In Press at Mycologia, preliminary version published on July 1, 2014 as doi:10.3852/13-317 Running head: Entyloma on Eryngium Revision of Entyloma (Entylomatales, Exobasidiomycetes) on Eryngium Kyrylo G. Savchenko1 Department of Evolutionary & Environmental Biology, University of Haifa, Mt Carmel, Haifa 31905, Israel, and G. Kholodny Institute of Botany of the NAS of Ukraine, 2 Tereshchenkivska Street, Kyiv 01601, Ukraine Lori M. Carris Department of Plant Pathology, Washington State University, Pullman, Washington 99164 Lisa A. Castlebury USDA-ARS, Systematic Mycology & Microbiology Laboratory, 10300 Baltimore Avenue, Beltsville, Maryland 20705 Vasyl P. Heluta M.G. Kholodny Institute of Botany of the NAS of Ukraine, 2 Tereshchenkivska Street, Kyiv 01601, Ukraine Solomon P. Wasser Department of Evolutionary & Environmental Biology, University of Haifa, Mt Carmel, Haifa 31905, Israel, and M.G. Kholodny Institute of Botany of the NAS of Ukraine, 2 Tereshchenkivska Street, Kyiv 01601, Ukraine Eviatar Nevo Department of Evolutionary & Environmental Biology, University of Haifa, Mt Carmel, Haifa 31905, Israel Abstract: The genus Entyloma consists of more than 160 species of smut fungi distributed worldwide on dicots, with Apiaceae being one of the main host families. This study aims to clarify the systematics and phylogeny of Entyloma on Eryngium (Apiaceae) with molecular and morphological data. Eleven species from Eryngium are discussed herein. Copyright 2014 by The Mycological Society of America. Four of them are described as new taxa: E. carmeli sp. nov. on Eryngium falcatum, E. eryngii-cretici sp. nov. on Eryngium creticum, E. eryngii-maritimi sp. nov. on Eryngium maritimum and E. ho-chunkii sp. nov. on Eryngium yuccifolium. Analysis of the internal transcribed spacer (ITS) region of rDNA is presented and supports the polyphyly of Entyloma on Eryngium. Key words: Entylomataceae, host diversity, phytopathogen, smut fungi, taxonomy INTRODUCTION Entyloma de Bary comprises more than 160 species of smut fungi that are found on all continents except Antarctica (Vánky 2011). It is characterized by the formation of sori embedded in the tissue of the vegetative parts of dicotyledonous plants, which look like dense white spots or rarely pustules and galls. The characteristic features of this genus include mostly hyaline, solitary or crowded teliospores, which germinate with Tilletia-like basidia, simple host-parasite interaction apparatus and simple septal pores with two membrane cups (de Bary 1874, Bauer et al. 1997, Vánky 2011). Due to minimal morphology and limited taxonomically valuable characters, the systematics of this group is dependent on host taxonomy. More than 70 Entyloma species are found on Asteraceae and more than 10 on each of Apiaceae, Papaveraceae, Ranunculaceae and Scrophulariaceae (Begerow et al. 2002, Vanky 2011). Almost half of the species parasitizing umbellifers (Apiaceae) have been described on Eryngium L. hosts. The first species from this group was E. eryngii on Eryngium campestre L., described as Physoderma eryngii Corda (Corda 1839) and transferred to the genus Entyloma when it was created by de Bary (1874). Additional species were described from North Africa; they are E. eryngii-dichotomi Maire on Eryngium dichotomum Desf. (Maire 1917), E. eryngii-tricuspidati Maire on Eryngium tricuspidatum L. (Maire 1919) and E. maroccanum Maire on Eryngium maroccanum Pitard (Maire 1937), from Europe; they are E. eryngii-alpini Vánky on Eryngium alpinum L. (Vánky 2009) and E. eryngii-plani Cif. on Eryngium planum L. (Ciferri 1924). Entyloma argentinense Speg. on Eryngium nudicaule Lam. was described from South America (Spegazzini 1925). Eryngium is a large genus in subfamily Saniculoideae with approximately 250 species distributed worldwide (Wollf 1913, Wörz 2005). Taxonomically Eryngium is divided into two subgenera: Monocotyloidea for species from the Americas and Eryngium for species from Eurasia. Molecular phylogenetic studies support this infrageneric classification and reveal three strongly supported additional clades within Monocotyloidea, the so-called “South American species”, “Mexican” and “Eastern