YOYOTTE, J., 2006. Herakles, a man of mighty deeds, forever questing for proofs of his courage and Ptolemy III foundation plaque, in Franck Goddio, Manfred Clauss (eds.), Egypt’s Sunken Treasures, Munich and London, Prestel, 72 and 140. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 383-387 Treasures of Portuguese archaeology Notes towards a preventive conservation project Trésors de l’Archéologie portugaise – notes pour un projet de conservation préventive Isabel Tissot*, Matthias Tissot*, Pedro Pedroso* and Luís Raposo** Abstract: he permanent exhibition “Treasures of Portuguese Archaeology” of the Portuguese National Archaeological Museum is composed mainly of gold and silver artefacts which present extensive surface corrosion. he construction of the exhibition room dates from the late 1980s, and it was conceived with the highest security standards in mind. However, at that time, the importance of the choice of compatible materials and equipments for the exhibition of metal artefacts was still unknown, and the same was true for the inluence in terms of corrosion development, even for gold objects, if an inappropriate material was selected. Preventive conservation directives and actions have been developed since then, and presently one is confronted with new paradigms for the exhibition and preservation of artefacts, and, moreover, of cultural heritage. A preventive conservation project, which is currently in its initial stages, was set up with the objective of identifying and understanding the pathologies observed, in order to correct the problems presented by the exhibition room. his article presents an overall view of the project, a description of the alterations observed, and a summary of the methods proposed for obtaining responses to the respective problems, with the purpose of correcting the room’s fragilities. Resumé : L’exposition permanente « Les trésors de l’archéologie portugaise» du Musée National d’Archéologie est essentiellement composée de pièces en or et en argent qui présentent une surface de corrosion étendue. La salle d’exposition date de la in des années 80 et a été conçue en respectant les normes restreintes de sécurité. Néanmoins, à cette époque, l’importance du choix de matériaux de construction et d’exposition des objets en métal était encore inconnue ainsi que l’inluence, si le choix est incorrect, que ces matériaux exercent sur le développement de mécanismes de corrosion, même dans le cas d’objets en or. Nous avons assisté depuis à de nouveaux développements et directives concernant la conservation préventive et aujourd’hui nous sommes confrontés aux nouveaux paradigmes concernant l’exposition et la préservation des objets de notre patrimoine culturel. Il a été mis en place un projet de conservation préventive pour identiier et comprendre les pathologies observées, ain de corriger les problèmes de la salle d’exposition. L’article présente un aperçu de l’ensemble du projet, la description des altérations observées et les méthodes proposées pour obtenir les réponses aux problèmes observés ain de corriger ces erreurs. Keywords: Ancient jewellery, gold and silver alloys alteration, preventive conservation, exhibition materials. Mots-clés : orfèvrerie archaïque, alliages en or et en argent, conservation préventive, matériaux d’exposition. 1. INTRODUCTION he Portuguese National Archaeological Museum (MNA) possesses in its collections a remarkable number of ancient jewellery objects, either originating from excavations, or, more frequently, bought either from goldsmiths or the people who found them. his collection, due to its representative quality, allows a general overview of the evolution of this art in the present Portuguese territory, from the begin- * Archeofactu – Rua do Cerrado das Oliveiras, nº14 2ºDto. 2610-035 Amadora, Portugal. (isabel.tissot@archeofactu.pt) ** Museu Nacional de Arqueologia – Praça do Império, 1400-206 Lisboa, Portugal. (mnarq.info@imc-ip.pt) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 389-392 390 nings of metallurgy until the Early Middle Ages (Parreira and Vaz Pinto, 1980). he collection is on display in the “Treasures of Portuguese Archaeology” permanent exhibition, inaugurated in 1980. he exhibition room was conceived and built according to the highest security standards, relecting the value of the collection, consisting mainly of gold artefacts (circa 700). Even today, the concept of value is deeply associated with the erroneous idea of the inalterability of gold artefacts; however, most of them are composed of gold alloys, in which the minor elements can corrode. Also, 20 years ago, preventive conservation was still in its early development stage; concerns related to the exhibition of objects and collections were focused on diferent issues, like security. he choice of stable and compatible materials for showcases was not a irst priority, because it was still largely unknown. In contrast, today, museum professionals possess at least a basic understanding of preventive conservation, even if, as a new conservation ield, it is still evolving and searching for models (Cassar, 2000). his evolution relects the interesting paradigm pertaining to preventive conservation parameters related to the exhibition of objects, and, in this particular case, of gold objects. Isabel TISSOT et al. the 15th century and is located 250 m north of the Tagus river and 8 km away from the river mouth. In 1980, in the east gallery, the Treasury Room was built, as a structure of aluminium, glass, wood, and wooden shavings. he walls and the showcases were covered by tissues and carpeting on the loor. he room has air conditioning equipment that regulates only the temperature at 20 ºC, and the relative humidity (RH), varies during the year between 40 and 80%, with an average value of 65%. he exhibition still has a current context, notwithstanding the fact that it presents a maladjusted and outdated aesthetic (Fig. 2). In 2000, a small room renewal was performed, consisting of the replacement of the showcase fabrics and of the room carpet, as they displayed colour alterations and deterioration. At that time, no structural changes were considered, and no adequate material analytical study was carried out. Nevertheless, some attention was given to the selection of materials, in an attempt to make them compatible with the ones from the objects themselves; more speciically, the application of fabric composed of pure cotton was planned. During this change, a conservation treatment of the artefacts was performed. 2. THE TREASURY ROOM – A SECURE ROOM TO PROTECT A PRECIOUS COLLECTION he National Archaeological Museum is situated at the Virgin of Belém Monastery, generally known as Jerónimos Monastery, in Lisbon, occupying an area of almost 200 m, corresponding to the original dormitory of the monastery (Fig. 1). he stone foundation of the monastery dates from Figure 1: he Portuguese National Archaeological Museum. Figure 1: Le musée ational d’Archéologie du Portugal. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 389-392 Figure 2: (See colour plate) Showcase from the exhibition “Treasures of Portuguese Archaeology” of the Portuguese National Archaeological Museum. Figure 2 : (Voir planche couleur) Les vitrines de l’exposition « Trésors de l’archéologie Portugaise » du musée National d’Archéologie. Treasures of Portuguese Archaeology. Notes towards a preventive conservation project Only a few years later, a signiicant alteration of the surface of the metal objects could be seen (Fig. 3). he silver alloy artefacts presented a tarnished surface with a variation between yellow, brown, violet, blue and black blue, characteristic of silver sulphide, the main component of silver atmospheric corrosion (Selwyn, 1990). he surface of some of the gold alloy objects presented a rose coloration in some areas, also with a multihued iridescent efect. Although representing, apparently, a random efect, the alteration tended to appear to a more signiicant extent in areas of issures, in surfaces that were in contact with the showcase fabrics, and in the objects with a silver content in the range of 20%, based on the spectrographic analyses presented in the studies carried out by Hartmann (1970; 1982), the only analytic studies performed until today on the objects from the collection. he alteration that was observed could be a result of the formation of silver sulphide from the silver component in the alloys (Járó and Tóth, 1997) and the sulphide source originating from the materials of the showcases. In order to understand the causes of this deterioration, a simple corrosiveness test, an accelerated corrosion test, based on the Oddy Tests (Hatchield, 2002; Tétreault, 1999) was carried out in order to assess the efects of the materials from the exhibition room on the artefacts. hese tests are merely indicative of whether the materials can be deleterious to the metals exposed. he test consisted in placing the materials – tissue, wood, wood shavings and carpet – in separate reaction vessels, some containing pure silver, and others pure copper test coupons (20 x 10 x 0.05 mm), maintained at 60 ºC with 100% RH over a 30 days period. At the end of this period, the metal coupons were assessed for visual alterations; the possible hazards are indicated by the severity of corrosion on the test coupons. From the results obtained, it is practically impossible to distinguish the slight alterations in the copper test coupons, indicating there was no severe corrosion. he silver coupons showed no alteration when placed together with the carpet, wood and wood shaving materials, but presented a severe tarnishing in the coupon tested with the tissue. hese results suggest that, although pure cotton tissue was considered an acceptable material for the exhibition, there is a problem with this material, which can be related to a dye or a fabric surface treatment, further testing being necessary to clarify this. Notwithstanding the fact that the tissue was identiied as one of the causes for the accelerated development of the corrosion processes registered on the artefacts, the deicient showcase system, which allows particle entrance and accumulation, should also be taken into consideration. 391 Figure 3: (See colour plate) Detail from a torc illustrating the corrosion of the surface. Figure 3 : (Voir planche couleur) Détail d’un torque illustrant la corrosion sur la surface. In light of these irst tests, and with the aim of extending further the study of the Treasury Room, the MNA decided to put together a project involving the study of the material, not only in view of the conservation issues concerned, but also of the collection as an assemble. his includes a detailed study of the materials and equipment from the current exhibition room, and air quality analyses, allowing an assessment of the characteristics of the materials as potential deterioration sources, through their correlation with the pathologies observed. he aim is to apply the preventive conservation guidelines for an exhibition site in order to guarantee the conservation and appraisal of the collection. his project will also consider the diagnosis of conservation conditions and the treatment of the objects on display, c. 700, as well as a further remodelling of the room, accompanied by the development of a preventive conservation programme that will follow the evolution of the objects in the remodelled room in the future. To this purpose, a team of specialists was assembled, where the synergies of various disciplines, including archaeology, preventive conservation, restoration and conservation, chemistry and physics, can be efectively targeted in an interdisciplinary endeavour to achieve the goals of such a project of preventive conservation. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 389-392 392 3. CONCLUSION Today, the conception, organization, selection of materials and showcases for cultural heritage exhibitions is a major and complex issue which requires detailed study, although some research guidelines already exist (Tétreault, 1994; Hatchield, 2000), and experimental rules for standardizing procedures in collections conservation have started to emerge (Borges de Sousa et al., 2007; Afnor, 2007). hese new trends in conservation can be associated with the most recent paradigms regarding preventive conservation parameters (Pereira, 2007), valuing the collection beyond its material value, as a cultural identity. his is an aspect which undoubtedly represents an inlection of the concept of heritage, which establishes cultural heritage and the assumption of preventive conservation as parts of the concept of the sustainability of objects and collections. he principal aim of sustainability is to achieve an acceptable quality of life for the population of the world, combined with the economic growth of communities, without depleting or damaging the natural resources needed to sustain future generations. his concept can be applied to the present global sense of preventive conservation and its relationship with cultural heritage (Cassar, 2000). Presently, in order to enhance the value of the collection, the MNA is planning a study aiming to identify, understand and correct the fragilities of the Treasury Room, ensuring the conservation of objects in the long term, and involving the new paradigms of conservation following the statement: ‘Cultural heritage as a contributing factor to human development’. References ASSOCIATION FRANÇAISE DE NORMALISATION, 2007. Normes expérimentale AFNOR XP X80-002 décembre 2007: Conservation ArcheoSciences, revue d’archéométrie, 33, 2009, p. 389-392 Isabel TISSOT et al. de biens culturels – Recommandations pour concevoir, aménager, choisir et utilise rune vitrine d’exposition des biens culturels. Saint-Denis-la-Plaine, AFNOR. BORGES DE SOUSA, C., AMARAL, J. and TISSOT, M., 2007. Plano de Conservação Preventiva Bases orientadoras, normas e procedimentos. Lisboa, Instituto dos Museus e da Conservação. CASSAR, M., 2000. Value of Preventive Conservation. European Preventive Conservation Meeting. Vantaa, Institute of Art and Design. JARÓ, M. and TÓTH, A., 1997. Deterioration of metal threads and other metallic decorations made of gold, silver, or gilt silver on Museum textiles – problems of their conservation, in Metal ’95: Proceedings of the International Conference on Metals Conservation. London, James & James, 201-208. HARTMANN, A., 1970. Prähistorische Goldfunde aus Europa. Studien zu den A,fängen der Metallurgie 3:1. Berlin, Mann. HARTMANN, A., 1982. Prähistorische Goldfunde aus Europa. Studien zu den Anfängen der Metallurgie 3:2. Berlin, Mann. HATCHFIELD, P., 2002. Pollutants in the Museum Environment: Practical Strategies for Problem Solving in Design, Exhibition and Storage. London, Archetype. PARREIRA, R. and VAZ PINTO, C., 1980. Tesouro da Arqueologia Portuguesa no Museu Nacional de Arqueologia e Etnologia. Lisboa, Secretaria de Estado da Cultura, Instituto Português do Património Cultural. PEREIRA, H., 2007. Contemporary trends in conservation: culturalization, signiicance and sustainability. City & Time 3(2): 15-25. SELWYN, L., 1990. Historical silver: storage, display and tarnish removal. Journal of the International Institute for Conservation – Canadian Group (IIC – GC) 15: 12-22. TÉTREAULT, J., 1994. Display Materials: he Good, the Bad and the Ugly. Edinburgh, Scottish Society for Conservation and Restoration (SSCR) Exhibition and Conservation Preprint. TÉTREAULT, J., 1999. Revêtements pour l’exposition et la mise en réserve dans les musées. Bulletin Technique nº 21. Ottawa, Institut Canadien de Conservation. Ancient gold patinas: experimental reconstruction Les patines anciennes de l’or : recréation expérimentale Alessandro Pacini* Abstract: he present study is based on the experimental reconstruction of samples of diferent carat gold foil, obtained by hand hammering, and subsequently subjected to diferent physical-chemical treatments with the aim of reproducing ancient patinas. he treated samples were then analysed by SEM-EDS in order to provide a valid comparison with the ancient goldsmiths’ techniques. Résumé : Ce travail se base sur la recréation expérimentale d’échantillons de feuilles d’or réalisées par martelage à la main et dont la composition en carats est variée. Ces feuilles sont ensuite soumises à divers traitements physico-chimiques pour reproduire des patines anciennes. Les échantillons subissant ces traitements ont été analysés par MEB-EDS de façon à obtenir une comparaison valide avec les techniques orfèvres anciennes. Keywords: Patina, gold patination, colouring, giallone, gilding Mots-clés : patine, patine d’or, coloration, giallone, dorure. 