USA” (Calviño et al. 2008, 2010). Species of Entyloma have been reported on a number of members of subgenus Eryngium and on species from the “South American” clade of Monocotyloidea (Vánky 2009, 2011). Of note, no Entyloma species have been reported on “Mexican” and “Eastern USA” Eryngium species. Moreover, there are only several records of Entyloma on North American Eryngium aquaticum L. (Anonymous 1960), Eryngium yuccifolium Michx. (Clinton 1902, Gilman and Archer 1929, Green 1942, Fischer 1953) and on South American Eryngium nudicaule (Spegazzini 1925) from the “South American” clade, which most likely is due to incomplete sampling. In the most recent treatment of Entyloma on Eryngium, Vánky (2009) assumed that the radiation and speciation in this group was high and that each Eryngium species possesses its own Entyloma parasite. Taking into account this hypothesis, several records from different parts of the world do not correspond to any known Entyloma species. Most were reported as E. eryngii. Specimens have been reported on Eryngium aquaticum (Anonymous 1960) and Eryngium yuccifolium (Gilman and Archer 1929, Greene 1942, Fischer 1953, Zundel 1953) from North America, on Eryngium amethystinum L. from Slovenia (Voss 1889–1892), on Eryngium macrocalyx Schrenk from Kazakhstan and Uzbekistan (Ramazanova et al. 1987), and on Eryngium maritimum L. from France and Scotland (Ainsworth and Sampson 1950, Viennot-Bourgin 1956). Reports also exist of E. eryngii-tricuspidati on Eryngium creticumWebb. from Greece and Israel (Rayss and Zwirn 1944, Pantidou 1973) and, as found recently by the lead author, on Eryngium falcatum Delar. from Israel. Therefore, a critical revision of this group is necessary. To date, only two species of Entyloma on Eryngium (i.e. E. eryngii on Eryngium campestre and E. eryngii-plani on Eryngium planum) have been used in molecular phylogenetic studies (Begerow et al. 2002). The aim of this study is to critically analyze known Entyloma species on Eryngium hosts based on morphological and molecular phylogenetic data using recent collections in the analysis where possible. This work includes an ITS phylogeny of some Entyloma on Eryngium with focus on the Israeli specimens on Eryngium creticum and Eryngium falcatum and describes four species that are proposed as new. An identification key is provided for Entyloma species on Eryngium. MATERIALS AND METHODS Specimen sampling and documentation.—Specimens in this study are summarized (TABLE I). Herbarium samples are deposited in BPI, DAOM, FT, G, HAI, K, KRAM, MICH, NY, WIS and WSP. Nomenclatural novelties are registered in MycoBank (Crous et al. 2004). The hologenetype concept follows the proposal of Chakrabarty (2010). Morphological examination.—Sorus and spore characteristics were studied with dried herbarium material. Specimens were examined by light microscopy (LM). The characters assessed included: (i) sorus characters—shape, size, color, position on the leaves (abaxial, adaxial, amphigenous); (ii) spore morphology—the average length of spores, the average width, spore wall thickness, thickness of the outer layer, thickness of the inner layer, character of spore wall—even or uneven, color of spore matrix, color of outer layer of the spore wall; (iii) anamorph (if present)—position on the leaves (abaxial, adaxial, amphigenous), color. Micrographs of sori were taken with a Discovery 20 stereo microscope and AxioCam digital camera (Carl Zeiss Microscopy, Thornwood, New York) imaging system or Carl Zeiss Stemi D 4 stereo microscope. For LM, small pieces of leaf tissue with sori were mounted in 90% lactic acid on a microscope slide, several drops of distilled water were added, the slide was heated to the boiling point and cooled. The softened sori were cut in long pieces and squashed with a lancet, covered, gently heated to boiling point to eliminate air bubbles and examined under a Carl Zeiss Axiostar and a Carl Zeiss Axioplan2 compound light microscope imaging system at 1000× magnification. LM photographs were taken with a Canon Power Shot G10 and an AxioCam digital camera. At least 50 spores were measured from each collection, and the variation is presented as a range, with extreme values given in parentheses. In the descriptions, mean and standard deviations (SD) were calculated from spores measured in all specimens and provided after the spore sizes. DNA extraction and sequencing.—The majority of sequences were obtained from GenBank with seven additional sequences generated for this analysis (TABLE I). Genomic DNA was isolated from sori removed from herbarium specimens and lysed in 1.5 mL tubes for 1 min with a FastPrep 24 (MP Biomedicals, Irvine, California.). Tubes were incubated in a water bath 5 h at 55 C, and DNA was extracted with DNeasy Plant Mini Kit (QIAGEN Inc., Valencia, California) following the manufacturer's instructions. All amplifications were performed in 20 mL aliquots on a GeneAmp 9700 thermal-cycler (Applied Biosystems, Foster City, California). ITS5 or ITS1 were used as the forward primer and ITS4 as the reverse primer (White et al. 1990). Standard cycling parameters with an annealing temperature of 57 C were used for amplification. PCR products were purified with ExoSAP-IT (USB, Cleveland, Ohio) according to the manufacturer's instructions and amplified with respective forward and reverse PCR primers with the BigDye 3.1 terminator kit (Applied Biosystems). Those products were sequenced on an ABI 3100 automated DNA sequencer. Phylogenetic analysis.—The chromatographic outputs were assembled and edited into consensus sequences with Sequencher 4.5 for Windows. The consensus sequences were aligned in MAFFT 6 (Katoh et al. 2002, Katoh and Toh 2008) under default settings. Ambiguously aligned regions were recorded with GBlocks 091.b (Castresana, 2000) and excluded from the analyses. MAXTREES were unlimited, branches of zero length and ambiguously aligned regions were excluded from the analyses. PAUP 40b10 (Swofford 2002) was used to conduct the parsimony analysis. Trees were inferred with the heuristic search options that were collapsed and all multiple parsimonious trees were saved. Descriptive tree statistics for parsimony (tree length [TL], consistency index [CI], retention index [RI], relative consistency index [RC] and homoplasy index [HI]) were calculated for trees generated under different optimality criteria. Bootstrap analysis (Hill and Bull 1993) was based on 1000 replications. Bayesian analysis using a Monte Carlo Markov chain (MCMC) technique was implemented in MrBayes 3.1.2 (Huelsenbek and Ronquist 2001, Ronquist and Huelsenbek 2003). Four MCMC chains were run simultaneously, starting from random trees, for 1 000 000 generations. Trees were sampled every 100th generation for a total of 10 000 trees. The first 2000 trees were discarded as burn-in of each analysis. Posterior probabilities (Rannala and Yang 1996) were determined from a majority-rule consensus tree generated with the remaining 8000 trees. This analysis was done four times, starting from different random trees to ensure that trees from the same tree space were being sampled during each analysis. Clades supported with a minimum of 95% Bayesian posterior probabilities are considered well supported. Trees were rooted with Entyloma atlanticum Massenot as in Begerow et al. (2002). RESULTS Morphology.—Morphological characters for Entyloma spp. on Eryngium are summarized (TABLE II). The main complication in taxonomic work with Entyloma is that it has few meaningful morphological characters that can be used in species delimitation. Therefore we chose 14 characters that can be easily accessed with light microscopy for evaluation of taxonomic utility. Spore size overlapped among all species with the exception of E. eryngii with larger spores and E. eryngii-maritimi with smaller spores. Spore wall thickness was a more useful criterion for species delimitation than spore length and width and varied 0.8– 1.5 µm in E. maroccanum to 2.5–7 µm in E. eryngii. Sorus color, size, position and structure were different in all species (TABLE II). Anamorphs were found on lesions of E. eryngii-plani, E. eryngii-alpini, E. maroccanum, E. carmeli and E. eryngii-tricuspidati and were most abundant on the latter species. We were unable to obtain specimens of E. argentinense and E. maroccanum; therefore, all of the information on those two species was taken from sources in the literature. Phylogenetic analysis.