1. INTRODUCTION In the ield of the recognition of fakes in jewellery, the laboratory An.te.a. deals with the typical patinas of archaeological jewellery. In general, such objects can be recognized by four typical patinas: Natural patinas due to the contact of the jewellery with earth, water or air. Non-intentional patinas from contact with substances associated with the life of the jewellery (blood, perfume, etc.). Restoration patinas (due to the process of cleaning). Artiicial patinas with the aim of forgery. he major part of jewellery patinas are caused by liquation and alteration of the copper and silver contained in the alloy and/or supericial deposits of corrosive products, soil, or decomposed organic material. Rarely, in particular conditions related to the position in which the object is deposited, alterations to the gold itself can also be veriied, caused by the formation of gold cyanide or complex compounds with chlorine, (Grimwade, 1999; Mazzeo, 2005) which result in an orangey-brown patina. he colours and efects of ancient jewellery patinas have always been the subject of imitation on the part of the jewellers who, in various eras, have tried to reproduce archaeological jewellery. In order to better identify these artiicial patinas, we carried out a reproduction of patination via chemical processes on some gold samples, based on old laboratory recipes. he results were then submitted to an SEM-EDS analysis by Daniela Ferro at the Institute for the Study of Nanostructured Materials (CNR-ISMN) in Rome. 2. EXPERIMENTATION he samples were made up of sheets 1 cm wide and 0.1 mm thick, obtained by hammering (M) and by a rolling mill (L). hree types of alloys similar to the ancient ones * An.te.a. – Laboratory of archaeometry and experimental archaeology. (ale.pacini@gmail.com) rec. Sept. 2009 ; acc. nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 393-396 394 have been used: Alloy A = Au 95%, Ag 4%, Cu 1% ; Alloy B = Au 93%, Ag 5%, Cu 2%; Alloy C = Au 90%, Ag 7%, Cu 3%. he samples are classiied by: letter of the alloy (A, B or C), type of work (M or L), and identiication number (1, 2 or 3). Some gold samples have been buried for 30 days at 16 °C in soil rich in organic substances, moistened with water, ammonia and blood, and submitted every seven days to a heat of 100 °C for an hour. At the end of the 30 days period, only one sample presented a supericial change with small red spots, and part of the surfaces had a dark yellow colour. A very intense colouring of the surface of sample CM2 was caused by pickling in a boiling solution of iron sulphate, ammonium chloride, sodium thiosulphate and potassium bitartrate in distilled water. he sample was subsequently heated over a lame, resulting in a toning down of the bright colours (Fig. 1). Sample AM1 was painted with a red iron oxide suspension in distilled water, lightly heated over a lame, and then cooled in water (Fig. 2). Sample BM3 was painted with an iron sulphate suspension in distilled water, heated to red by a lame and cooled in water, and then brushed (Fig. 3). Colouring is a process that can be performed on jewellery items with the purpose of forgery, by pickling, which dissolves the less precious metals on the surface of the jewellery, leaving a few microns of thickness of pure gold. Classical Alessandro PACINI Figure 2: (See colour plate) Sample AM1. Figure 2 : (Voir planche couleur) Échantillon AM1. Figure 3: (See colour plate) Sample BM3. Figure 3 : (Voir planche couleur) Échantillon BM3. Figure 1: (See colour plate) Sample CM2. Figure 1 : (Voir planche couleur) Échantillon CM2. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 393-396 pickling is carried out with a watery solution of sodium chloride, potassium nitrate and alum, but there are also very ancient recipes (for example No. 67 of the Leiden Papyrus – Cafaro and Falanga, 2004) for colouring jewellery. Sample BL1 was immersed for a few seconds in a solution of boiling water, hydrochloric acid, saltpetre and sodium chloride, in order to obtain the typical opaque surface and light corrosion associated with antique jewellery. Around the middle of the 19th century, the archaeological jewellery of Castellani was coloured by using chemical baths Ancient gold patinas: experimental reconstruction to obtain the famous “giallone” of Etruscan gold (Castellani, 1827: 161-164), also employed in this case as a treatment to enrich the surface with a chemical gilding. he colouring was caused by dissolving the metal alloys on surfaces, or it could also be carried out (better on low carat gold, as in the case of tumbaga) by boiling the article in an acid vegetable juice, such as that of the plant Rumex acetosa, or in a solution of salt and vinegar. he colourings, in archaeological fakes, serve above all to uniform the colour of the surface, hiding the eventual brazing made with alloys richer in silver, and thus the resulting lighter colour. he soil deposits often present in antique jewellery are imitated by using a very ine clay mixed with resin or vegetable paste, easily recognizable by the uniform distribution on the precious piece and its light consistency. 3. RESULTS OF THE ANALYSIS ON SAMPLES AM1 AND BM3 AM1 395 Figure 5: X-ray EDS spectrum of sample AM1. Figure 5 : Spectre de rayons X par EDS de l’échantillon AM1. elements are the ochre components and the other products used in the experimental procedure. A part of the qualitative spectrum highlights the iron peak, while the other elements of ochre are a little more higher than the background; this observation could allow us to put forth the hypothesis of the formation of the ferric oxide as being responsible for the rose colouring of the patina (Fig. 5). he surface observed by SEM using backscattered electrons highlights a structure with well formed crystal grains with a darker inter-granulated area. he subsequent EDS microanalysis carried out on the areas indicated in the illustration (Fig. 4) revealed the values of the elements reported in Table 1. In area No. 1, the composition corresponds to the initial alloy, while in the inter-granulated area there is evidence of a decrease of silver and copper, and of the simultaneous formation of ferrous composites. he other minor he supericial morphology highlights a rough aspect, almost spongy, with some formation of crystals. he EDS analysis on the area indicates a square igure, elaborate in respect of the single elements of the golden alloy, revealing an impoverishment only of the silver, probably due to the formation of the silver sulphide then taken from the surfaces (Fig. 6). Figure 4: EDS microanalysis of sample AM1. Figure 4 : Micro-analyse EDS de l’échantillon A1. Figure 6: EDS microanalysis of sample BM3. Figure 6 : Micro-analyse EDS de l’échantillon BM3. BM3 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 393-396 Alessandro PACINI 396 Area C O Na Mg Al Si K Ca Fe Cu 0.54 3.50 95.95 2 30.11 11.18 0.08 0.19 0.21 0.5 0.18 0.78 0.31 0.54 1.88 53.90 3 40.79 13.13 0.21 0.19 0.48 1.5 0.29 1.81 0.39 0.43 1.26 39.05 4 8.75 23.21 0.18 1.01 1.15 2.64 0.15 0.28 0.95 0.49 2.14 58.23 1 Ag Au Table 1: Composition of sample AM1; concentrations are given in wt%. Tableau 1 : Composition de l’échantillon AM1 ; concentrations en %. Spectrum C N 2.73 O Na Cl K Fe Cu Ag Au 1 7.84 2.91 0.63 1.51 2.97 81.40 2 15.36 13.35 4.58 1.76 2.63 62.33 3 29.13 10.13 1.69 0.87 2.04 54.34 4 15.20 11.03 3.18 1.69 2.27 66.33 5 9.42 6.84 1.74 1.75 2.61 77.64 6 8.88 11.27 3.08 1.47 2.58 69.54 3.18 0.5 0.77 0.53 0.3 Table 2: Composition of sample BM3; concentrations are given in wt%. Tableau 2 : Composition de l’échantillon BM3 ; concentrations en %. Table 2 presents the composition (in terms of weight percent) of all the elements present, showing the presence of iron in noticeable quantity, and thus attesting the formation of a red colour ferric product (oxide). 4. CONCLUSION he processes efected on the two samples analysed produced a red colour similar to that of the natural patina, but which is however diferent, due to the lack of signiicant alterations to the golden alloy underneath the iron oxide. his type of artiicial patinas can therefore be revealed by analytical methods. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 393-396 References CASTELLANI, A., 1827. Sul pulimento dell’oro. Antologia, giornale di scienze, lettere e arti XXV: 161-164. CAFFARO, A. and FALANGA, G., 2004. Il papiro di Leida: un documento di tecnica artistica e artigianale dell’IV secolo d.C. Salerno, Arci ed., Postiglione. GRIMWADE, M., 1999. Arricchimento supericiale delle leghe d’oro per gioielleria. Gold technology 26: 16-23. MAZZEO, R., 2005. Patine su manufatti metallici, in Le patine: genesi, significato e conservazione. Nardini ed., Firenze, Kermesquaderni, 29-43. L’illusion de l’or Imitation de dorures dans les peintures murales médiévales he illusion of gold Imitation of gilding in mediaeval mural paintings Aurélie Mounier*, Floréal Daniel* et Françoise Bechtel* Résumé : Au Moyen Âge, les métaux et en particulier l’or, associés dans les peintures murales à certaines couleurs, outre leur rôle iconographique et esthétique, sont porteurs d’une forte charge symbolique en fonction de leur localisation dans l’image. Or, les études récentes de vestiges de dorures dans les peintures murales des édiices religieux ont montré que des feuilles d’étain, à la place d’or, sont identiiées sur la plupart des sites. Une couche de vernis sur la surface de l’étain permettait de lui donner un aspect doré et cette « fausse dorure » était connue de tous les ateliers. Ces techniques sont décrites dans les traités anciens (héophile, xiie siècle ; Cennini, xive siècle ; Denys, xie siècle). La grande diiculté aujourd’hui est de détecter la présence de ces dorures qui ont disparu lorsque les conditions de conservation ont été défavorables. Grâce à la luorescence des liants, l’examen sous lumière UV permet de retrouver les zones originellement dorées. Des dorures sur les peintures murales de la chapelle de l’ancien logis abbatial de Moissac (xiie siècle), de l’église de Nogaro (xie siècle) et de la cathédrale de Cahors (xiiie siècle) ont été découvertes (Mounier et al., 2008). L’analyse des prélèvements (stratigraphies, analyses élémentaires par MEB/EDS, des pigments par spectrométrie Raman et les liants par IRTF) réalisés dans ces vestiges de dorure informe sur la technique et les types de métaux utilisés. Pour les historiens d’art, c’est toute une partie de l’interprétation iconographique qui peut être précisée puisque l’application d’une dorure est l’indice d’une hiérarchisation des personnages et/ou d’une volonté de valoriser certains éléments de l’image. Par ailleurs, élargi à l’organisation de l’atelier et la société, l’emploi d’une « fausse dorure » s’explique par des raisons économiques ; autant de facteurs déterminants qui doivent être pris en considération. Abstract: In the Middle Ages, metals, and gold in particular, associated in wall paintings with certain pigments and colours, in addition to their iconographic and aesthetic function, also had a strong symbolic meaning related to their localization in the image. Recent studies of traces of mediaeval gildings still present in mural paintings have shown that tin leaves, instead of gold, can be identiied in most of the sites. A varnish layer allowed giving them a gold aspect. he use of this ‘false gilding’ was common and well-known by workshops. he techniques pertaining to it are described in ancient treatises. Today, the main issue at stake is to detect traces of gildings which have disappeared when conservation conditions were unfavourable. Examination under UV light allows detection of the originally gilded areas. he gildings in the mural paintings of the chapel in the ancient abbey home of Moissac (12th century AD), the paintings in Nogaro Church (11th century), and on the Cahors Cathedral (13th century) have been discovered according to this procedure. he analyses of samples collected from these gildings provide valuable information about the techniques and the types of metals used. he analyses of microsamples (stratigraphy, SEM/EDXS, Raman spectrometry for the analysis of pigments and FTIR for the binders) show that a gilding or ‘false gilding’, no longer visible today, was often originally applied. For art historians, this allows clarifying part of the iconographic interpretation, as the application of gilding provides information about the hierarchy of the characters and/or the desire to emphasize certain elements of the representation. Moreover, from the point of view of workshop organization, the use of ‘false gilding’ can be related to economic considerations (or supplying facilities). * Institut de Recherche sur les Archéomatériaux (IRAMAT – CRPAA, UMR 5060), Université Bordeaux – CNRS. (fdaniel@u-bordeaux3.fr), (bechtel@ubordeaux3.fr), (mounieraurelie33@yahoo.fr) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 397-403 398 Aurélie MOUNIER, Floréal DANIEL, Françoise BECHTEL Mots-clés : Argent, dorures, étain, fausse dorure, fausse dorure, dorure, or, argent, étain. Keywords: False gildings, gildings, gold, silver, tin. 1. INTRODUCTION « Illusion : Sens abusés par les apparences. » La réalisation de cette illusion afective qui pousse à représenter matériellement ce qui paradoxalement ne peut être vu (monde irréel, Dieu…) constitue la gageure que s’imposent les concepteurs d’images à l’époque médiévale. Une des constantes est le recours à l’or (comme, par exemple, les fonds dorés des icônes). Chargé symboliquement en raison de son coût, de son inaltérabilité, de sa brillance, l’or est choisi pour souligner un attribut symbolique, religieux ou profane (auréole, glaive, leur de lys…). Il faut considérer également l’illusion perceptive guidée par la volonté, quelles qu’en soient les raisons (économiques ou de conservation), de tromper, en substituant à l’or d’autres métaux moins nobles qu’on recouvrait d’un vernis pour lui donner l’apparence du métal le plus précieux. C’est à cette double illusion que nous sommes confrontés lorsqu’il s’agit d’interpréter les dorures, vraies ou fausses, détectées dans les peintures murales médiévales. Toutefois, le premier obstacle à surmonter pour l’étude de ces éléments de décor métalliques est la détection des traces de dorure subsistant dans les peintures murales. Cette question est résolue par l’examen sous ultraviolets et l’analyse de micro-prélèvements permettant respectivement de localiser et de caractériser ces décors métalliques en vue d’une restitution de l’image originelle. Pour l’interprétation de ces vestiges de dorures, une aide est fournie par les sources écrites et les études de cas relevés dans la bibliographie. La technique de la dorure est connue et a été décrite dans de nombreux traités de l’Antiquité. Pour le Moyen Âge, on retient ceux du moine Denys de Fourna (xie siècle), du moine héophile (xiie siècle) et de Cennino Cennini (xive siècle). De l’étude de cette documentation, il résulte que deux principales techniques ont été utilisées : à la détrempe et à la mixtion. Dans le premier cas, le pigment est mêlé à un liant organique de type colle ou gomme arabique et permet de faire adhérer la feuille métallique qui peut être de l’or, de l’argent, de l’étain. Dans le second cas, l’emploi d’un mordant est nécessaire. Souvent, on emploie l’huile ArcheoSciences, revue d’archéométrie, 33, 2009, p. 397-403 de lin qui doit être associée à un siccatif comme la litharge (PbO), ain de coller la feuille de métal. Ces deux techniques difèrent donc principalement par le liant organique employé qui, en général, est identiiable in situ par sa luorescence sous éclairage UV. Une étude précédente (Mounier et al., 2009) avait montré que les caractéristiques de la luorescence dépendent du liant employé et que l’observation sous UV est un bon moyen de détecter les éventuelles traces de dorures encore conservées. Dans l’étude présentée ici, la problématique générale consiste à montrer l’importance de la dorure au Moyen Âge dans les peintures murales, et à travers quelques exemples, d’en retrouver leur technique d’application et d’étudier le lien avec la symbolique et l’iconographie. Nous nous sommes limités géographiquement à l’Aquitaine historique 1 sur une période allant du xie au xve siècle. Les recherches bibliographiques et l’analyse de divers sites nous ont permis de dresser un inventaire qui montre que les dorures, vraies ou fausses, sont assez présentes et leur technique très diversiiée. Bien que certains articles d’histoire de l’art et des récits de voyageurs témoignent de la présence de dorures dans les peintures murales, il convient de rester prudent à leur lecture, car bien souvent, le terme de dorure est employé pour décrire des auréoles peintes à l’ocre jaune. Enin, d’autres sites ont été étudiés et analysés par nous-mêmes et diférents types de dorures ont été identiiés. Certains ne présentent que de la dorure à l’or, à l’argent ou à l’étain, d’autres combinent deux ou trois types. 2. MÉTHODOLOGIE Échantillons Vingt-huit micro-prélèvements (<1mm) ont été réalisés dans diférents sites : les peintures de l’absidiole sud de l’église de Nogaro ; celles de la voûte en berceau plein cintre de la chapelle de l’ancien logis des abbés de Moissac ; 1. Région dont le territoire était compris entre la Garonne, les Pyrénées et l’océan Atlantique. 399 L’illusion de l’or. Imitation de dorures dans les peintures murales médiévales un bloc polychrome et doré provenant de l’ancien couvent des Cordeliers d’Oloron Sainte-Marie ; les décors de la façade de la cathédrale Saint-Étienne de Cahors ; les peintures funéraires de la chapelle Sainte-Anne de la cathédrale Saint-André de Bordeaux ; l’ensemble peint de l’église SaintSauveur de Saint-Macaire (Tableau 1). Ils ont été réalisés en collaboration avec les restaurateurs 2 dans les zones qui présentaient une luorescence sous éclairage UV. Méthodes physico-chimiques Les analyses élémentaires ont été réalisées par microscopie électronique à balayage (MEB, Jeol JSM-6460LV). Cet appareil, qui permet d’obtenir des images en électrons secondaires (contraste topographique) et en électrons rétrodifusés (contraste chimique), est doté d’un analyseur de rayons X à dispersion d’énergie (EDS) permettant l’étude de la composition élémentaire des échantillons. Ce système est associé à un logiciel de microanalyse (Oxford INCA 300) permettant l’acquisition et le traitement des spectres. Les prélèvements ont été soit directement déposés sur des pastilles de carbone, soit enrobés dans une résine (Sody 33), découpés transversalement et polis pour l’étude de la succession de couches de mortier, de pigments et de la feuille métallique. Site La possibilité de travailler en pression partielle (15 Pa et 20 Pa) a permis d’éviter la métallisation des échantillons et de pouvoir les réutiliser pour des analyses ultérieures. Le spectromètre Raman est un appareil Renishaw RM2000, avec un laser rouge (633 nm) ou vert (535 nm). Une caméra CCD permet la visualisation de l’échantillon en lumière naturelle sur un écran d’ordinateur et le logiciel informatique GRAMS permet de traiter les spectres. Les spectres obtenus ont été comparés à ceux de bases de données publiées (Bell et al., 1997). 3. RÉSULTATS L’examen et l’analyse des échantillons montrent la diversité des types de dorures qu’il est possible de rencontrer : à l’or, à l’argent, à l’étain pour donner un efet doré ou argenté (utilisé comme support pour un autre métal ou destiné à être recouvert de vernis). La dorure à la feuille d’or Les exemples de dorure à l’or sont nombreux et se multiplient jusqu’à l’époque Baroque. Par la suite, le lien avec la Nombre de prélèvements Localisation des peintures Zone des prélèvements Nogaro (xiie, Gers) Cul-de-four de l’absidiole sud, Christ en Majesté 2 Manteau du Christ Etoile Moissac (xiie, Tarn-etGaronne) Voûte en berceau plein cintre, Christ en Majesté 2 Auréole du Christ Auréole d’un symbole Évangéliste Cahors (xiii , Lot) Portail de la façade occidentale, sous le linteau XVIIe dégagé 8 Auréoles Instruments de musique Fleurs de lys Fleurettes Oloron Sainte-Marie (xive, Pyrénées-Atlantiques) Bloc conservé au SRA, DRAC de Bordeaux 8 Voile Vierge Architecture Frise Bordeaux (xiv , Gironde) Chapelle Sainte-Anne, peintures funéraires d’Arnaud de Puylehaut et de Pons de Pommiers 7 Rayons du soleil Bouton de rose Couronne et auréole de la Vierge Étoiles Saint-Macaire (xive, Gironde) Cul-de-four de l’abside centrale, scène de l’Apocalypse 1 Glaive e e Tableau 1 : Échantillonnage de dorures dans les peintures murales de divers sites du grand Sud-Ouest du xiie au xive siècle. Au total, 28 microprélèvements ont été réalisés ain d’identiier les matériaux et les techniques d’application. Table 1: Sampling of gildings in the murals of diverse sites in the South West of France. 