—The ITS dataset consisted of 36 ingroup and one outgroup taxa. The alignment was deposited in TreeBASE 14656 and contained 655 characters. Of those characters 477 were invariable, 82 variable but parsimony uninformative and 96 parsimony informative. The different runs of Bayesian and the MP analyses yielded consistent topologies. A representative tree generated from the Bayesian analysis is presented to illustrate the results (FIG 1). One clade (BP below 70, PP = 87) containing all species on Apiaceae (E. eryngii, E. carmeli, E. eryngii-plani, E. eryngii-cretici, E. magocsyanum Bubák) grouped together with other species on Asterids. Of note, E. calceolariae Lagerh., a parasite on Calceolaria L. (Calceolariaceae, Lamiales), also was clustered in the clade with the species on Apiaceae. Within this clade were isolates of E. eryngii (PP = 98%), E. carmeli (PP = 100%) and E. eryngii-cretici (PP = 100%). Moreover, E. eryngii-plani and E. carmeli were clustered together with E. magocsyanum from Tordylium cordatum Poir., forming three separate lineages. TAXONOMY Entyloma argentinense (Speg.) Cif., Annls mycol. 26(1/2):57 (1928) Ustilago argentina (Speg.) Speg., Revista Argent. Bot. 1(2–3):151 (1925) Entyloma eryngii (Corda) de Bary var. argentinensis Spegazzini, Revista Argent. Bot. (1):147 (1925) Sori in leaves forming orbicular or ellipsoidal, amphigenous, slightly swollen, dark brown spots, 1–2 mm diam. Anamorph not observed. Spores crowded, subglobose, ellipsoidal to subpolyhedrally irregular, 9–13 × 10–17 µm, yellow to pale yellowish brown; wall twolayered, uneven, 1.5–5 µm thick, smooth. Host range: Eryngium nudicaule section Foetida. Distribution: South America (Argentina). Type: Argentina. La Plata: on Eryngium nudicaule, Sep, Oct 1912, 1916, 1917, 1920, LPS 3367. Notes: Material not seen. Description is taken from the original (Spegazzini 1925) and from Vánky (2011). Entyloma carmeli Savchenko, Carris, Castl., Heluta, Wasser & Nevo, sp. nov. FIG. 2a–c. MycoBank MB807872 Etymology: derived from Mt Carmel, the type locale. Sori in leaves forming roundish to ellipsoidal, amphigenous spots, flat, scattered over the leaf surface, dark brown, 0.2–4 mm diam. Anamorph occasionally present, amphigenous, not abundant, yellowish white. Spores crowded, subglobose, ellipsoidal, 12– 17 × 9–15 µm diam, [av. ± SD, 12.1 ± 1.8 × 13.5 ± 2.4 µm} with yellowish brown to pale golden matrix. Spore wall two-layered, uneven, 1.5–4 µm thick, with outer layer 1.5–3 µm thick, and inner layer 0.5–1 µm thick. Outer layer pale golden, smooth. Habitat: Pinus halepensis Mill.-Arbutus andrachne L. forests, shady areas. Host range: Eryngium falcatum section Plana. Distribution: Asia (Israel). Type: Israel. Haifa: Carmel National Park, 0.5 km E of Haifa University campus, on Eryngium falcatum, 01 Apr 2011, leg. K.G. Savchenko, HAI 4521. Additional specimens examined (paratypes): Israel. Haifa: Carmel National Park, on the road to Usfiya, 2 km N of Haifa University campus, on Eryngium falcatum, 20 Apr 2011, leg. K.G. Savchenko, HAI 4522. Notes: The most striking character of E. carmeli is the dark brown, flat sori. This species differs from other Eurasian species of Entyloma on Eryngium section Plana by thicker spore walls. Of interest, the holotypes of E. carmeli and E. eryngii-cretici were found less than 100 m from each other, occupying different ecological niches (i.e. humid, shady forest for the former species and dry, rocky slopes for the latter species). Entyloma carmeli differs from E. eryngii-cretici by slightly larger spores (av. 13.5 vs. 12.0 µm) and thicker spore walls (1.5–4 vs. 1.5–2.5 µm). In addition, an outer spore layer of E. carmeli is yellowish brown and of E. eryngii-cretici is pale golden yellow. Entyloma eryngii (Corda) de Bary, Bot. Ztg. 32:105 (1874) Fig. 2, j, k Physoderma eryngii Corda, Icon. Fung. (Prague) 3:3 (1839) Protomyces eryngii Fuckel, Jahrb. Nassauischen Vereins Naturk. 23–24:75 (1870) Sori in leaves forming roundish to ellipsoidal, polygonal, amphigenous, flat, scattered over the leaf surface, often gregarious, pale brown, brown, pustules cracked at maturity, 0.3–1.5 mm diam or larger when confluent. Anamorph not seen. Spores crowded, subglobose, ellipsoidal, often irregular, 9–15 × (10) 11–20 µm diam [av. ± SD, 12.9 ± 2.7 × 14.8 ± 3.2 µm, n = 550/11] with subhyaline to pale yellowish matrix. Spore wall two-layered, uneven, 2.5–7 µm thick, with outer layer 2–6 µm thick, and inner layer 0.5–1 µm thick. Outer layer pale yellowish, smooth. Host range: on Eryngium campestre section Campestria. Distribution: Europe (Austria, Czech Republic, Finland, France, Germany, Hungary, Italy, Poland, Romania, Russian Federation, Slovakia, Slovenia, Spain, Ukraine), Asia (Georgia). Type: Czech Republic. Prague: on Eryngium campestre, autumn 1838, leg. A.C.I. Corda. Type not seen. Additional specimens examined: Austria. Niederösterreich: Steinfeld, ca. 6 km N of Vienna "New city", W of the train line between the Theresienfeld stop and Felixdorf train station, MTB 8163/1, on Eryngium campestre, 11 Jul 1997, leg. Th. Barta, WSP 70820; Marchfeld, ca. 15 km SE from Gänserndorf, ca. 3.5 km SW from Marchegg, near the train depot Breitensee, MTB 7767/1, 09 Jul 1999, leg. Th. Barta, WSP 70822. Czech Republic. Moravia: Brünn, Aug 1879, Prof. Niessl, Thümen, Mycoteca Universalis 1713, FT 5322/2; Freyn, Aug 1930, leg. J. Hruby, WSP 18106. Germany. Hamburg: Cuxhaven, Deich bei der Kugelbake, 20 Aug 1925, leg. A. Ludwig, WSP 2091; Thuringia: inter Hachelbich et Berka fr. Sondershausen, 20 Jun 1906, leg. G. Oertel, DAOM 79831. Hungary: Pr. pag. Balatonfüred ad lacum Balaton, Comit. Veszprém, 08 Aug 1974, leg. K. Vánky, Ust. Ex. 130, FT 5322/1. Italy. Parma: Vigheffio presso, Sep 1883, leg. G. Passerini, Erb. Critt. Ital. Ser. II. 1353, FT 5322/3; Verona: Scorgnano prope Tregnano, June 1906, leg. C. Massalongo, Saccardo Mychoteca Italica 1646, FT 5322/3. Spain. Llanes: Oviedo, Jul 1919, leg. P. Unamuno, BPI 175263; Isla del Mino: Sta. Maria de Castrelo, Orense, 28 Jul 1931, leg. P. Unamuno, BPI 175280. Notes: Entyloma eryngii is the most widely distributed species of this group. It differs from all the other species by the formation of pustulate, cracked, sometimes bullate sori, generally larger spores and thick, uneven spore walls. Specimens from Eryngium hosts were reported under this name over the past century. Thus, it was reported on Eryngium ampelostrathum from Slovenia (Voss 1889–1892), Eryngium aquaticum from Iowa (Anonymous 1960), Eryngium macrocalyx from Kazakhstan and Uzbekistan (Ramazanova et al. 1987) and on Eryngium yuccifolium from Iowa and Wisconsin (Clinton 1902, Gilman and Archer 1929, Greene 1942, Fischer 1953, Zundel 1953). All specimens on the latter host belong to a different species, E. ho-chunkii, described in this paper. It is not possible to determine whether other specimens belong to E. eryngii without morphological examination. Unfortunately we were not able to locate those collections in any of the major world herbaria. Nevertheless, in that the respective hosts of these taxa belong to different sections of Eryngium (i.e. Panniculata for E. aquaticum, Halobia for Eryngium macrocalyx), it is likely that those collections represent undescribed species of Entyloma. It should be noted that Entyloma on North American Sanicula L. species have been reported as E. eryngii (Fischer 1953, Ellett 1959). Entyloma saniculae Peck, another species complex, also occurs on Sanicula species (Savchenko et al. in prep). Therefore, all collections of Entyloma eryngii on Sanicula need reexamination. Entyloma eryngii-alpini Vánky, Mycotaxon 110:304 (2009); Fig. 2d–f Sori in leaves forming roundish or polyangular, amphigenous, flat, gregarious spots, initially yellowish white and later yellowish brown, 0.2–2 mm diam or larger when confluent. Anamorph rarely present on the abaxial side of the leaves, not abundant, white. Spores crowded, globose, subglobose, ovoid, ellipsoidal, 8–12 × 9–14 µm diam, [av. ± SD, 10.8 ± 1.9 × 12.4 ± 2.2 µm, n = 300/6] with hyaline, subhyaline, pale yellowish matrix. Spore wall two-layered, even to uneven, 1.5–2.5 µm thick, with outer layer 1.5–2 µm thick, and inner layer 0.5 µm thick. Outer layer hyaline, pale yellowish