28 microsamples were taken to identify materials and techniques. 2. Jean-Marc Stoufs à Moissac et Cahors, Didier Legrand à Nogaro, Rosalie Godin à Saint-Macaire. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 397-403 400 symbolique de la représentation se distend. L’or sera utilisé de façon plus systématique, même dans des zones secondaires et à des ins simplement décoratives. Dans notre échantillonnage, la dorure à l’or est la technique la plus représentée. Un premier exemple illustre cette technique de dorure : il s’agit des peintures murales funéraires d’Arnaud de Puylehaut (xive siècle), conservées à la cathédrale SaintAndré de Bordeaux. Dans le registre supérieur, Dieu dans sa gloire, est représenté sous la forme d’un soleil (Fig. 1 et 2). L’extrémité des rayons les plus courts se termine par un bouton de rose doré (Gaborit, 1997). Les analyses par spectrométrie de rayons X couplée au MEB et à la spectrométrie Raman nous ont permis d’identiier les matériaux de ces dorures. Sur la pierre est appliquée une couche colorée rouge orangée, un mélange de calcite, d’ocre rouge et d’un peu de cinabre. Par-dessus, une ine couche contenant du plomb certainement mêlée à un liant organique sert de couche adhésive à la feuille métallique. L’or est localisé sur le registre supérieur des peintures, sur la scène représentant Dieu, là où la symbolique est la plus forte. De la dorure à la feuille d’or est également présente sur les peintures murales du cul-de-four de l’absidiole sud de Figures 1 et 2 : (Voir planche couleur) À gauche : représentation du soleil dans le registre supérieur des peintures murales funéraires d’Arnaud de Puylehaut (chapelle Sainte-Anne de la cathédrale Saint-André de Bordeaux). À droite : détail d’un bouton de rose peint à l’extrémité des rayons les plus courts du soleil. (Clichés : A. Mounier). Figures 1 and 2: (See colour plate) On the left: representation of the sun in the upper register of the funerary wall paintings of Arnaud de Puylehaut (Saint-Anne’s chapel in the cathedral of Saint-André in Bordeaux). On the right: detail of a rosebud painted on the extremities of the shorter sunbeams (Photos: A. Mounier). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 397-403 Aurélie MOUNIER, Floréal DANIEL, Françoise BECHTEL l’église de Nogaro datées du xie siècle. La scène représentée est un Christ en Majesté, entouré du Tétramorphe. La dorure est localisée sur le manteau du Christ. Les analyses ont permis de montrer que la stratigraphie est composée de cinq couches : un support de calcite et silice, une couche rouge mêlant du cinabre à de l’ocre rouge (par mesure d’économie, on peut le supposer), une couche d’oxyde de plomb sous la couche de mixtion plus ine d’ocre rouge et de blanc de plomb ain de faire adhérer la feuille d’or. Il est intéressant de noter que, dans les deux cas, le pigment le plus onéreux, le cinabre, est associé à l’or. Un dernier exemple illustrant cette technique est le bloc retrouvé lors des fouilles de l’ancien couvent des Cordeliers d’Oloron-Sainte-Marie, aujourd’hui conservé au Service Régional d’Archéologie de la DRAC Aquitaine. Il s’agirait d’un vestige d’un ensemble funéraire du xive siècle. La scène représentée est une Vierge au voile doré, tenant un cierge. Elle se tient sous un élément d’architecture. Les quatre faces du bloc sont peintes et dorées (Gaborit, 2002). La stratigraphie d’un prélèvement efectué sur la face principale montre les couches successives suivantes : sur le support de pierre, une couche de calcite et de minium puis une ine couche de L’illusion de l’or. Imitation de dorures dans les peintures murales médiévales 401 Figures 3 et 4 : (Voir planche couleur) Stratigraphie de l’échantillon prélevé au revers du bloc polychrome provenant d’Oloron SainteMarie (BDX 12999). À gauche : en lumière naturelle (microscope polarisant, x200). À droite : image au MEB (20 kV, mode électrons rétrodifusés, x120, pression partielle : 20Pa). Les trois couches ont été identiiées par MEB/EDS et spectrométrie Raman. De bas en haut : 1- support blanc composé de calcite ; 2- couche d’ocre jaune (goethite) ; 3- feuille d’or. Figures 3 and 4: (See colour plate) Stratigraphy of the sample collected from the obverse of the polychrome block from Oloron Sainte-Marie (BDX 12999). On the left: viewed under natural light (polarising microscope, 200x). On the right: SEM image (20 kV, BSE mode, 120x, low vacuum: 20Pa). he three layers were identiied by SEM/EDX and Raman spectroscopy. From the bottom: 1- white support composed of calcite; 2- layer of yellow ochre (goethite); 3- gold foil. calcite, une couche d’ocre rouge sur laquelle est appliquée la feuille d’or. Sur la frise, du côté droit du bloc, la feuille d’or est appliquée sur une couche constituée de cinabre mêlé à la calcite déposé sur un support de calcite et d’ocre rouge Fig. 3 et 4). Sur des échantillons prélevés à d’autres endroits, la couche colorée est un mélange de minium, cinabre et hématite. La dorure est volontairement associée à des pigments onéreux comme le cinabre ou le lapis-lazuli, pigment bleu identiié sur la face principale. Toutefois ces pigments sont souvent mélangés à d’autres matériaux plus communs comme les ocres ou comme du noir de carbone sous la couche bleue, ain d’économiser les pigments comme l’azurite ou le lapis-lazuli. La dorure à la feuille d’argent Sur les peintures funéraires de la chapelle Sainte-Anne à la cathédrale Saint-André de Bordeaux, des feuilles d’argent noircies, très altérées ont été identiiées à divers endroits sur les rayons du soleil (Daniel et Mounier, 2010). L’argent est appliqué sur une mixtion rouge orangée composée de minium et d’ocre rouge. Sous cette couche, se trouve une autre constituée de calcite et d’ocre sur un support de calcite. L’espace entre les rayons du soleil est garni de petits anges qui présentent le même aspect noirci sur les ailes et les auréoles. Celles-ci étaient probablement dorées. Si on considère le thème iconographique, le soleil représentant Dieu, l’argent était certainement à l’origine recouvert d’un vernis jaune pour donner une illusion dorée. Sur ces mêmes peintures, dans le registre inférieur, la couronne et l’auréole d’une Vierge à l’enfant sont « dorées » à la feuille d’argent. Celle-ci est appliquée sur une couche de minium et d’ocre rouge. En règle générale, par souci de réalisme, les auréoles et couronnes sont dorées. L’argent a pu être verni pour donner l’aspect de l’or mais a pu également être utilisé pour sa propre couleur. Le choix des métaux est déini en fonction de la hiérarchisation des scènes. On préfère l’or dans le registre supérieur illustrant Dieu, et l’argent pour la zone médiane, où la symbolique est moins forte. Un dernier exemple est donné avec les peintures murales de l’église Saint-Sauveur de Saint-Macaire (xive siècle). Elles sont localisées sur le cul-de-four de l’abside centrale. Ici, l’argent a été employé par souci de réalisme puisqu’il est localisé sur le glaive tenu dans la bouche de Dieu. La feuille d’argent a été appliquée sur une couche orangée composée d’ocre rouge et de blanc de plomb, sur une couche de calcite et de blanc de plomb. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 397-403 402 Aurélie MOUNIER, Floréal DANIEL, Françoise BECHTEL La dorure à la feuille d’étain L’étain en tant que support pour un autre métal ou destiné à être verni Dans l’église Saint-Nicolas de Nogaro, l’étain a été utilisé, selon une technique décrite dans les textes anciens, en sous-couche à la feuille d’or. Il a été employé pour réaliser les petites étoiles qui parsèment le fond de la mandorle du Christ. Sur un support à fresque, deux couches colorées ont été appliquées (de l’ocre rouge et de l’azurite qui correspond au bleu du fond de la mandorle). Ensuite, une couche de mixtion est constituée d’oxyde de plomb et un liant organique identiié comme étant de l’huile de lin (Mounier et al., 2008). Enin, la feuille d’étain, assez épaisse, sert de base à la feuille d’or beaucoup plus ine. Cette technique est principalement utilisée en raison de l’état de surface irrégulier de la peinture qui ne facilite pas l’application d’une feuille métallique trop mince. L’utilisation de l’étain permet également de donner du relief et d’économiser l’or. Parfois, l’étain a pu être employé en substitut à l’or. Les traités anciens décrivent cette pratique. On parle alors de « fausse dorure ». À Moissac, par exemple, de la dorure à la feuille d’étain a été trouvée sur les auréoles du Christ et des évangélistes. Leur localisation sur des personnages importants laisse penser qu’à l’origine, ces dorures à l’étain étaient dorées à l’aide d’un vernis jaune appliqué sur le métal. De nombreuses recettes de ce « vernis » appelé auripetrum sont données. Il s’agit généralement d’un mélange de cire, huile de lin, pigments tels que le sang-dragon ou le safran conférant au métal un aspect doré (héophile, xiie siècle ; Cennini, xive siècle ; Denys xie siècle). Sur le portail de la façade de la cathédrale de Cahors, des dorures ont été découvertes sur des leurettes et des leurs de lys peintes au xiiie siècle, bien conservées grâce à un linteau qui les avait recouvertes au xviie siècle. Les leurs de lys sont constituées d’une feuille d’étain collée sur une couche d’azurite, puis une couche de blanc de plomb et calcite et enin le support de pierre (Fig. 5). La leur de lys est un symbole royal. On la trouve plus souvent dorée à l’or. Dans le cas des leurettes du portail, l’étain est utilisé soit en substitut à l’or ou, si l’efet recherché est argenté, l’étain a pu être choisi pour ses qualités de conservation supérieures à celles de l’argent. L’étain comme substitut à l’argent L’analyse de prélèvements efectués sur des leurettes hexalobes rouges décorant le portail occidental de la cathédrale de Cahors, a permis de montrer que la stratigraphie corresArcheoSciences, revue d’archéométrie, 33, 2009, p. 397-403 Figure 5 : Image au MEB (électrons rétrodifusés) de la stratigraphie de l’échantillon prélevé dans une leur de lys ornant le portail de la cathédrale de Cahors. 1- feuille d’étain (40 µm) ; 2- ine couche d’azurite ; 3- couche de blanc de plomb (grains clairs) et de calcite et couche préparatoire. Les deux dernières ne sont pas clairement délimitées. Figure 5: Image under the SEM (BSE) of the sample collected from the leur-de-lis decorating the gate of the Cahors Cathedral. 1- tin foil (40 µm); 2- thin layer of azurite; 3- layer of white lead (light grains) and of calcite and preparation layer. he latter two are not clearly delimited. pond à la technique de la dorure. Sur la couche préparatoire, une couche de blanc de plomb et de calcite a été appliquée sous la couche colorée rouge composée de cinabre ain de faire adhérer une feuille d’étain. Nous pensons ici que l’efet escompté était de rendre ces leurettes argentées. Il s’agissait, à travers ces quelques exemples, de montrer d’une part la diversité des types de dorures et d’autre part que le choix de type de métal est orienté par des raisons symboliques, économiques ou par souci de réalisme. Les dorures sont souvent associées à des pigments onéreux et l’on constate une hiérarchisation des matériaux (pigments, feuilles métalliques). L’examen des peintures murales de la chapelle de l’ancien logis abbatial de Moissac montre que l’organisation et la répartition des matériaux est réléchie. Le lapis-lazuli, le cinabre et les dorures sont localisés sur le Christ ; l’azurite sur la Vierge et l’aérinite (un pigment bleu bon marché) (Daniel et al., 2008) sur le fond des scènes. La dorure est associée à des pigments chers sur des zones où la symbolique est forte, sur des personnages que l’on veut mettre en valeur. L’illusion de l’or. Imitation de dorures dans les peintures murales médiévales 4. INFORMATIONS TECHNIQUES COMPLÉMENTAIRES Les caractéristiques techniques de ces diférentes dorures méritent également d’être considérées. L’épaisseur des couches métalliques varie entre 1 à 3 µm pour la feuille d’or ; entre 40 et 60 µm pour les feuilles d’étain et de 2 à 10 µm pour les feuilles d’argent. Bien qu’il faille tenir compte du degré d’usure de la feuille après plusieurs siècles, l’épaisseur est assez homogène et plus ou moins identique d’un site à l’autre. La technique la plus représentée est la technique à la mixtion. Le liant, de nature lipidique, a été identiié par infrarouge à transformée de Fourier comme pouvant correspondre à de l’huile de lin. L’huile est dans tous les cas associée au plomb pour son pouvoir siccatif, sous forme de pigment (minium, blanc de plomb…). 5. CONCLUSION Cette étude montre qu’à l’époque médiévale les dorures étaient certainement beaucoup plus présentes qu’on ne le pense sur les peintures murales. La combinaison de l’examen sous ultraviolets et de l’analyse permet l’identiication des matériaux et des techniques de la dorure et montre leur grande variété. Les résultats obtenus sont en accord avec la description des techniques de dorures dans les traités de recettes anciennes. Trois métaux sont employés : l’or, l’argent et l’étain le plus souvent, selon la technique de la mixtion, moins sensible à l’environnement, à l’humidité, surtout quand ces décors sont en extérieur. La compréhension des techniques, du choix des matériaux et dorures en fonction des thèmes iconographiques constitue une aide précieuse aux historiens d’art. Elle permet de comprendre la valeur symbolique et l’importance relative de certains personnages ou décors. Néanmoins, au-delà de l’étude des matériaux et techniques et de l’iconographie, il manque des informations essentielles d’ordre historique pour comprendre le choix de ces matériaux : les ressources économiques du commanditaire, le savoir-faire technique, la disponibilité des matériaux précieux qui sont en dehors du champ purement technique ou de l’image. La présence d’un vernis, l’auripetrum, donnant un aspect doré aux feuilles d’argent ou d’étain n’a pas été identiié sur ces sites mais il serait intéressant de poursuivre les recherches et la caractérisation des vernis qui vraisemblablement couvraient les « fausses » dorures. 403 Remerciements Cette étude a été rendue possible grâce au soutien de la région Aquitaine et aux conservateurs de la DRAC Aquitaine qui nous autorisent à efectuer les micro-prélèvements sur sites et aux restaurateurs, pour leur expérience de terrain. Bibliographie BELL, I. M., CLARK, R. J. H. et GIBBS, P. J., 1997. Raman spectroscopic library of natural and synthetic pigments (Pre- ~1850 AD), Spectrochimica Acta Part A, 53, (2): 2159-2179. CENNINI, C., XIVe siècle, réédité en 1991. Il libro dell’arte, BergerLevrault, Paris, 188-313. DANIEL, F., LABORDE, B., MOUNIER, A. et COULON, E., 2008. Le pigment d’aérinite dans deux peintures murales romanes du Sud-Ouest de la France, Archéosciences 32: 83-91. DANIEL, F. et MOUNIER, A., 2010. Alteration of gilts on mediaeval mural paintings, Communication présentée à l’ICOM-CC (Interim Meeting), Multidisciplinary Conservation: A holistic view for historic interiors, Rome, 23-26 mars 2010. DENYS, H., 1845. Le guide de la peinture, Traduit par Paul Durand « Manuel d’iconographie chrétienne, grecque et latine », Paris, Didron. GABORIT, M., 1997. Découvertes de peintures murales dans la cathédrale Saint-André de Bordeaux, Revue Archéologique de Bordeaux, XXXVIII: 67-101. GABORIT, M., 2002. Des hystoires et des couleurs, peintures murales en Aquitaine. XIIIe-XIVe siècle, Bordeaux, Conluences. MOUNIER, A., DANIEL, F. et BECHTEL, F., 2008. Gilding techniques in mural paintings in the Romanesque period in France, Actes du colloque « Archéométrie 2008 », 37th International Symposium on Archaeometry, 12-16 mai 2008. Sienne, Italie, sous presse. MOUNIER, A. DAYET, L., BELIN, C. et DANIEL, F., 2009. Étude de la luorescence des liants employés dans les dorures sur peintures murales médiévales, 18e colloque du Groupe des Méthodes Pluridisciplinaire Contribuant à l’Archéologie, 6-10 avril 2009, Montpellier « Ressources, Sociétés, biodiversité», Archéosciences, sous presse. THÉOPHILE (Moine), XIIe siècle, réédité en 1996. Essais sur divers arts, Publié par le Cte Charles de l’Escalopier, réédition Laget, Nogent-Le-Roi, 39-44. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 397-403 Gold leafs in 14th century Florentine painting Feuilles d’or dans la peinture florentine du XIVe siècle Giovanni Buccolieri*, Alessandro Buccolieri*, Susanna Bracci**, Federica Carnevale*, Franca Falletti**, Gianfranco Palamà*, Roberto Cesareo*** and Alfredo Castellano* Abstract: Gold leafs are typically present in paintings and frescoes of the Italian Renaissance in the 13th and 14th centuries. he chemical composition and thickness of gold leafs provide important information toward a better understanding of the technology of that epoch. he present paper discusses the results of non-destructive analysis carried out with a portable energy dispersive X-ray luorescence (ED-XRF) equipment on the 14th century panel Annunciation with Saints Catherine of Alexandria, Anthony Abbot, Proculus and Francis by the painter Lorenzo Monaco. Résumé : L’application de feuilles d’or sur des tableaux et des fresques est typique de la Renaissance italienne aux XIIIe et XIVe siècles. Leur composition chimique et leur épaisseur fournissent des informations importantes pour la compréhension des techniques de cette période. Ce travail discute les résultats des analyses non-destructives réalisées avec un système portable de luorescence X à dispersion d’énergie (ED-FX) sur le panel du XIVe siècle Annonciation avec Saints Catherine d’Alexandrie, Anthony Abbot, Proculus et Francis par le peintre Lorenzo Monaco. Keywords: Gold leafs; Florentine painting; non-destructive analysis; ED-XRF. Mots-clés : Feuilles d’or, peinture lorentine, analyse non-destructive, ED-FX. 1. INTRODUCTION All populations have always used metals to make ornaments, such as leafs, on paintings or other objects of historical art. he most frequent metals employed were gold, silver, lead, copper, tin, and iron (Scott and Dodd, 2002; Duran et al., 2008). Gold leafs are typically present in paintings and frescoes of the Italian Renaissance in the 13th and 14th centuries, as background, haloes, or decorations. he non-destructive determination of gold leaf composition and thickness in paintings and frescoes is of primary importance for a better understanding of the technological development of the respective epoch. Generally, when a gold leaf is studied by ED-XRF analysis, an X-ray spectrum is obtained, typically containing two Cu-K lines, three Au-L lines and two Ag-K lines. From a quantitative analysis of these lines, the concentrations of Cu, Au and Ag can be determined. Furthermore, the thickness of the Au leaf may be obtained by a detailed study of the internal ratio of the elements. In fact, the X-ray intensity ratios IKα/IKβ, or ILα/ILβ and ILα/ILγ, for an element in a multilayer sample, depend on the composition and * Università del Salento, Dipartimento di Scienza dei Materiali, via per Monteroni, 73100, Lecce, Italy. (giovanni.buccolieri@unisalento.it) (alessandro. buccolieri@unisalento.it) (carnevale.federica@alice.it) (gianfranco.palama@unisalento.it) (alfredo.castellano@unisalento.it) ** C.N.R. – I.C.V.B.C., via Madonna del Piano, 50019, Sesto Fiorentino, Firenze, Italy. (s.bracci@icvbc.cnr.it) *** Struttura Dipartimentale di Matematica e Fisica, via Vienna n. 2, 7100, Sassari, Italy. (cesareo@uniss.it) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 405-408 406 thickness of the layer in which the element is situated, and on the composition and thickness of the superimposed layer (or layers). Multilayered samples are common in archaeometry, for example in the case of pigment layers in paintings or of gilded alloys. In the present paper, experimental results of an ED-XRF analysis carried out on the panel “Annunciation with Saints Catherine of Alexandria, Anthony Abbot, Proculus and Francis” (Fig. 1) by the Florentine painter Lorenzo Monaco (born Piero di Giovanni, c. 1370-1425) are reported. Lorenzo Monaco (Tartuferi and Parenti, 2006) was an artist whose work bridged the 14th and 15th centuries, between the Trecento art of Duccio and Giotto and the Quattrocento painting of Masaccio and Fra Angelico. His art is considered a synthesis of the Sienese and Florentine Trecento styles and was also one of the most important examples of international Gothic in Italy. In 1391, Lorenzo Monaco took his vows as a monk of the Camaldolese monastery of Santa Maria degli Angeli. He rose to the rank of deacon, but in 1402 he was enrolled in the painters’ guild with the name of Piero di Giovanni (Lorenzo Monaco means ‘Lorenzo the Monk’) and was living outside the monastery. he mon- Giovanni BUCCOLIERI et al. astery was renowned for its illuminated manuscripts, and several book miniatures, now in the Laurentian Library in Florence, are attributed to him. However, the artist was primarily a painter of altarpieces, examples of which are in the National Gallery in London and at the Uizi in Florence. His main works in fresco are the scenes of the Life of Mary in the Bartolini Chapel of Sta Trinità, Florence. His style is characterized by a luminous beauty of colouring, and a graceful, rhythmic low of the line, and is in complete contrast to his great contemporary Masaccio, representing the highest achievement of the last lowering of Gothic art in Florence. Energy dispersive X-ray luorescence (ED-XRF) analysis is a common non-destructive method employed in archaeometry for quantitative analysis and for the determination of the thickness of coatings (Nygard et al., 2004; Roldán et al., 2006). ED-XRF analysis generally involves an area of a few mm2 and a thickness between microns and fractions of mm. herefore, the analysis is supericial and dependent on surface conditions (Cesareo et al., 2004). In the case of multiple layer samples, it is not possible to rely on the use of standards, and therefore one has to resort to theoretical estimates of the XRF Figure 1: (See colour plate) Annunciation with Saints Catherine of Alexandria, Anthony Abbot, Proculus and Francis (ca. 1410-1415). Panel 210 x 229 cm by Lorenzo Monaco. Galleria dell’Accademia, Florence, Italy. he points on the painting shows the regions analyzed. Figure 1 : (Voir planche couleur) Annonciation avec Saints Catherine d’Alexandrie, Anthony Abbot, Proculus et Francis (ca. 14101415). Panel de 210 x 229 cm par le peintre Lorenzo Monaco. Galleria dell’Accademia, Florence, Italie. Les points sur le tableau correspondent aux régions analysées. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 405-408 407 Gold leafs in 14th century lorentine painting intensity. he accuracy of the method can be estimated by using carefully characterized calibration samples. he aim of the present study is the evaluation of the chemical composition and thickness of gold leafs by using a portable ED-XRF device in order to obtain important information permitting a better understanding of the technology of that epoch. 2. EXPERIMENTAL SET-UP he analysis was performed using a portable energy dispersive X-ray luorescence (ED-XRF) instrument designed and built at the University of Salento (Cesareo et al., 1999). It is composed of an X-ray tube (produced by MOXTEK), with a Pd-anode operating at 15 kV voltage and 3 μA current, a Si-PIN detector, thermoelectrically cooled, with a Be window of 25 µm, a resolution of 190 eV at 5.9 keV and a pocket multi-channel analyzer (both produced by AMPTEK). A stratigraphic study could be carried out following one hypothesis, where the gold thickness could be approximately evaluated by determining the ratio Fe-Kα/Fe-Kβ, because these lines are diferently absorbed by gold. In fact, the Fe-Kα X-rays (with an energy of 6.4 keV) are more absorbed than the Fe-Kβ X-rays (7.0 keV), since the mass attenuation coeicients are μα = 380 cm2·g-1 and μβ = 316 cm2·g-1, respectively. In Figure 3, the symbol x (g cm-2) indicates the radiation paths of Fe-Kα and Fe-Kβ into gold, and the symbols I0 and I indicate the intensity of the radiation before and after the layer of gold, respectively. 3. RESULTS AND DISCUSSION ED-XRF spectra of the analyzed paintings show the presence of gold, iron and calcium in the entire painting, while silver, generally present in the composition of gold leafs, was not detected. he presence of iron is due to the use of bolus (iron silicate hydroxide), employed by the artist as a glue to superimpose the gold leafs (Baldinucci, 1681; Cennini, 1971). Calcium is probably used by the artist as priming, as in the case of ‘bianco San Giovanni’. Figure 2 shows the ED-XRF spectra of regions 1, 2 and 3. Fe-Kα 3250 3000 2750 Zone 1 Zone 2 Zone 3 2500 Figure 3: Simulation for determining the thickness of the coating of gold on iron. Figure 3 : Simulation pour l’évaluation de l’épaisseur de la couche d’or sur le fer. 2250 Au-M 1000 750 Au-Lβ Fe-Kβ 1250 Au-Lα 1500 500 250 Au-Lγ 1750 Ca-Kα Counts 2000 0 1 2 3 4 5 6 7 8 9 10 11 Energy (keV) Figure 2: ED-XRF spectra of regions 1, 2 and 3. Figure 2 : Spectres ED-FX des régions 1, 2 et 3. 12 13 14 15 In region 3 (Fig. 2), there is a single leaf (of thickness x), while in the near region 2 there are two leafs (of thickness 2x). herefore, by using equation 1 on these areas, the ratio (Iα/Iβ)0 was calculated for iron. his value is equal to 7.42 and it has been used for the determination of the gold thickness in diferent regions of the painting; the values obtained are reported in Table 1. he gilding technique employing gold leafs uses the superposition of several layers of gold leafs, particularly in areas where covering with one ArcheoSciences, revue d’archéométrie, 33, 2009, p. 405-408 408 Giovanni BUCCOLIERI et al. Region Total thickness (μm) No. of layers hickness (mm) 1 3.5 4 0.9 4 3.1 3 1.0 5 2.3 2 1.2 6 1.6 2 0.8 7 2.0 2 1.0 8 1.1 1 1.1 9 0.9 1 0.9 10 2.8 3 0.9 11 1.1 1 1.1 Table 1: Determination of gold thickness. Tableau 1 : Détermination de l’épaisseur des feuilles d’or. single layer is diicult to carry out. he gold leafs were glued on the substrate with bolus, a compound containing iron (Cennini, 1971). Experimental data obtained by ED-XRF measurements on a single layer showed that the thickness of the gold leafs used by the artist is between 0.9 and 1.1 μm. Measurements carried out on apparently thicker areas led us to hypothesize that these areas are made by superimposing several layers of equal thickness. he thicknesses analyzed show the remarkable technological capacity of the artists of the time (batiloro). As shown above, it was possible to obtain gold leaf thicknesses of about one micrometer from a gold coin. Assuming that the batiloro used an Italian lorin coin (3.54 g), from one such coin it was possible to obtain gold leafs with an area between 1500 and 2000 cm2. 4. CONCLUSION Experimental results show the presence of gold, iron and calcium all over the surface of the painting. he presence of iron may be due to the use of a layer that contained a pigment of iron under the gold leaf, while calcium was probably used by the artist as priming. Moreover, the reported data show that the thickness of a surface gold leaf may be determined by a diferential absorption of deeper elements. hus, energy dispersive X-ray luorescence appears as a powerful, fast, inexpensive, non-destructive and portable method of analysis that can be used successfully in archaeometry in order to determine the chemical composition and the thickness of gold leafs on paintings or other objects of historical and artistic interest. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 405-408 References BALDINUCCI, F., 1681. Vocabolario Toscano dell’arte del disegno. Firenze: Santi Franchi. CENNINI, C., 1971. Il libro dell’arte. Vicenza, Neri Pozza. CESAREO, R., CASTELLANO, A., BUCCOLIERI, G. and MARABELLI, M., 1999. A portable apparatus for energy-dispersive X-ray luorescence analysis of sulfur and chlorine in frescoes and stone monuments. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 155: 326-330. CESAREO, R., CASTELLANO, A., BUCCOLIERI, G., QUARTA, S., MARABELLI, M., SANTOPADRE, P., LEOLE, M. and BRUNETTI, A., 2004. Portable equipment for energy dispersive X-ray luorescence analysis of Giotto’s frescoes in the Chapel of the Scrovegni. Nuclear Instruments and Methods in Physics Research Section B, 213: 703-706. DURAN, A., PEREZ-RODRÍGUEZ, J.L., JIMENEZ DE HARO, M.C., HERRERA, L.K. and JUSTO, A., 2008. Degradation of gold and false golds used as gildings in the cultural heritage of Andalusia, Spain. Journal of Cultural Heritage 9: 184-188. NYGARD, K., HAMALAINEN, K., MANNINEN, S., JALAS, P. and RUOTTINEN, J.P., 2004. Quantitative thickness determination using x-ray luorescence: application to multiple layers. X-Ray Spectrometry 33: 354-359. ROLDÁN, C., COLL, J. and FERRERO, J., 2006. EDXRF analysis of blue pigments used in Valencian ceramics from the 14th century to modern times. Journal of Cultural Heritage 7: 134-138. SCOTT, D.A. and DODD, L.S., 2002. Examination, conservation and analysis of a gilded Egyptian bronze Osiris. Journal of Cultural Heritage 3: 333-345. TARTUFERI, A. and PARENTI, D. (eds.), 2006. Lorenzo Monaco – A bridge from Giotto’s heritage to the Renaissance. Florence, Giunti. Gilding on wall paintings from a period between the 14th-16th centuries AD in the Lombard lake region La dorure sur peintures murales entre le XIVe et le XVIe siècles dans la région des lacs lombards Giovanni Cavallo* and Marcel Verda** Abstract: he use of gildings on wall paintings enjoyed its largest development in Italy between the 13th and 14th centuries AD. his work covers a few important artefacts from the Lombard lake region, dating from a period between 1360 and 1520, and painted by the Seregnesi, Da Tradate and De Passeri studios. Its purpose is to understand the gilding techniques by means of scientiic examinations (OM, SEM/EDS, FTIR), integrated with information from ancient treatises. Two main gilding techniques have been used in this context: the composite foil (tin and gold), and the single gold foil, in both cases ixed on the wall by a type of oil-based adhesive with the addition of inorganic pigments, mostly lead white and ochres, called missione. he scientiic results are in agreement with the recipes contained in the ancient treatises, particularly in the Libro dell’Arte by Cennino Cennini. Extending the research to other case studies in Tessin and Italy, it is clear that the gilding technique was not inluenced by the geographical and historical context, but by the inal appearance required by the decoration on the wall surface. he use of tin in the composite foil permitted the gold foil to be burnished and engraved. Résumé : L’emploi de la dorure sur des peintures murales en Italie a eu son apogée entre le XIIIe et le XIVe siècle. Ce travail traite les objets peints dans la région des lacs lombards, entre 1360 et 1520, réalisés dans les ateliers des Seregnesi, des Da Tradate et d’Andrea de Passeri, en vue d’une compréhension de la technique de la dorure au moyen de techniques scientiiques (MO, MEB/EDX, FTIR) et des informations obtenues par étude des traités anciens. Deux techniques sont principalement utilisées: la feuille composée (étain et or) et la simple feuille d’or ; dans les deux cas les feuilles sont ixées à la surface avec un adhésif huileux auquel sont ajoutés des pigments inorganiques, le plus souvent du blanc de plomb et des ocres, appelé missione (mixtion). Les résultats scientiiques trouvent une bonne correspondance avec les recettes mentionnées dans les traités anciens, en particulier dans le Libro dell’Arte de Cennino Cennini. En étendant les recherches à d’autres exemples dans le Tessin et en Italie, il paraît évident que la technique n’a pas été inluencée par le contexte historique ou géographique, mais par l’aspect inal de la surface murale employée. L’utilisation de l’étain dans la feuille composée permettait de brunir et graver la surface. Keywords: gilding, gold, missione, Tessin, tin, wall paintings. Mots-clés : dorure, or, missione (mixtion), Tessin, étain, peinture murale. * University of Applied Sciences of Southern Switzerland, Dept. Environment, Constructions and Design, Institute of Materials and Constructions Trevano – P.O. Box 12 CH-6952 Canobbio (Tessin). (giovanni.cavallo@supsi.ch) ** Architect and Conservator – Via San Nicolao 3 CH-6816 Bissone (Tessin). rec. Aug. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 409-415 Giovanni CAVALLO, Marcel VERDA 410 1. INTRODUCTION he present research was initiated during the conservation project of the wall painting representing the Cruciixion scene painted by Andrea De Passeri in 1513 in the St. Alessandro Church in Lasnigo (Lombardy, Northern Italy). Traces of the original gold foil, in the form of irregular and fragile lakes, were found on the halos of the Virgin and the adoring angels (Bonizzoni et al., 2007). Further in-depth studies (Cavallo and Verda, 2008) revealed in detail the technique that was adopted: a composite metal foil – gold and tin, probably ixed with glue – was adhered to the surface of the wall painting by means of a type of adhesive called missione. his adhesive was a mixture of linseed oil with the addition of pigments with drying capacity, such as lead white and white bones; yellow ochres were used in small amounts. he research was extended to the period between 1360 and 1520 in the Lombard lake region, corresponding in the past to the present day Tessin (Southern Switzerland) and Lombardy (NW Italy), when the city of Milan played a crucial historical role and became one of the most important centres of European culture. In the fourteenth century – Trecento – many famous artists, such as Giovanni di Benedetto from Como, Giovannino de’ Grassi, and Balbello from Pavia, worked for religious and aristocratic clients, also supported by the miniaturists and goldsmiths gathered under the Scuola of St. Eligio. he peak of this political and cultural life took place during the government of Gian Galeazzo Visconti, who extended the control in the North-eastern and Central Italy. After his death in 1402, a period of political instability forced many artists to move principally towards Eastern Italy. he stagnant artistic situation received a renewed impulse during the reign of the Sforza dynasty in the second half of the 15th century, and particularly under Ludovico il Moro and his wife, Beatrice d’Este, when Milan attracted important literates and artists, such as Bramante and Leonardo da Vinci. he decline of the visconteo-sforzesco government took place in 1500, when the French occupied Milan (Gregori, 1997; 1998). Returning to our research on gildings, the examined artworks are part of the aforementioned political and cultural panorama. he attention was focused on the lunette (Fig. 1) attributed to Giovanni da Milano (1360) preserved in the Santa Maria delle Grazie Oratory at Mendrisio (Tessin), and to the cycle of wall paintings in the church of St. Carlo at Prugiasco-Negrentino (Tessin), dating from the last decade of the 15th century, where the Antonio Da Tradate and Seregnesi studios painted the apses. he Cruciixion scene painted by Andrea De Passeri at Lasnigo (1513) is also part of this work. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 409-415 Figure 1: (See colour plate) he Oratory of Santa Maria delle Grazie (Mendrisio, Southern Switzerland). Detail of the lunette attributed to Giovanni da Milano (1360). Figure 1 : (Voir planche couleur) Oratoire de Santa Maria delle Grazie (Mendrisio, Tessin). Détail de la lunette attribuée à Giovanni da Milano (1360). he main goals of the research addressed the following questions: did the geographic and temporal contexts inluence the gilding technique? Was the gilding technique inluenced by other important studios in Italy, operating mainly in Tuscany? What kind of diferences and similarities was it possible to establish with the main studios in Italy? Are there correspondences between the analytical results and the ancient recipe books? Was the identiication of the original technique possible in the cases of overlapping layers, due to past restoration works? 