smooth. Host range: Eryngium alpinum section Alpina. Distribution: Europe (Italy, Switzerland). Type: Switzerland. Graubünden: Frauenkirch near Davos, 1525 m, on Eryngium alpinum cult., 12 Aug 1987, leg. A. Bolay, G 00111412. Additional specimens examined: Switzerland. Graubünden: Basse Engadine, Ftah, on Eryngium alpinum, 20 Aug 1987, A. Bolay, G 00111416; Portein, Domlesch, 09 Jul 1987, leg. A. Bolay, G 00111417; Brigels, 07 Aug 1987, leg. A. Bolay, G 00111413; Savognin, 29 Jul 1987, leg. A. Bolay, G 00111414; Malix, 28 Jul 1987, leg. A. Bolay, G 00111415; Jenisberg, 20 Aug 1987, leg. A. Bolay, G 00111418. Notes: Entyloma eryngii-alpini differs from E. eryngii, the species to which it was first attributed by Servazzi (1950), in sorus and spore morphology. There are subtle differences between E. eryngii-alpini and another European species, E. eryngii-plani, in the color of young sori, which are yellowish white in the former species and greenish yellow in the latter. Another difference is the position of anamorph, which is amphigenous in E. eryngii-alpini and adaxial in E. eryngii-plani. Furthermore, Eryngium alpinum and Eryngium planum belong to different sections of Eryngium and were in different clades in the phylogentic analysis by Calviño et al. (2008). Entyloma eryngii-cretici Savchenko, Carris, Castl., Heluta, Wasser & Nevo, sp. nov. FIG. 2g–i MycoBank MB807873 Etymology: derived from the host plant Eryngium creticum. Sori in leaves forming roundish or ellipsoidal, amphigenous, flat to swollen spots, scattered over the leaf surface, yellowish brown, fawn, often gregarious, 0.2–4 mm diam. Anamorph absent. Spores crowded, subglobose, ellipsoidal, 9–13 × 10–(13) 15 µm diam, [av. ± SD, 10.6 ± 1.7 × 12.0 ± 2.5 µm, n = 150/3] with pale yellowish to pale golden matrix. Spore wall two-layered, even to uneven, 1.5–2.5(3) µm thick, with outer layer 1.5– 2(2.5) µm thick, and inner layer 0.5 µm thick. Outer layer pale yellowish, smooth. Habitat: open forests of Quercus calliprinos Webb, dwarf-shrubs, dry slopes. Host range: Eryngium creticum section Plana. Distribution: Europe (Greece), Asia (Israel). Type: Israel. Haifa: Carmel National Park, 0.5 km E of Haifa University campus, dwarf shrubs, dry slope, on Eryngium creticum, 08 Apr 2011, leg. K.G. Savchenko, HAI 4525. Additional specimens examined (paratypes): Israel. Golan Heights: Yehudiyya Reserve, 1 km N of the main entrance, open Quercus calliprinos forest, dry slope, on Eryngium creticum, 26 Apr 2011, leg. K.G. Savchenko, HAI 4526; Upper Galilee: Mt Meron, 3 km W of kibbutz Kfar Shamayi, 05 May 2001, leg. T. Pavlichek, HAI 4527. Notes: Entyloma eryngii-cretici differs from Entyloma eryngii in sorus and spore morphology. It differs from all of the other species with hosts belonging to section Plana—E. carmeli, E. eryngii-dichotomi, E. eryngii-plani, E. eryngii-tricuspidati—by the absence of anamorph, from E. eryngii-dichotomi by thicker spore walls, 1.5–2.5(3) µm vs 1–1.5 µm, and from E. carmeli by the color of sori. In addition, species from section Plana for which sequences were available—E. carmeli, E. eryngii-cretici, E. eryngii-plani—did not form a monophyletic group in the phylogenetic analysis (FIG 1). Thus far, Entyloma on Eryngium creticum have been reported from Israel (Rays and Zwirn 1944, Savchenko et al. 2010) as E. eryngii or Entyloma sp. and from Greece (Pantidou 1973) as E. eryngii. This smut is probably widespread in the eastern Mediterranean but is overlooked. Entyloma eryngii-dichotomi Maire, Bull. Soc. Hist. Nat. Afr. N. 8:146 (1917) Fig. 2e, m Sori in leaves forming polyangular, amphigenous, flat, scattered or gregarious, slightly thickened, pale brown spots, 0.5–2.5 mm diam or larger when confluent. Anamorph abundant, amphigenous, white to yellowish white. Spores crowded, subglobose, ellipsoidal, subpolyhedrally irregular, 8–11 × 10–14 µm diam [av. ± SD, 9.5 ± 1.6 × 12 ± 2.1 µm, n = 50/1] with subhyaline to pale yellowish matrix. Spore wall two-layered, uneven, 1–2 µm thick, with outer layer 1–1.5 µm thick, and inner layer 0.5 µm thick. Outer layer subhyaline to pale yellowish, smooth. Host range: Eryngium dichotomum section Plana. Distribution: Africa (Algeria). Type: Algeria. Zéralda near Alger, on Eryngium dichotomum , 01 May 1915, leg. R. Maire, Mycoth. Bor.-Afric. No. 277, NY 00738484. Notes: The distinctive feature of E. eryngii-dichotomi is the abundant Tilletiopsislike anamorph with 3–6 µm long ballistoconidia formed on sterigmata. Anamorph occurs on every sorus. This species was found only in North Africa (Algeria). Entyloma eryngii-maritimi Savchenko, Carris, Castl., Heluta, Wasser & Nevo, sp. nov. FIG. 3d, e MycoBank MB807874 Etymology: derived from the host plant Eryngium maritimum. Sori in leaves forming ellipsoidal or polygonal, adaxial, flat to swollen, greenish gray spots, scattered over the leaf surface, 0.1–3 mm diam. Anamorph absent. Spores crowded, subglobose, ellipsoidal, subpolyhedral to irregular, 6–9 × 7–10 µm diam, [av. ± SD, 7.6 ± 1.4 × 8.3 ± 1.8 µm, n = 100/2] with hyaline matrix. Spore wall two-layered, uneven, 1–2 µm thick, with outer layer 1–1.5 µm thick, and inner layer 0.5 µm thick. Outer layer hyaline, smooth. Host range: Eryngium maritimum section Halobia. Distribution: Europe (France, Scotland, Wales). Type: France. Seine-Et-Oise: Grignon, Botanic Garden, on Eryngium maritimum, 20 Sep 1935, leg. Anonymous, BPI 175289. Additional specimens examined (paratypes): United Kingdom. Wales: Carmarthenshire (Dyfed), on Eryngium maritimum, 03 Feb 1996, leg. N. Stringer, K(M) 106003. Notes: Entyloma eryngii-maritimi can be distinguished easily from other species by its small spores and greenish gray sori. It was found in western Europe only: France (Viennot-Bourgin 1956), Scotland (Ainsworth and Sampson 1950, Legon et al. 2005) and Wales (Legon et al. 2005). Entyloma eryngii-plani Cif., Boll. Soc. Bot. Ital. 2:54 (1924) Fig. 3a–c Sori in leaves forming circular or polyangular, amphigenous, flat, thickened, scattered or sometimes gregarious, initially yellowish green later yellowish, yellowish brown spots, 0.5–3 mm diam or larger when confluent. Anamorph rarely present on the adaxial side of leaves, not abundant, whitish. Spores crowded, often in chains, globose, subglobose, ovoid, ellipsoidal to slightly irregular, 9–13 × 11–16 µm diam [av. ± SD, 10.9 ± 1.7 × 12.4 ± 2.3 µm, n = 350/7] with hyaline matrix. Spore wall two-layered, even to uneven, 1–2.5 µm thick, with outer layer 1–2 µm thick, and inner layer 0.5 µm thick. Outer layer hyalineyellowish, smooth. Host range: Eryngium planum section Plana. Distribution: Europe (Germany, Italy, Poland, Romania, Russian Federation, Ukraine), Asia (Kazakhstan; Kyrgyzstan; Russian Federation, Siberia). Type: Italy. Alba, near Cuneo, on Eryngium planum, leg. R. Ciferri. Type not seen, probably lost. Additional specimens examined: Poland. Saldowice prope Pulawy, on Eryngium planum, 23 Jul 1935, leg. J. Kochman, TUR 060378; Lublin Province: near Zmudz Reserve, SE of Chelm, 14 Jul 2006, leg. J. & M. Piątek, KRAM F; Pulawy, 28 Aug 1958, leg. J. Kochman, NY. Romania. Transsilvania: pr. oppid. Sovata. 46°35′N, 25°03′E. c. 500 m, 27 Jun 1983, leg. T., U. & K. Vánky, Ust. Ex. 507, NY. Russian Federation. Omsk province: near Omsk, Aug 1922, leg. Murashkinsky, BPI 175291 (as E. eryngii); 14 Jul 1927, leg. Ziliny, BPI 175290. Ukraine. Ivano-Frankivsk region: Polonica-Prutec near Mykulychyn, Jul 1914, leg. A. Wróblewski, BPI 175293 (as E. eryngii). Notes: Entyloma eryngii-plani differs from E. eryngii in having smaller spores, thinner spore walls and flat sori. It is similar to E. eryngii-cretici but has a slightly different color of spore matrix and an adaxial anamorph, while the latter species has no known anamorph. Furthermore, in the phylogenetic analysis, the sequence of E. eryngii-plani was in a different lineage from E. carmeli and E. eryngii-cretici, another species on hosts from section Plana. Entyloma eryngii-plani was reported from Germany to Siberia and is likely to have a wide distribution in temperate regions of Europe and Asia. Entyloma eryngii-tricuspidati Maire, Bull. Soc. Hist. Nat. Afr. N. 10:138 (1919) Fig. 3f, g Sori in leaves forming polygonal, polyangular, amphigenous, flat, scattered, coffee brown spots, 0.5–1 mm diam, cryptic. Anamorph absent. Spores singular, subglobose, ellipsoidal to