2. METHODS Two approaches have been adopted in studying the gilding techniques and materials. he irst entails a critical reading of the ancient recipe books and the second the scientiic investigation of selected samples. Gilding on wall paintings from a period between the 14th-16th centuries AD in the Lombard lake region 411 Ancient recipe books Samples Location and description Ancient manuscripts report the materials and methods used for gilding. In the De Coloribus et artibus Romanorum, par. XLV (Alexander, 1967), Heraclius refers to the method used for gilding upon tin. In the Libro dell’Arte, par. XCVII (Brunello, 1971), Cennino Cennini mentions how to cut the gilt tin. De Diversis artibus by heophilus, par. XXI (Cafaro, 2000), the Libro dell’Arte by Cennini, par. XCI and XCII (Brunello, 1971), and the Esperimenta de Coloribus by Jehan Le Begue, par. 105 (Alexander, 1967) all report indications of the methods used in attaching the metal foils (composite foil). Finally, the De Coloribus faciendis by Petrus De S. Audemar, par. 190 (Alexander, 1967) refers to the technique employed for laying the gold foil on a wall. he interpretation of the manuscripts is diicult; the linguistic style, useful for the comprehension of the Medieval and Late Medieval outlook, is also a cause of terminological ambiguity, compromising the real interpretation of the text. An additional issue is the discontinuity of the information: some recipes are described carefully, as in the Libro dell’Arte by Cennini, others supericially. GM1 Mendrisio, lunette attributed to G. da Milano. Gilding fragment from the halo of St. Caterina. GM3b Mendrisio, lunette attributed to G. da Milano. Gilding fragment from the halo of the Virgin. NE1 Prugiasco-Negrentino, Seregnesi’s studio. Gilding fragment from the halo of the Apostle. NE3 Prugiasco-Negrentino, Da Tradate’s studio. Gilding fragment from the halo of the Virgin Mary. CR9 Lasnigo, Andrea De Passeri. Gilding fragment from the halo of the Virgin Mary. CR10 Lasnigo, Andrea De Passeri. Gilding fragment from St. Alessandro’s sword. Scientific investigations Scientiic investigations have been carried out on selected samples collected in Switzerland (Mendrisio and PrugiascoNegrentino) and Italy (Lasnigo). Optical microscopy (OM) of cross-sections in relected light (Plester, 1956) has been used to identify the stratigraphy, and for a preliminary characterization of each layer. Scanning Electron Microscopy (SEM) coupled with an EDS detector (Vega Tescan electron microscope) has been used for the determination of the elemental composition of each layer detected under the optical microscope. he instrumental setup was 20 kV, vacuum mode HV, working distance 9 mm. Infrared spectroscopy (FTIR) has been used for determining the organic (and inorganic) fraction of the samples; a Perkin Elmer Spectrum One BM working in the range 4000-600 cm-1 has been used. Table 1 reports the description of the studied samples. 3. RESULTS Optical microscopy (OM) Table 2 reports the cross-section stratigraphy (optical microscopy in incident light). Table 1: Samples, location and description. Tableau 1 : Échantillons, localisation et description. he data show two types of gildings; the use of composite and single foil. It should be taken into account that the samples collected at Mendriso and Negrentino come from restored gildings, while the one in Lasnigo is the original gilding. However, the stratigraphy corresponding to the sample NE1 (Fig. 2) contains information both on the original technique and the restored one. Infrared Spectroscopy (FTIR) Infrared Spectroscopy has been used for understanding the composition of the adhesive applied to attach the metal foil or the composite foil on the wall. All the examined samples show the presence of C-H absorption bands at 2918, 2845, 1736, 1460 and 1402 cm-1, C-O at 1243 and 1100 cm-1, and C-C at 723 cm-1, characteristic of linseed oil. Calcium carbonates show the characteristic bands at 1410, 873 and 712 cm-1; the presence of white lead is conirmed by the absorption bands at 3533, 1046, 681 cm-1. Traces of calcium oxalates (samples CRs) are marked by the characteristic absorption at 1320 cm-1. he reported data are in accordance with the use of missione, a type of adhesive made with linseed oil and pigments with drying capacity. Ca-oxalates could be referred to the mineralization of the animal glue used to attach the two metal foils. Microanalysis (SEM/EDS) Microanalysis has been carried out on each layer detected by OM, with the exclusion of the support layer (plaster, labelled 0); the results are reported in Table 3. he microanalysis clearly shows the presence of diferent techArcheoSciences, revue d’archéométrie, 33, 2009, p. 409-415 Giovanni CAVALLO, Marcel VERDA 412 Sample GM1 GM3b NE1 NE3 CR9 CR10 Layers Description 0 Lime-based plaster 1 Brown layer, thickness 100 µm. 2 Bright metal foil. 0 Lime-based plaster 1 Brown layer, thickness 100 µm. 2 Bright metal foil. 0 Lime-based plaster. 1 Brown layer, thickness 20-50 µm. 2 Non continuous grey-black layer, thickness 20 µm. 3 Red layer, thickness 30 µm. 4 Bright metal foil. 5 Red layer, thickness 30 µm. 0 Lime-based plaster. 1 Yellow to brownish layer, thickness 100 µm. 2 Bright metal foil. 1 Brownish layer, thickness 100 µm. 2 Grey-black layer, thickness 20 µm. 3 Irregular traces of a bright metal foil. 0 Lime-based plaster. 1 Brownish layer, thickness 100 µm. 2 Grey-black layer, thickness 20 µm. 3 Irregular traces of a bright metal foil. Figure 2: (See colour plate) Sample NE1: cross-section. Figure 2 : (Voir planche couleur) Échantillon NE1: coupe transversale. Table 2: Optical microscopy in incident light. Tableau 2 : Microscopie optique sous lumière incidente. nologies for gilding adopted by the artists working in the Lombard lake region. he presence of the composite foil (Sn plus Au) on the original gilding decorating the halos at Lasnigo (samples CRs) applied on a layer containing Pb-based pigments was determined. Here, the presence of P and Ca is associated with bones white. Tin thickness ranges from 12 to 23 µm; the metal shows longitudinal and transversal micro-cracks due to processing. Gold thickness ranges from 1.50 to 3.00 µm; the metal exhibits an irregular surface and is locally detached from the tin metal foil. Samples collected at Mendrisio (Fig. 3) show the application of the gold foil directly on the oil-based adhesive. Gold foil thickness is around 1.20 µm; the spatial continuity is well preserved. Samples from Negrentino (NEs) show two distinct technologies; the use of the gold foil (thickness around 1.20-1.50 µm) on the murals painted by the Antonio da Tradate studio (Fig. 4) and tin metal foil (thickness ranging from 10.5 to 20.5 µm), attesting the use of the composite foil in the original gilding on the surface painted by the Seregnesi studio. his gilding was restored twice, with a notable change of the technology. A brass foil ArcheoSciences, revue d’archéométrie, 33, 2009, p. 409-415 Figure 3: Sample GM3b: BSE picture with indication of the investigated areas. Figure 3 : Échantillon GM3b : image BSE avec indication de la surface étudiée. (Cu+Zn+Fe), with thickness ranging from 1.0 to 4.5 µm, was applied on the bole (aluminium silicate). A later restoration work shows the use of gold with a thickness around 1.30 µm on a new layer of bole. 4. DISCUSSION AND CONCLUSION he analytical data, integrated with information from the ancient manuscripts and other case studies referred in the Gilding on wall paintings from a period between the 14th-16th centuries AD in the Lombard lake region Sample GM1 GM3b NE1 NE3 CR9 CR10 Layer Chemical elemental composition Ca Mg Al Si Pb Fe Cu Au Ag Sn Zn P tr tr tr + +++ tr tr - - - - - 2 - tr tr - - tr tr +++ tr - - - 1 + tr tr + +++ + - - - - - - 2 tr tr tr tr - tr tr +++ - - - - 2 - - - - - - - - - +++ - - 3 + - + +++ - tr - - - - - - 4 - - - - - tr +++ - - - ++ - 5 + - + +++ - tr - - - - - - 6 - - - - - - - +++ - - - - 1 +++ - - tr +++ tr - - - - - - 2 - - - - - - - +++ - - - - 1 ++ tr - tr +++ + - - - - - ++ 2 - - - - - - - - - +++ - - 1 413 3 - - - - - - - +++ - - - - 1 ++ tr - tr +++ + - - - - - ++ 2 - - - - - - - - - +++ - - 3 - - - - - - - +++ - - - - Figure 4: Sample NE3: BSE picture with indication of the investigated areas. Figure 4 : Échantillon NE3 : image BSE avec indication de la surface étudiée. specialist literature, allow us to deine the gilding techniques in the period under consideration, between the irst half of the 14th century and the beginning of the 16th century. Composite foil has been used in the Cruciixion scene painted by Andrea De Passeri in Lasnigo (1513). Indications on the use of the composite foil have also been detected Table 3: Qualitative chemical elemental composition of the gildings layers (+ + + high; + + medium; + low; tr traces). Tableau 3 : Composition chimique élémentaire qualitative des couches des dorures (+ + + fort ; + + moyen ; + bas ; tr traces). in the stratigraphy of the wall paintings executed by the Seregnesi in Negrentino (last decade of the 15th century), where the tin foil is applied on the oil-based missione. We may suggest that the gold, or another precious metal such as silver, fell down because of the loss in binding power of the animal glue between the gold and the tin; according to Matteini and Moles (1990), the thickness of the animal glue was of the order of a few microns. Other famous cases in the same area are the Teodolinda Chapel in the St. Giovanni Battista Cathedral in Monza – Tales of the Queen Teodolinda painted by Franceschino, Gregorio and Giovanni Zavattari in 1444 (Cassanelli and Conti, 1991) – and the Baptist Tales in the Castiglione Olona Baptistery painted by Masolino da Panicale in 1435 (Marani, 1997). he use of tin is also mentioned on the Tabernacolo in via del Leone in Florence, painted in 1356 by Tommaso di Stefano, called Giottino (Matteini and Moles, 1982), the Trinità painted by Masaccio in Santa Maria Novella in Florence between 1425 and 1427, and the St. Giovanni Baptistery in Siena between 1447 and 1489. he technological process of the composite foil applied on the wall by means of an oil-based missione is recommended by Cennino Cennini (Brunello, 1971). he use of a double metal foil is required for gold burnishing: a strong burnishing (Alexander, 1967) could produce abrasions of the gold foil, as suggested by St. Audemar; therefore, the tin foil was used to attenuate this inconvenience. According to Heraclius (Alexander, 1967), the process employed to ix ArcheoSciences, revue d’archéométrie, 33, 2009, p. 409-415 Giovanni CAVALLO, Marcel VERDA 414 the gold with tin was carried out on a desk, using a suitable wooden board. he use of the single foil applied on an oil-based missione layer has been detected at Negrentino (on the works of Da Tradate’s studio and on the lunette attributed to Giovanni da Milano, 1360, at Mendrisio). We should take into consideration that at these two sites, restoration works have been carried out during the middle of the 20th century. Considering the thickness of the gold foil, its recent use appears clear; the thickness ranges between 1.20 and 1.50 µm, lower than that of the gold foil in the Lasnigo church, where the metal has a thickness up to 3.00 µm. Other examples of gildings with a single foil are the Passion of Christ on the chancel arch in Santa Maria delle Grazie Bellinzona, Tessin (end of the 15th century), attributed to Foppa’s studio (Moles, 1998), and in the Cathedral of Monza (Matteini and Moles, 1991). he presence of wax mixed with resin as an adhesive for the composite foil has been detected only on the wall painting representing the Last Supper (uncertain attribution), painted around 1520 in the St. Ambrogio Church at Ponte Capriasca (Gilardi, 1993). During the period towards the end of the 14th century, and particularly during the 15th century, the interest of the aristocracy was oriented towards a court sensibility; this new attitude will condition the choices of many Lombard artists invited to recreate on the walls the magical atmosphere of the illuminated manuscripts. he artists were asked to use the gold on wall paintings; they were aware of the great technical problems linked with the use of metal foils on the wall. he recipe books could play an important role for solving such problems. he actual distribution of these books is an open issue; the costs for copying were high, and sometimes the contents, moving between diferent ields, such as medical science and alchemy to artistic practice, coupled with the diferent idioms being used, and the limitations deriving from the translations, limited their accessibility to the artists. It is more probable that oral tradition was the main vehicle employed to hand down past practices, also supported by the skill achieved during training in other studios. In conclusion, the microanalysis of selected cross-sections was very useful for the identiication of the gilding technique, especially when integrated with infrared spectroscopy carried out on the adhesive layers. his qualitative information is per se suicient for understanding the gilding technique; a quantitative approach is very important when provenance studies of the raw materials, characterizations of the compositional homogeneity versus heterogeneity of the metals, or their typological classiication are fundamental issues for the research. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 409-415 he authenticity of the adopted techniques is very clear for some of the investigated artefacts; the thickness of the gold foil represents a key factor for distinguishing between original and renewed gildings. he use of non-invasive analytical methods such as portable µ-XRF represents a practical way for determining the thickness of the gold foil. Furthermore, restored gildings sometimes reveal the marker of the original technique, as conirmed by the use of tin metal; in these cases, a cautious approach should be adopted during the interpretation of the analytical data. Finally, it seems that the use of a gilding technique rather than another is not inluenced either by chronology or by the regional context: the choice was depending on the inal appearance of the gilding on the wall painting and the options available for processing. References ALEXANDER, M. (ed.), 1967. Original treatises on the arts of painting. New York, Dover Publications. BONIZZONI, L., BRUNI, S., GARGANO, M., GUGLIELMI, V., LUDWIG, N., MONETA, A. and MILAZZO, M., 2007. Indagini sugli afreschi della chiesa di S. Alessandro a Lasnigo (CO). Atti del 4. Congresso Nazionale di Archeometria, Bologna, Patron Ed., 115-127. BRUNELLO, F. (ed.), 1971. Il libro dell’arte di Cennino Cennini. Vicenza: Ed Neri Pozza. CAFFARO, A. (ed.), 2000. Teoilo Monaco. De diversis artibus. Salerno, Ed. Palladio. CASSANELLI R. and CONTI, R., 1991. Monza, La cappella di Teodelinda nel Duomo. Milano, Ed. Electa. CAVALLO, G. and VERDA, M., 2008. Historical, microanalytical evidences and limits of non-invasive technologies in studying gildings on 16th century wall paintings, in CD Proceedings of the IX International Conference ART2008, Jerusalem. GREGORI, M., 1997. Pittura a Milano dall’Alto Medioevo al Tardogotico. Milano, Ed. Cassa di Risparmio delle Provincie Lombarde. GREGORI, M., 1998. Pittura a Milano, Rinascimento e Manierismo. Milano, Ed. Cassa di Risparmio delle Provincie Lombarde. GILARDI, A., 1993. Ponte Capriasca. Il Cenacolo restaurato. Lugano, Ed. Fondazione H. Dietler-Kottmann. MARANI, P. C., 1997. Conservazione e valorizzazione degli afreschi nella Provincia di Varese, in Proceedings of the conference held in Varese in 1995, Provincia di Varese. MATTEINI, M. and MOLES, A., 1982. Tommaso di Stefano detto Giottino, Madonna col Bambino in U. Baldini (ed.), Metodo e scienza, Firenze, Ed. Sansoni. Gilding on wall paintings from a period between the 14th-16th centuries AD in the Lombard lake region MATTEINI, M. and MOLES, A., 1990. Le tecniche di doratura nella pittura murale, in C. Danti C., M. Matteini, A. Moles (eds.), Le pitture murali: tecniche, problemi, conservazione. Firenze, Ed. Centro Di, 121-126. MATTEINI, M. and MOLES, A., 1991. Le “Storie” di Teodelinda nel ciclo degli Zavattari: alcune indagini preliminari sulla natura dei materiali pittorici e sulla loro collocazione stratigraica, in 415 R. Cassanelli, R. Conti, Monza, La cappella di Teodelinda nel Duomo. Milano, Ed. Electa, 164-171. MOLES, A., 1998. Analisi chimiche e strumentali preliminari alle prove di pulitura. 1st Report. Bellinzona, Uicio Beni Culturali. PLESTER, J., 1956. Cross-section and chemical analyses of paint samples, Studies in Conservation 2: 110-157. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 409-415 Gold leaf analysis of three baroque altarpieces from Porto Analyse de feuilles d’or de trois retables baroques de la ville de Porto Ana Bidarra*, João Coroado** and Fernando Rocha* Abstract: During the second half of the 17th century and the irst half of the 18th century, Portugal experienced a rising altarpiece production and an increased taste for gold application in church decoration. Some examples of this tendency can be found in the northwest of Portugal, particularly in the baroque altarpiece productions of Porto, Braga, Vila Real and Viana do Castelo. Porto was the irst area studied and three gilded wood-carved altarpieces from diferent churches were selected: São Bento da Vitória (SB), Santa Clara (SC) and São Francisco (SF). Qualitative and semi-quantitative analyses of the samples were performed by optical microscopy (OM) of the cross-section using relected and polarised light, and by scanning electron microscopy (SEM) in combination with energy dispersive spectroscopy (EDS). he main advantages and disadvantages of the use of SEM-EDS were discussed and the results regarding the presence of major and trace elements compared. Résumé : Pendant la deuxième moitié du XVIIe siècle et la première moitié du XVIIIe siècle, au Portugal, est observée une production croissante de retables ainsi qu’une préférence pour l’application d’or dans la décoration des églises. Quelques exemples de cette tendance peuvent être observés dans le NordOuest du Portugal, en particulier dans le cas des retables produits à Porto, Braga, Vila Real et Viana do Castelo. Porto a été la première région étudiée et trois retables en bois doré provenant de diférentes églises ont été sélectionnés : São Bento da Vitória (SB), Santa Clara (SC) et São Francisco (SF). Des analyses qualitatives et semi-quantitatives de prélèvements ont été réalisées par microscopie optique (MO) de sections, en lumière réléchie et polarisée, et par microscopie électronique à balayage (MEB) combinée à un système de spectroscopie en énergie dispersive (EDS). Les avantages et inconvénients de l’utilisation du MEB-EDS sont discutées et les résultats concernant la présence d’éléments majeurs et traces comparés. Keywords: Gold leaf, baroque altarpieces, major and minor elements, SEM-EDS. Mots-clés : feuille d’or, retables baroques, éléments majeurs, éléments traces, MEB-EDS. 1. INTRODUCTION Baroque art was one of the most convincing ways used by the church to defend and impose its faith. It was an example of scenic art, where light, movement and sound have key roles, appealing to the senses (Ferreira-Alves, 1989). One of the most magniicent art forms took the shape of gilded wood-carved altarpieces covering partially or totally the interior of churches, in complex and structured architectures, working alone or in association with ceramic tiles (azulejos) or paintings. he craftsmen involved in the process were masters of this art. he carvers, sculptors, painters or gilders were well organised, and diferent workshops competed with each other to produce the best works. * Departamento de Geociências and GeoBioTec Research Centre – Universidade de Aveiro, Campus Universitário de Santiago, PT, 3810-193 Aveiro, Portugal. (anabidarra@portugalmail.com), (frocha@geo.ua.pt) ** Departamento de Arte, Conservação e Restauro and GeoBioTec Research Centre – Instituto Politécnico de Tomar, Quinta do Contador, Estrada da Serra, PT, 2300-313 Tomar, Portugal. (jcoroado@ipt.pt) rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 417-421 418 The city of Porto was one of the most important Portuguese production centres and had a great inluence in the surrounding regions. According to Robert C. Smith’s chronology, the Portuguese baroque wood working can be divided in two major periods: barroco nacional (c. 16801729) and barroco joanino (c. 1729-1750) (Smith, 1962). An example from the irst period is the altarpiece from São Bento da Vitória, while examples of works from the second period are the São Francisco and Santa Clara altarpieces. Although there are several studies on Portuguese altarpieces from this period, addressing national or regional productions, these researches are focused on the typological, social, economic and religious aspects of these works. In terms of the study of the materials and techniques applied, the approach used to date is mainly based on the analysis of documents, such as treatises and orders from the manufacturing period, with references to the materials (wood, gold, pigments, varnishes) and to where and how they should be applied (Ferreira-Alves, 1989; 2004). However, there is a gap related to a direct study, focused exclusively on the gold leaf and its origin and characterisation. he present work aims to provide a new approach to Portuguese baroque altarpieces, by focusing on the study of the gold leaf and its elemental analysis, and identifying a distinct ingerprint. his will be accomplished by the determination of the concentration of speciic minor and trace elements, which should be characteristic of the metal and its provenance: platinum and palladium contents in South American gold ores (Colombian and Brazilian - Peruvian gold with lower contents); platinum contents in contemporary Eastern European gold; palladium, tin and antimony in other Latin American gold ores; and trace elements typical of telluride in gold from India after the 16th century (Guerra and Calligaro, 2004). Due to the fact that most of the gold sources from the past are geo-chemically unknown, the accurate identiication of the characteristic elements is a diicult task (Guerra, 2004). he evolution of gold processing is another obstacle, since the addition of other metals to improve the workability of the material, to reduce costs, or even to change colour, might contribute to the loss of information provided by trace elements. 2. METHODS he samples were selected from discrete but representative areas in the main altarpieces, such as the back of the columns or the steps of the throne. he collected samples came from high and hidden places, diicult to reach, where contamination should be at the minimum. For cross-secArcheoSciences, revue d’archéométrie, 33, 2009, p. 417-421 Ana BIDARRA, João COROADO, Fernando ROCHA tion analysis, the samples were mounted in polyester resin (BYLAPOX 3085 A and B (2:1)) and were polished using a Struers Planopol-V device. he observation and photography of the samples (100x magniication) were obtained using optical microscopy (OM), with polarised and transmitted light (Zeiss Stemi, 2000-C), with an external artiicial light system Zeiss KL 1500 LCD, image acquisition with digital camera AxioCam MRcS, and an Axio Vs 40 V4.4 Carl Zeiss Vision GmbH acquisition and treatment software. he mounting and observation of the samples followed the procedures applied to the analysis of easel paintings or polychrome sculptures (Khandekar, 2003). he study of the cross-sections allows determining the number, thickness, adhesion and cohesion of layers, the shape and size of the particles (Calvo, 2003), and the presence of more than one gilding; it also allows comparing the ground layer and the bole layer. For the scanning electron microscopy (SEM), a Hitachi SU-70 equipped with a Bruker AXS energy dispersive spectrometer (EDS) with a Quantax 400 EDS system with liquid nitrogen free and XFlash Silicon Drift Detector (SDD) was used. he semi-quantitative results were based on a peak-to-background ZAF evaluation method (P/BZAF), ZAF representing a matrix correction, mainly based on analytical expressions for atomic number (Z), dependent X-ray yield, self-absorption (A), and secondary luorescence enhancement (F), provided by the Esprit software. A 15KV acceleration voltage and a current intensity of 32 mA were applied. he samples were coated with carbon. he use of SEM in combination with EDS increases sensitivity for lighter elements, and the spatial resolution for spot analysis is higher. Additionally, it allows the examination of the microstructure with line scans or two-dimensional mappings of element concentrations (Hein and Degrigny, 2008; Guerra, 2004). he mapping acquisition was carried out under the same conditions as the EDS spectrum, with no background elimination. Two sets of samples were analysed in terms of the study of the cross-sections and the surface. In the irst case, the study was focused on the bole and ground layers (elemental results not presented), and in the second case the study was focused on the gold and its physical and chemical aspects, such as homogeneity, abrasion, voids and elemental distribution. he areas were selected in terms of their homogeneity and lack of voids; a 300x400 µm area was scanned using a 300x magniication, and the elemental and semi-quantitative results were obtained after three measurements. he semi-quantitative results were normalised to 100%. Gold leaf analysis of three baroque altarpieces from Porto 3. RESULTS he OM and SEM analyses revealed that the gold leaf from the SB, SC and SF altarpieces share important features, such as surface morphology, as well as extreme thinness and uniformity. he presence of ground and bole layers was visible, typical of traditional gilding (Martínez, 1997; SerckDewaide et al., 2004), with good adhesion and cohesion between them (Fig. 1). he analysis of SC samples raised several diiculties, due to the considerable presence of wax over the gold layer. he main diferences were obtained in the EDS analysis; for the determination of the gold/silver/ copper alloy composition, the results were normalised to 100%, revealing the use of a very pure gold (between 22 and 23K). he most signiicant diferences were identiied in the relative silver percentage, varying between 4.4% (SF) and 0.9% (SC) (Table 1). he major elements were common to all samples (Fig. 2); aluminum and silicon were components of the bole layer, a clay layer rich in aluminum-silicates and iron. he results obtained from the study of minor elements should be carefully analysed, as due to the lower concentrations, the readings could lead to error. For instance, when studying Table 2, the sample SB revealed the presence of platinum; however, the high concentration of gold and the proximity of Lα1 lines (9.711 KeV of Au and 9.441 KeV of Pt) and all other lines of both elements could be responsible for the overlap of the platinum peacks. he same thing happened when analysing SB silver and palladium (2.984 KeV and 2.838 KeV, respectively, for Lα1 lines), as both picks were practically overlapping. he same interpretation was given for the relatively high percentage of mercury in the SC sample: the mercury peaks were near the gold peaks (9.987 KeV and 419 9.711 KeV, for Lα1 lines of Hg and Au, respectively), and the automatic values were generated by the software deconvolution of the results (Goldstein, 1975). By analysing the EDS data, it was possible to characterise the gold from SB as a 22K gold and to identify the presence of palladium and tellurium; SC had the highest carat, almost 23, with the presence of palladium and titanium and SF samples is a 22K gold with the presence of minor elements such as tin, tellurium and zinc. he mapping of the surface provided the topographic distribution of selected elements in a sample, with brighter areas where the element concentration is higher and faded areas for the lower concentrations. For SB samples, the gold, silver and copper distribution was clear (Fig. 3). When mapping SC for gold, silver and palladium, the distribution showed a clear distribution of gold and silver, and a palladium distribution with background interference. 4. DISCUSSION AND CONCLUSION he major elements of gold alloys – gold, silver and copper – can be determined using the SEM-EDS technique. However, the interpretation of trace elements such as palladium, platinum, tellurium, tin, antimony or zinc raises several doubts concerning the certainty of their presence. his uncertainty is due not only to the very low concentration of these elements, but also to the interpretation of the spectrum peaks, where some elements such as platinum or tellurium have very close peaks to gold and calcium, making an accurate interpretation of the results very diicult. Other elements, such as ruthenium, rhodium, chromium or lead, were not detected. he presence of silicon, aluminum, iron, Figure 1: SEM image (2000 x). Crosssection of samples from SB (a), SC (b), and SF (c). 1 – Gold leaf; 2 – Bole layer; 3 – Ground layer. Figure 1: Image MEB (x2000). Section de prélèvements de SB (a), SC (b), et SF (c). 1– feuille d’or ; 2 – bol ; 3 – souscouche Table 1: Concentration of the alloy composition and gold carat in relative percentages (SB – São Bento; SC – Santa Clara; SF – São Francisco). Tableau 1 : Composition des alliages et teneurs en or en pourcentages relatifs (SB – São Bento ; SC – Santa Clara ; SF – São Francisco). Sample Au % Ag % Cu % Carat SB 91.99 2.9 5.11 22.08 SC 94.91 0.91 4.18 22.78 SF 91.65 4.41 3.94 22 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 417-421 Ana BIDARRA, João COROADO, Fernando ROCHA 420 Figure 2: (See colour plate) EDS spectrum of major elements. Figure 2: (Voir planche couleur) Spectre EDS des éléments majeurs Figure 3: (See colour plate) São Bento (SB): SEM image (a) and EDS mapping of the surface – Au (b), Ag (c), and Cu (d) (300x). Figure 3: (Voir planche couleur) São Bento (SB) : Image MEB (a) et cartographie EDS de la surface – Au (b), Ag (c) et Cu (d) (300x). Sample Hg wt% Error % Pd wt% Error % Pt wt% Error % Sn wt% Error % Te wt% Error % Ti wt% Error % Zn wt% Error % SB nd — 0.96 0.6 4.96 0.9 nd — 1.23 0.2 nd — nd SC 3.17 0.8 1.02 0.7 nd — nd — nd — 1.31 0.1 nd — — SF nd — nd — nd — 2.91 0.6 1.45 0.2 nd — 2.10 0.2 Table 2: Weight percentage of minor elements – comparison between relative percentages (SB – São Bento; CS – Santa Clara; SF – São Francisco). Tableau 2 : Concentrations en pourcent des éléments mineurs – comparaison entre les pourcentages relatifs (SB – São Bento; SC – Santa Clara; SF – São Francisco). calcium and potassium were mainly contributions from the bole and ground layers. A spot analysis reduces the percentage of these elements, but not in a signiicant way. Since the results concerning the study of trace elements are not satisfactory, other procedures are now in order, such as inductiArcheoSciences, revue d’archéométrie, 33, 2009, p. 417-421 vely coupled plasma mass spectrometry (ICP-MS) and the use of synchrotron radiation (SR). Overall, this study is part of a larger project that aims to study the compositional characterisation of the gold leaf, the typiication of its aging, and to establish provenance Gold leaf analysis of three baroque altarpieces from Porto relations, allowing a better approach to the conservation and restoration of these coatings. hrough this study, it will be possible to complement the information concerning the three selected altarpieces, and also to compare them with other altarpieces where the same masters have worked. Acknowledgements Portuguese Foundation for Science and Technology (FCT) for PhD grant FRH/BD/38593/2007. GeoBioTec Research Centre – Aveiro. Technological Centre of Ceramic and Glass (CTCV) – Coimbra. References CALVO, A., 2003. Conservación y Restauración - Materiales técnicas y procedimientos – De la A a la Z. Tercera edición. Barcelona, Ediciones del Serbal. FERREIRA-ALVES, N.M., 1989. A Arte da Talha no Porto na Época Barroca. Artistas e clientela. Materiais e técnica. Vol.I e II. Porto, Arquivo Histórico, Câmara Municipal do Porto. FERREIRA-ALVES, N.M., 2004. O douramento e a policromia no Norte de Portugal à luz da documentação dos séculos XVII e XVII. Policromia – A escultura policromada religiosa dos séculos XVII e XVIII. Estudo comparativo das técnicas, alterações e conservação em Portugal, Espanha e Bélgica. Actas do Congresso Internacional, Lisboa 29, 30 e 31 de Outubro de 2002. Lisboa, Instituto Português de Conservação e Restauro, 17-22. GOLDSTEIN, J.I., 1975. Electron beam – specimen interaction, in J.I. Goldstein and H. Yakowitz (eds.), Practical scanning elec- 421 tron microscopy. Electron and ion microprobe analysis. New York, Plenum Press, 49-94. GUERRA, M.F., 2004. Fingerprinting ancient gold with proton beams of diferent energies, Nuclear Instruments and Methods in Physics Research Serie B 226: 185-198. GUERRA, M.F. and CALLIGARO, T., 2004. Gold traces to trace gold, Journal of Archaeological Science 31: 1199-1208. HEIN, A. and DEGRIGNY, C., 2008. he application of nondestructive technologies for the damage assessment of metal objects, in V. Argyropoulos (ed.), Metals and Museums in the Mediterranean, Protecting, Preserving and Interpreting. Athens, he PROMET Consortium, 125-140. KHANDEKAR, N., 2003. Preparation of cross sections from easel paintings, Reviews in Conservation 4: 52-64. MARTÍNEZ, E., 1997. Tratado del dorado, plateado y su policromia. Tecnología, conservación y restauración. Valencia, Departamento de Conservación y Restauración de Bienes Culturales, Universidad Politécnica Valência. SERCK-DEWAIDE, M., RABELO, E., SANYOVA, J., RIBEIRO, I., BARREIRO, A., ROMÃO, P., LE GAC, A., NARANJO, M.C., GONZALEZ, M.J., LOPEZ, A.C., ARCAUTE, E.R., RAMOS, R.G., OLANO, M.B., SALVARREDI, I.B., IBARRARAN, E.M., 2004. Les techniques utilisées dans l’art baroque religioux des xviie et xviiie siècles au Portugal, en Espagne et en Belgique, in A.I. Seruya (ed.), Policromia – A escultura policromada religiosa dos séculos XVII e XVIII. Estudo comparativo das técnicas, alterações e conservação em Portugal, Espanha e Bélgica. Actas do Congresso Internacional, Lisboa 29, 30 e 31 de Outubro de 2002. Lisboa, Instituto Português de Conservação e Restauro, 119-155. SMITH, R., 1962. A talha em Portugal. Lisboa, Livros Horizonte. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 417-421 he main altarpiece of the Old Cathedral of Coimbra (Portugal): characterization of gold alloys used for gilding from 1500 to 1900 Le retable majeur de l’ancienne cathédrale de Coimbra (Portugal): caractérisation des alliages d’or employés dans la dorure de 1500 à 1900 Agnès Le Gac*, Ana Isabel Seruya*, Michel Lefftz** and Adília Alarcão*** Abstract: his paper surveys gilding methods and materials found in the original polychromy and three subsequent coatings applied to the main altarpiece of the Old Cathedral of Coimbra, between 1502 and 1900. Twenty samples from gilded surfaces were examined with optical and scanning electron microscopy for leaf thickness, and analyzed with energy-dispersive X-ray spectroscopy for alloy composition (semi-quantitative evaluation). By crosschecking the documentary data (archives and technical sources) with the analytical data, the results point out several aspects: 1) two recurrent techniques can be distinguished: water-based burnished gilding on bole, and oil-based matte gilding on mordant; 2) gold leaf thickness is clearly below one micron over time; 3) in 1502, 1583 and 1685, the hand-beaten gold corresponds to the highest grade of the metal (up to 23 carat gold). his degree of ineness corresponds to the required purity and its transcendental meaning in the religious context in which the precious metal was used. In 1900, the ternary alloy found, equivalent to 20 carat gold, is more pertinent with the restoration then carried out and perhaps with the need to save money. Research on the concentrations of characteristic trace elements present in the gold alloys is being pursued. Résumé : Le retable majeur de la Sé Velha de Coimbra a fait l’objet de quatre revêtements polychromes, parfaitement datés, de 1502 à 1900. Les dorures qui les caractérisent ont été étudiées à partir de vingt prélèvements, pour évaluer l’épaisseur des feuilles d’or (microscopie optique et électronique à balayage), la nature et la composition des alliages d’or (spectroscopie de rayons X à dispersion d’énergie). En croisant l’ensemble des données historiques, techniques et analytiques, il est possible de tirer plusieurs conclusions : 1) deux techniques de dorure sont récurrentes : la dorure brunie, à l’eau et appliquée sur bol, et la dorure mate, à l’huile et appliquée sur « or de couleur » ; 2) l’épaisseur des feuilles d’or est nettement inférieure au micron quelle que soit l’époque considérée ; 3) en 1502, 1583 et 1685, l’or battu est in, supérieur à un or 23 carats. Ce titre est parfaitement en accord avec la pureté que l’on attendait alors du précieux métal et l’expression de transcendance dont celui-ci était le symbole. En 1900, l’alliage ternaire employé, à peu près équivalent à un or 20 carats, répond davantage aux besoins de la restauration entreprise à cette date et, peut-être, à un souci d’économie. Une recherche sur la concentration des éléments traces, caractéristiques de l’or suivant sa provenance, est en cours. Keywords: Altarpiece, analysis, gilding, gold alloys, manufacturing technologies, polychromy. Mots-clés : Alliages d’or, analyses, dorure, polychromie, retable, techniques anciennes. * Departamento de Conservação e Restauro, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa – Campus da Caparica, 2829-516 Caparica, Portugal. (alg@fct.unl.pt) ** Facultés Universitaires Notre-Dame-de-la-Paix – Rue de Bruxelles 61, B-5000 Namur, Belgique. *** Museu Nacional Machado de Castro – Av. António José de Almeida, 208-RC-Esq., Coimbra, Portugal. rec. Sept. 2009 ; acc. Nov. 2009 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 423-432 424 Agnès LE GAC et al. 1. INTRODUCTION One of the most prestigious creations of its time, the main altarpiece of the Old Cathedral of Coimbra (Portugal) was commissioned in 1499 by Bishop D. Jorge de Almeida. he work was carried out by two migrant masters who came from Southern Netherlands, but worked in the Iberian Peninsula: Olivier de Gand (sculptor) and Jean d’Ypres (painter-gilder) (Fig. 1). his imposing Portuguese-Flemish altarpiece (thirteen meters high) still exhibits the colour scheme of the 16th century, which used abundant gold and blue, typical of lamboyant Gothic. In fact, it was totally or partially coated four times over the ive hundred years of its existence. Research into the historical records allows us to date each campaign very precisely: in 1502 (gothic and original polychromy) (Garcia, 1923), in 1583 (refurbishment restricted to a few parts) (Garcia, 1923), in 1685 (new baroque polychromy) (Louro, 1983; Le Gac, 2004a) and in 1900 (complete restoration) (Vasconcelos, 1930). Gold coatings were applied systematically during these four periods and, in all cases, in the form of gold leaf. he use of very thin hand-beaten gold leaf 1 was already established on wooden sculptures in Northern Europe during the 11th century (Serck-Dewaide, 1991). Numerous publications concerning sacred polychrome artworks contain reports of the use of metallic leaves applied on wooden substrates. he collective work entitled Gilded Wood (Bigelow et al., 1991) remains the seminal publication in this ield. However, studies focussing on the quality of the alloys, on the beating process, and on the thickness remain rather scarce, despite the pioneering work undertaken in Portugal by Natalia Alves 2 (1989), Maria Teresa Marques 3 (1998) and Isabel Ribeiro (Moura et al., 2004). he main altarpiece of Coimbra provides us with an excellent opportunity to examine the trade practices of gilders in Portugal between 1500 and 1900, and more speciically: 1) To study the diferent techniques applied for gilding the altarpiece. 1. Several specialists who talk about hand-beaten gold (Bigelow et al., 1991) point out two terms in English: ‘foil’, to refer to thin metal that can support itself, holds its shape when bent, and may be attached mechanically; and ‘leaf ’, to refer to even thinner limp metal that requires some adhesive to be laid on a substrate. 2. Natália Alves compiled many deeds executed by notaries related to the gilding of altarpieces in the North of Portugal in the 17 th and 18th centuries, and found several technical aspects by crosschecking the data with goldbeaters’ practices. his documentary investigation should be confronted with the tangible aspects of the altarpieces in question. 3. his research, dealing with the qualitative and quantitative analysis by TEM-EDS of the gold alloys used in Portugal on several baroque altarpieces, has unfortunately not been published. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 423-432 Figure 1: (See colour plate) Main altarpiece of the Old Cathedral of Coimbra (1502). Figure 1 : (Voir planche couleur) Retable majeur de la Sé Velha de Coimbra (1502) © J. Pessoa. he main altarpiece of the Old Cathedral of Coimbra (Portugal)… 2) To verify if the sets of technical particulars from each coating are pertinent to the written historical records and sources on gilding. In addition to medieval and Renaissance treatises, consulted for Western European practices, priority was given to ordinances and contracts written in Portugal from the 16th to the 19th century. 3) To search for information on the gold products, as well as on their speciic output. 4) To study the caratage (ct) and thickness of the gold leaf wherever possible, taking into account the fact that only traces of the original polychromy and its refurbishment of 1583 remain today, as they were almost completely stripped of in 1685. he baroque re-polychromy, which is the stage most visible today, together with the restoration carried out in 1900 are therefore the two interventions which best allow systematic research. 2. METHODS Twenty samples were taken from the gilded parts of the altarpiece (architectural structure, several statuettes, and statues from the irst and third tiers). hey were mounted in synthetic resin and prepared as cross-sections. All layers were irst examined by optical microscopy and scanning electron microscopy (SEM) for high magniication images. he composition of the gold alloys was ascertained using energydispersive X-ray spectroscopy (EDX), which permits major element point analysis and semi-quantitative evaluations. Because the samples were very small (0.5-1 mm3) and the layers so well adhered to one another, the gold leaf could not be separated 4 from the surrounding materials, either below (such as bole) or above (such as paint: lead white, verdigris, vermilion). We therefore tried to ascertain the compounds of these surrounding layers by X-ray difraction (XRD) and by SEM-EDX, and to consider how their elemental composition (Ca, Fe, Pb, Hg, Cu, Sb, Ti, etc.) might interfere with that of the gold leaf and with the level of impurities found within the gold. 3. RESULTS AND DISCUSSION Techniques Depending on the type of surface to be covered (architectural structure or statuary of the altarpiece), the means 4. By washing, as M.T. Marques (1998) did when studying by TEMEDS gold leafs sampled from baroque altarpieces. 425 employed, and the inal efect required, two recurrent gilding techniques can be distinguished: l) Water gilding, on bole or ‘assiette à dorer’, composed of ochre and/or clays bound in a proteinaceous medium (usually an animal glue or egg white), whose presence is meant to allow the gold leaf to be ixed by moistening the bole. Once dry, the gilded part can be polished with an animal tooth or an agate stone to obtain a brilliant and shiny inish, hence its name of ‘burnished gilding’. For this method, the wooden substrate has to be properly primed with a thick white ground, which ofers better compressive strength during the burnishing. In Southern Europe, such a whiting consists of calcium sulphate in its anhydrite form (gesso) and/ or bihydrite form (gypsum), applied with a proteinaceous binder. 2) Oil gilding, over mordant or ‘or de couleur’, composed of a fat layer made of a drying-oil containing pigments rich in metallic oxides. he gold leaf was laid on when this layer was nearly dry. It cannot be polished, hence its generally accepted name of ‘matte gilding’ 5. his technique requires less care in preparation and execution. In both cases, there will be an inluence on the tone of the gold from the coloured underlayers. In 1502 we observe: 1) water gilding on the structure of the altarpiece, where the gold leaf is applied to an extremely ine bole of a very pale orange colour (Fig. 2); 2) oil gilding on tin leaf in the production of ‘appliqué relief brocades’ 6 which still exist in the niches in the irst, third and fourth tier, and that perhaps once also covered the rich garments of the igures. Here, the gold leaf is laid on a translucent and reddish oil layer containing lead-based pigments. While a translucent red glaze, essentially organic, brings out the contrast of the brocade pattern in the irst tier, we found evidence of a translucent greenish copper glaze in one crosssection from the third tier. In 1583, we observe only oil gilding, applied on the architectural structure without prior whiting, to refresh some parts. he topography of the gold leaf is very irregular. It has been attached to a thick dark red layer containing coarsely ground lead-based pigments, clays and iron-oxides (Fig. 2). 5. Even though the surface of a non-burnished water gilding could achieve a poor relective efect as well, leaving a low gloss surface, sometimes quite dull, without any brilliancy. 6. ‘Appliqué relief brocade’ refers to both the imitation of sumptuous brocades and the technique itself, by reproducing the relief design and gluing it to wooden and gessoed panels or sculptures. Following a certain pattern unit, motifs ranging from 4 cm to 20 cm high were engraved in wood or metal, allowing a production of a great number of gilded tin leaves. he richly woven texture of fabrics embroidered in gold thread was imitated by placing the leaves side by side. For further information, see Serck-Dewaide (1991). ArcheoSciences, revue d’archéométrie, 33, 2009, p. 423-432 426 he entire process clearly indicates that Fernão da Costa, who carried out the work in only eight days, did not have enough time to execute the gilding more carefully by applying water gilding and covering larger areas. In 1685, as stipulated in the contract of 1684, all the carved elements of the structure and the statuary were treated with burnished gilding (with the exception of lesh tone, hair, and beard), which was laid on an orange bole (Fig. 2). Water gilding was all the more imperative for the clothing once the technique of sgraito (‘estofado’ in Portuguese) was also required by the commissioner. his other technique, closely connected with gilding (described in detail by Nunes [1615] and Pacheco [1649]), reproduced the brocade fabrics still in fashion. It involved painting a colour in egg tempera over burnished gilding and stripping of the paint selectively in diferent design patterns to expose the shiny gold beneath (Serck-Dewaide, 1991). A small raised hand moulded texture, meant to imitate to perfection the weft looped weave of Italian or Spanish brocades, was applied to complete the surfaces of the clothing (Le Gac, 2004b). Such inal decorations are made in wax, hence water repellent; therefore, they are the only ones covered with oil gilding (not mentioned in the contract). Between the wax and the gold leaf, we found a very ine ‘or de couleur’, whose complex elemental composition led us to believe that palette scrapings could have been boiled with oil. In 1900, during the restoration, 1) three movable groups of the predella were completely or partially replaced with new elements and then water gilded. he gold leaf was laid on a pinkish bole and carefully polished. 2) Matte gilding was used on the damaged elements of the architectural structure (moulding, stars glued on the lat areas, or small ixed igures). his matte gilding is applied on a thick oily layer of a light yellow colour 7, rich in lead, barite, cadmium and chrome compounds, which reveal the wide range of pigments available. In places, there are indications of a hurried application and lack of care: pigmented oil splashes over the blue background, and wrinkles on the gold leaf (it appears to have been laid on too soon, before the ground had dried out suiciently). In contrast, the water gilding has been applied with a high degree of craftsmanship. Whether this indicates two separate gilders, or is a factor of viewing distance (the oil gilding is much further away), it is not clear. 7. We verify thus that gilders continued to use an ‘or de couleur’, whereas, at the end of the 18th century, they were inclined to use ‘mixtion’, an oil-based substance leaving an extremely thin layer when dry. In Portugal, at the very beginning of the 20th century, Castro da Silva (1900) reiterates both uses of ‘or-couleur’ and ‘mordente’. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 423-432 Agnès LE GAC et al. Figure 2: (See colour plate) Cross-section of the gilding (R-O1). Optical microscope, magniication 100x. Figure 2 : (Voir planche couleur) Coupe transversale d’échantillon (R-O1) comportant trois dorures successives (MO 100x). Gold-beating process and raw material Up until the 18th century, gold leaf came from mint gold coins rather than from raw gold, described as ‘or vierge’. Sources, either concerning Portugal or other countries in Europe, refer to the use of “ducados” (Cennini, ca. 1400), “scudi” (Vasari, 1550), “cruzados” (Livro dos Regimentos 1572), “piastres” or “anciennes monnaies d’Espagne” (Diderot and d’Alembert, 1752), or very simply to “moedas de ouro” (Brandão, 1985). While some coins were undoubtedly pure gold, others would have been alloys. his suggests that goldbeaters must have known how to reine debased coin species by cupellation and quarting (Guerra and Gondonneau, 1998). hickness of the gold leaf and its performance Early sources indicate that the thickness of the metal leaves was of great importance because: 1) it afects the spreading capacity and performance of the leaf; 2) a very thin leaf creates an illusion at a lower cost, since it can be applied to materials of inferior quality, giving the impression that they are made out of solid gold or silver; 3) while thicker leaves are well adapted to lat surfaces, thinner leaves are more suitable for carved elements. Gold and silver, in thinly beaten leaves, will cover larger areas with a much reduced weight of metal. Table 1 summarizes the documents where we found one or more references to the type of gold used (mint gold coins or raw he main altarpiece of the Old Cathedral of Coimbra (Portugal)… 427 ArcheoSciences, revue d’archéométrie, 33, 2009, p. 423-432 Table 1 : Weight* of gold used for the beating process and output by square meters obtained with the beaten gold leaves, according to several documentary sources. Tableau 1 : Poids* d’or utilisé pour la batte, et « rendement supericiel » des feuilles battues à partir de quelques sources anciennes. 428 gold), the weight of gold required for the beating process, the number of leaves obtained and/or their dimensions. he number of leaves obtained from an approximate weight of 28 g increases signiicantly over time. he number of leaves rises from 500-700 in the 1st century (Plinius) to 1190 in the 14th century (Cennini), to 1366 in 1550 (Vasari), to 1600-2000 in 1690 (Furetière), until 2407-2880 in 1750 (Savary des Bruslons). he increase of the area covered is also signiicant, successively from 726-931 m2 in the 16th century (Vasari, 1550), to 1114 m2 during the 17th century (Furetière, 1690), and 1587-1899 m2 in the 18th century (Savary des Bruslons, 1750). he observation of all the gold leaves of the main altarpiece of Coimbra with SEM (magniication up to 5000x) reveals thicknesses clearly below 1 µm, varying between 0.24 and 0.45 µm. hese dimensions are not absolute, since there are other parameters to consider: 1) gold, even burnished, presents an irregular surface on a micrometric scale; 2) if laking of, the gold leaf seems to be thicker than in reality; 3) overlaps can falsify the indings as well; 4) if sample cross-sections are not precisely positioned, any slight deviation can be suicient to modify the apparent thickness of all the layers. Although Cennini (ca. 1400) recommended “the use of thicker gold on panels and thinner gold on mouldings and leaves”, we have not studied this aspect. In the altarpiece, the lat surfaces are reserved for the application of the colour blue. he other parts of the architectural structure and the statuary are all carved in the round, resulting in more or less complex volumes and an unclear demarcation between lat and rounded areas. Caratage (ct) and gold leaf quality On the entire set of results obtained by SEM-EDX, Au is present in the highest quantity. Its purity is such that, between 1502 and 1685, gold appears to have been the sole metal acceptable. he elemental composition of the gold shows very minor amounts of Cu and sometimes, but rarely, Ag. Most of the time, the readings for Cu and Ag fall below the margin of error. Nevertheless, every time a peak for Cu or Ag appeared in the spectra, even if very weak, those elements were taken into account, in case they were present as traces. he original gilding (1502) has a high Au content (wt%), more than 99%; the one from 1583 shows a little Cu (one analysis, among the four we carried out, indicated that it contains 1.6% Cu); and Cu was found as well in 1685, from circa 0.5% to 4.5%, according to the speciic decorative surfaces (water or oil gilded). hese results indicate that either high purity native gold alloys were used for making ArcheoSciences, revue d’archéométrie, 33, 2009, p. 423-432 Agnès LE GAC et al. the gold leaf, when Cu reaches a concentration of ca. 1% (Guerra and Calligaro, 2003), or that Cu was intentionally introduced in the fabrication of the very thin leaf to modify its hardness 8, and, perhaps, colour. he last, 1900 gold leaf is of lower quality. It clearly corresponds to an Au-Ag-Cu alloy – the so-called ‘ternary alloy’ obtained when the three metals are intentionally mixed together – in which Au is nearing 83%, and which is approximately equivalent to a 20 ct gold alloy. In interpreting the results from the perspective of the History of Art in general and of the history of the altarpiece in particular, it is useful to consider the following aspects: At the end of the 15th century, and according to guild regulations, for Brussels in 1454 and Antwerp in 1470 (Pagès-Camagna, 2002) for example, only ine gold could be used by the painter-gilders on altarpieces. Jean d’Ypres would not only have been fully aware of such restrictions, but would have also imported this practice with him (as he did for many other techniques applied by the workshops of Brabant that we found on the altarpiece of Coimbra). In Portugal, however, the control over materials was the privilege of the commissioner. Taking into account the aweinspiring spectacle Bishop D. Jorge de Almeida sought to produce with his imposing project (Vasconcelos, 1930), he had no reason not to impose the greatest reinement, and therefore the use of a very pure gold 9. his aspect has been veriied by the present study. In 1583, the amounts of money assigned to the purchase of materials and the manpower necessary for refurbishing the altarpiece were limited. Only 1100 real were allocated for the gold leaf. At that time, a “milheiro” (a “thousand”, i.e. ten booklets of a hundred gold leaves each) cost between 3000 and 5000 real, depending on where it was obtained (Lisbon or Oporto), but perhaps also on how the gold was produced, its thickness and ineness. As the analyses show, very pure gold has been utilized, indicating that Fernão da 8. Fully aware that “Gold in its perfect pure form is soft” and that “the more metals are hard, the more they can have a beautiful polish” (Diderot and d’Alembert, 1765), we wonder if the presence of copper in gold leaves was meant to make the alloys slightly harder and turn the gilded surfaces more brilliant when burnished. In any case, it appeared that the intentional addition of copper did not hinder beating the gold to obtain leaves of an extreme thinness. 9. Diferent types of gold leaf were fabricated at that time and are specially mentioned to be used for artistic purposes in contracts written in Portugal in the 15th and 16th centuries (Garcia, 1923). he ‘ouro meão’ or ‘ouro meado’ – the so-called ‘Zwischengold’ in German, ‘oro di metà’ in Italian and ‘party gold’ in English – made up part of the set. he ‘ouro meão’ corresponds to a gold leaf and a silver leaf beaten together, equal in strength to a single metal leaf, whose gold side is used as the outer surface. Considering its deceptive appearance and low cost, associated with its poor aging, its use for artworks was proscribed in Lisbon in 1539, and again in 1572. he main altarpiece of the Old Cathedral of Coimbra (Portugal)… Costa did not compromise by using a low-grade and cheaper alloy; he rather respected the transcendental meaning of the most precious metal above all others in the religious context in which it had to be used. At the end of the 17th century, the gilding of the altarpiece was a matter of priority for Bishop D. João de Mello, and while the changes represented the cutting edge of the new baroque polychromy, the dichromatic (gold and blue) character of the gothic coating was preserved. he contract signed in 1684 insisted on the gold being “subido”, i.e. of a very high grade, “bem corado”, colourful, and “brunido”, burnished on all the parts to be gilded (carved elements of the structure and statuary). hese adjectives, used to characterize “the gold” 10 occur in a great number of Portuguese contracts of the 17th and 18th centuries (Brandão, 1984-1985; Alves, 1989; Le Gac in Serck-Dewaide et al., 2004). he contract is very explicit: Manuel da Costa Pereira, the painter-gilder who was in charge, had to use ine gold, of a saturated yellow, and obtain a brilliant inish using the water-gilding technique. he mention of the colour is signiicant because, up to the 18th century, the quality of gold or silver was very often evaluated based on their visual aspect 11. Any reference to colour would also be intended to diferentiate ‘ine gold’ (24 ct gold) from ‘common gold’ (23.25 ct gold, redder, containing up to 12 grains of silver [2.08%] and 6 grains of copper [1.04%]), and from ‘pale gold or green gold’ (16 ct gold, a lot less yellow, containing up to a third of silver). hese three types of beaten gold leaf were oicially recognised at the beginning of the 18th century (Savary des Bruslons, 1723; Diderot and d’Alembert, 1752). It should be pointed out again that the gold caratage is not speciied in the altarpiece contract. he mention of the gold caratage is practically – not to say totally – omitted in the contracts of this period 12; hence the importance of inding a deed signed by the same artist in 1686 for the gilding of another altar10. In French, as well as in Portuguese, the word ‘gold’ (‘or’ / ‘ouro’) in use at this time means ‘gilding’ (‘dorure’ / ‘douramento’), the latter being rather reserved to textile art (gold or silver wire, in lamella and ilé). 11. Until the 18th century, the sworn assayers from he Royal Mint used the touchstone (a small tablet of dark stone) to assay precious metal alloys. On one hand, they left on the touchstone a visible trace of each reference gold alloy, fabricated as a little stick and whose caratage was well known (known as ‘toucheaux’ in French); and on the other hand, they did the same with the metal alloy they had to control. “According to the degree of conformity they found between the color of this latter metal and the color of a determined reference alloy, they were able to decide on its purity” (Diderot and d’Alembert, 1765). We thus understand better why commissioners accorded such extreme importance to the colour of the gold alloy. 12. Among hundreds of contracts they red, Natália Alves found only three (dated 1712 and 1737) that refer to the gold alloy caratage, more speciically to a “23.5 carat gold” (Alves, 1989) and Fausto Sanchez Martins (1994) found one contract (dated 1710), that stipulates the use of a “twenty-three and two thirds carat gold”. 429 piece, in which it was stipulated “to use a pure twenty fourcarat gold similar to the one employed on the main altarpiece of the Cathedral [of Coimbra]” (Garcia, 1923). his demand is all the more exceptional because it refers to a perfectly pure gold. According to Furetière (1690) and Savary des Bruslons (1723), this was extremely diicult to obtain: “gold can never be reined up to this [level] – 24 carats – there are always some small fractions missing”. Diderot and d’Alembert (1768) add: “it is enough that the test tube reports 23 carats to the gold leaf be reputed ine, as the missing weight cannot be prejudicial to the highest-grade gold”. We therefore conclude, considering the results we obtained by SEM-EDX, that Manuel da Costa Pereira respected the terms of the contract. Taking into account the enormous amount of gold present, the gilding shows great consistency, both in quality and execution, despite the participation of many individuals organised in a workshop, as was usual at this period for large-scale operations. Such homogeneity is dependent irst on the goldbeater, in Portugal, who had to produce all the “milheiros” necessary for the gilding of a single artwork (Brandão, 1985; Alves, 1989). It seems, however, that the diferences observed in the gold caratage between water gilding and oil gilding were likely due to the painter-gilder himself or one of his craftsmen. he results for oil gilding clearly show that it was a deliberate choice to use a low-grade gold on statuettes, where gilding was required on the small wax ornaments (which had been irst applied to the garments of the small igures). We were not able to establish if the use of ca. 21.5 ct gold leaf was a matter of economy – a low-grade gold cannot be noticed on very small objects – or if it was rather an aesthetic concern, that of differentiating burnished gold from matte gold. he caratage of the gilding used in the 1900 restoration does not comply with the purest commercial grade accessible at the time (a 23 ct gold), that was especially manufactured, among other gold alloys for indoor exposure, in Lisbon, by A Favrel Lisbonense (Varella, 1901) (Table 2). Both caratages identiied by SEM-EDX (20 ct and c. 19.5 ct) correspond to Varella’s yellow greenish gold alloys: either “cidrão claro” (“light citron” – 20 ct gold) or “verde 1ª” (“extra ine green” – 19 ct gold) 13. Once again, was the quality of the gold leaf chosen for an economic reason or an aesthetic one? he gold alloy used in the gilded surfaces would have resulted in an obvious change of hue, from an original warmer tone to a pale one, reinforced by the very light yellow ground over which the gold leaf was laid. While the restoration appears 13. he list of colours available in England in 1929, published by Andrew Lins (1991), shows that the designation of gold leaves according to their alloy quality varies from one country to another, and from date to date. he one that we are considering is not standardised. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 423-432 430 Agnès LE GAC et al. well blended nowadays, we wondered whether, at the time of application, there had been an attempt to visually ‘integrate’ the new gilding. In the samples of the most recent oil gilding, we did not ind any tinted layer above it which would have provided a certain patina. Further investigation: Origin of the gold supply hese preliminary results, compared with the precisely dated documents to which we had access, and with published studies on the circulation of gold in the past (Godinho, 1971; Guerra and Calligaro, 2003), led us to formulate some theories about the provenance of this metal: for the irst metal leaf (1502), it is possible that the gold came from Northern Africa, for the second (1583) from Western Africa, for the third (1685) via the Spanish from the mines of Colombia, and for the fourth (1900) from Brazilian ores. It proved diicult to conirm the provenance of the gold in this study for various reasons: the very thin layers of gold (0.24-0.45 µm) available for study, and their location, surrounded by materials whose elemental composition could compromise the characterisation of trace elements in the gold. Indicators, such as the presence of Pt in the gold leaf from 1685, or Pd in the leaf from 1900 (Guerra and Calligaro, 2003), could provide a certain provenance. However, attempts to identify them using particle induced X-ray emission (PIXE) or synchrotron radiation X-ray luorescence spectroscopy (SR-XRF) did not yet provide satisfactory results. his study is currently being pursued. On the main altarpiece of the cathedral of Coimbra, different gilders applied the procedures of their time while perpetuating an age-old craft tradition that changed little since the one described by Cennini (ca. 1400). his study has revealed new information, but in order to understand how widespread these practices were, we must rely on further research. We look forward to further information on three Oporto gilded altarpieces provided in the study of Bidarra et al. (2009) in order to broaden our understanding of the gilding techniques used in Portugal, particularly during the baroque period. Acknowledgments Table 2: Gold alloys for indoor exposure, available at A Favrel Lisbonense in 1901. Tableau 2 : Alliage d’or en feuilles pour dorure d’intérieur, disponibles chez A Favrel Lisbonense en 1901. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 423-432 he authors are extremely grateful to Isabel Nogueira, of the Instituto Superior Técnico / Instituto de Ciência e Engenharia de Materiais e Superfícies (Lisbon), for providing the analytical data by SEM-EDX, to Maria José Vinagre, of the Departamento de Estudo dos Materiais / Instituto dos Museus e da Conservação (Lisbon), for providing the analytical data by XRD, and to Maria Filomena Guerra, of the Centre he main altarpiece of the Old Cathedral of Coimbra (Portugal)… de Recherche et de Restauration des Musées de France, for accepting the challenge to try to measure the concentration of characteristic trace elements in the gold leaf (by PIXE, with the AGLAE accelerator). hey especially thank Marie-Louise Hugon and Josephine de Linde for translating this paper from French into English, and Leslie Carlyle for technical content editing. References ALVES, N.M.F., 1989. A arte da talha no Porto na época barroca. (Artistas e clientela. Materiais e técnica), Volumes 1 and 2. Porto, Câmara Municipal do Porto. BIDARRA, A., COROADO, J. and ROCHA, F., 2009. Gold leaf analysis of three baroque altarpieces from Porto. ArcheoSciences 33. BIGELOW, D. et al. (eds.), 1991. Gilded Wood: Conservation and History. Madison, Connecticut, Sound View Press. BRANDÃO, D. DE P., 1984-1985. Obra de talha dourada, ensamblagem e pintura na cidade e na diocese do Porto. Documentação I, séculos XV a XVIII-Vol. I / Documentação II, 1700-1725 – Vol. II / Documentação III, 1726-1750 – Vol. III. Porto, [s.n.]. CASTRO DA SILVA, F.L.T. DE, 1900. Algumas indicações sobre a arte de dourar. Lisboa, Typograia do Commercio. CENNINI, C., [CA. 1400] 1991. Le Livre de l’art (Il libro dell’arte). Traduction critique, commentaires et notes par Colette Déroche. Paris, Editions Berger-Levrault. CORREIA, V. (ed.), 1926. Livro dos Regimentos dos Oiciaes mecanicos da mui nobre e sepre leal cidade de Lixboa [1572]. Coimbra: Imprensa da Universidade. DIDEROT, D. and D’ALEMBERT, J.L.R., 1752 / 1765 / 1768. Encyclopédie, ou dictionnaire raisonné des Sciences, des Arts et des Métiers, par une Société de gens de Lettres. Paris, Chez Briasson, David, Le Breton & Durand, Vol. II (1752) / Neufchastel, Chez Samuel Faulche & Compagnie, Vol. VIII (1765) and XI (1768). FÉLIBIEN, A., 1676. Des principes de l’architecture, de la sculpture, de la peinture et des autres arts qui en dépendent. Avec un dictionnaire des termes propres à chacun de ces arts. Paris, Chez Jean-Baptiste Coignard. FURETIÈRE, A., 1690. Dictionnaire Universel, contenant generalement tous les Mots François, tant vieux que modernes & les Termes de toutes les Sciences et des Arts, Volumes 1-3. La Haye, Chez Arnout & Régnier Leers. GARCIA, P.Q., 1923. Artistas de Coimbra. Documentos para as suas biograias. Coimbra, [Imprensa da Universidade]. GODINHO, V.M., 1971. Os descobrimentos e a economia mundial, Vol. 1-3. Lisboa, Editorial Presença (2nd edition). GUERRA, M.F. and CALLIGARO, T., 2003. Gold cultural heritage objects: a review of studies of provenance and manufacturing technologies. Measurement in Science and Technology 14: 1527-1537. 431 GUERRA, M.F. and GONDONNEAU, A., 1998. As tecnologias de fabrico das ligas amoedáveis e a proveniência dos metais amoedados: perspectivas abertas pelas novas técnicas de análise elementar, in F.A. Magro (ed.), Actas IV Congresso Nacional de Numismática. Lisboa, Associação Numismática de Portugal, 369-402. LE GAC, A., 2004(A). Le retable majeur de la Sé Velha de Coimbra (Portugal) et sa polychromie de 1685, in Retables in situ. Conservation & restauration. Preprints des 11es journées d’étude de la Section Française de l’Institut International de Conservation, Roubaix, 24-26 juin 2004. Champs-sur-Marne, SFIIC, 111-124. LE GAC, A., 2004(B). Les techniques de polychromie des “Broderie d’application” et des “bouclés par la trame” – Etude préliminaire, in A.I. Seruya (ed.), POLICROMIA. A escultura policromada religiosa dos séculos XVII e XVIII. Estudo comparativo das técnicas, alterações e conservação em Portugal, Espanha e Bélgica. Actas do Congresso Internacional, 29-31 Outubro 2002. Lisboa, Instituto Português de Conservação e Restauro, 79-94. LINS, A., 1991. Basic properties of gold leaf, in Gilded Wood. Conservation and History. Sound View Press, Madison, Connecticut, 17-23. LOURO, J.M.S., 1983. O restauro do retábulo da Capela-mor da Sé Velha no séc. XVII. Mundo da Arte 13: 60-63. MARTINS, F.S., 1994. A arquitectura dos primeiros colégios jesuítas de Portugal: 1542-1759. PhD hesis, Faculdade de Letras, Porto, 2 volumes. MARQUES, M.T.R., 1998. Análise do revestimento metálico de talha dourada por MET/EDS. Relatório de Stágio / Licenciatura (M.A. in Science), Universidade de Lisboa, Portugal (Unpublished). MOURA, C., BARREIRO, A. and RIBEIRO, I., 2004. Os bustosrelicário. Alguns casos notáveis da escultura seiscentista de barro em Alcobaça, in POLICROMIA. A escultura policromada religiosa dos séculos XVII e XVIII, 63-70. NUNES, P., [1615] 1982. Arte da Pintura. Symmetria, e Perspectiva. Fac-similé de l’édition originale, Introduction de Leontina Ventura. Porto, Editorial Paisagem. PACHECO, F., [1649] 1990. Arte de la Pintura. Madrid, Editorial Cátedra. PAGÈS-CAMAGNA, S., 2002. La polychromie des sculptures brabançonnes du Musée du Louvre sous l’œil du microscope, in Retables brabançons des XVe et XVIe siècles. Actes du Colloque du Musée du Louvre, 18-19 mai 2001. Paris, La documentation Française, 103-126. SAVARY DES BRUSLONS, J., 1723/1750. Dictionnaire Universel de Commerce. Volumes 1-3. Paris, Chez Jacques Estienne. SERCK-DEWAIDE, M., 1991. he history and conservation of the surface coating on European gilded-wood objects, in D. Bigelow et al. (eds.), Gilded Wood: Conservation and History. Madison, Connecticut, Sound View Press, 65-78. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 423-432 432 SERCK-DEWAIDE, M. [et al. / BELGIUM], CAMPOY, M. [et al. / SPAIN], LE GAC, A. [et al. / PORTUGAL], 2004. Les techniques utilisées dans l’art baroque religieux des xviie et xviiie siècles au Portugal, en Espagne et en Belgique, in A.I. Seruya (ed.), POLICROMIA. A escultura policromada religiosa dos séculos XVII e XVIII. Estudo comparativo das técnicas, alterações e conservação em Portugal, Espanha e Bélgica. Actas do Congresso Internacional, 29-31 Outubro 2002. Lisboa, Instituto Português de Conservação e Restauro, 119-155. ArcheoSciences, revue d’archéométrie, 33, 2009, p. 423-432 Agnès LE GAC et al. VARELLA, J.N., 1901. Preço corrente de A Favrel Lisbonense. Fabrico de ouro e prata em folhas e em pó. Lisboa, Typograia do Commercio. VASARI, G., [1550] 1996. Le tecniche artistiche. Introduzione e commento di Gerald Baldwin Brown. Vicenza, Neri Pozza Editore. VASCONCELOS, A. DE, 1930. Sé-velha de Coimbra (Apontamentos para a sua história). Coimbra, Imprensa da Universidade.

RELATED PAPERS

RELATED TOPICS

Log In

Aurum: Special Issue of ArcheoSciences 33 (Guerra & Rehren, Eds, 2009) - full volume (16 MB)

Aurum: Special Issue of ArcheoSciences 33 (Guerra & Rehren, Eds, 2009) - full volume (16 MB)

Related Papers

RELATED PAPERS

RELATED TOPICS