subpolyhedrally-irregular, 8–10 × 9–13 µm diam [av. ± SD, 8.8 ± 1.5 × 11.6 ± 1.9 µm, n = 100/2] with subhyaline, pale yellowish matrix. Spore wall two-layered, uneven, 2–2.5 µm thick, with outer layer 1.5–2 µm thick, and inner layer 0.5 µm thick. Outer layer subhyaline, smooth. Host range: Eryngium tricuspidatum section Plana. Distribution: Africa (Algeria). Lectotype: Algeria. Chiffa Gorge, on Eryngium tricuspidatum, 03 Apr 1915, leg. R. Maire, NY 0738486b. Additional specimens examined: Algeria. Miliana, near Ain-n-Sour, on Eryngium tricuspidatum, 31 Jun 1917, leg. R. Maire, NY 0738486a. Notes: Maire (1919) and Vánky (2009) noted that Entyloma eryngii-tricuspidati differs from E. eryngii-dichotomi by the absence of an anamorphic stage. However, there is also a consistent difference in the spore wall thickness, which is 2–2.5 µm in E. eryngiitricuspidati and 1–2 µm in E. eryngii-dichotomi. So far Entyloma eryngii-tricuspidati is known only from Algeria. Entyloma ho-chunkii Savchenko, Carris, Castl., Heluta, Wasser & Nevo, sp. nov. FIG. 3h–j MycoBank MB807875 Etymology: derived from the Ho-Chunk tribe of Native Americans in Wisconsin and Iowa where the species was found. Sori in leaves, delimited by veins, amphigenous, as flat, linear, elongated, scattered over the leaf surface, pale yellowish spots, 1–20 mm long and 0.5–2 mm wide. Anamorph absent. Spores singular or crowded, not firmly agglutinated, globose, subglobose, ellipsoidal, subpolyhedrally-irregular, 7–10.5 × 8–13 µm diam [av. ± SD, 9.6 ± 1.6 × 10.5 ± 2.1 µm, n = 350/7] with hyaline matrix. Spore wall two-layered, even, 1.5–2 µm thick, with outer layer 1–1.5 µm thick, and inner layer 0.5 µm thick. Outer layer hyaline, smooth. Host range: Eryngium yuccifolium section Panniculata. Distribution: North America (United States). Type: United States. Wisconsin: Paoli, Dane County, on Eryngium yuccifolium, 29 Sep 1940, leg. H.C. Greene, BPI 175298; Isotype in MICH 00073560. Additional specimens examined (paratypes): on Eryngium yuccifolium: United States. Iowa: Decorah, on Eryngium yuccifolium, 27 Jun 1885, leg. Holway, BPI 175295; Wisconsin: Paoli, Dane County, 04 Sep 1940, leg. H.C. Greene, BPI 198976; Dane County, right of way near Paoli, 25 Jun 1944, leg. H.C. Greene, WIS; Dane County, Paoli, 06 Jun 1948, leg. H.C. Greene, WIS; Dane County, Basco, 01 Sep 1940, leg. H.C. Green, WIS. Notes: This smut was reported first from Iowa by Clinton (1902) as E. eryngii, and at that time it was the only known Entyloma species on Eryngium. Later H.C. Green (1942) collected several specimens in Wisconsin and, following Clinton's species concept, attributed them to the European species E. eryngii. However, Entyloma ho-chunkii is the only Entyloma species found on North American Eryngium hosts. In addition to a different geographic distribution, it differs from all of the other species by linear, not circular or polygonal, sori, smaller spores and host belonging to a different section of the genus. Entyloma ho-chunkii looks similar to E. eryngii-maritimi in terms of the small, hyaline spores. Nevertheless, the spores of the latter are smaller (av. 7.6 × 8.3 vs. 9.6 × 10.5), with an uneven spore wall thickness, while the spore walls of E. ho-chunkii are consistently uniform in thickness. Entyloma maroccanum Maire, Mém. Soc. Sci. Nat. Maroc. 45:48 (1937) Entyloma eryngii var. maroccanum (Maire) Cif., Quaderno Ist. Bot. Univ. Pavia 27:111 (1963) Sori in leaves forming circular or angular, well delimited, amphigenous, slightly thickened spots, 1–1.5 mm diam, pale yellowish brown. Spores densely agglomerated, globose, ellipsoidal to subpolyhedrally irregular, 7–11 × 9–13.5 µm, subhyaline to pale yellow. Spore wall even or slightly uneven, 0.8–1.5 µm thick, smooth. Anamorph present. Host range: Eryngium maroccanum section Plana. Type: Morocco. Moyen-Atlas: Ras-el-Ma, near Azrou, 1600 m., on Eryngium maroccanum, 18 Apr 1923, leg. R. Maire, Herb. Maire, MPU. Notes: Type not seen, description is taken from the original. Entyloma sp. on Eryngium macrocalyx Sori in leaves forming circular or polyangular, amphigenous, flat, thickened, yellowish brown, pinkish purple spots, 0.5–2 mm diam. Anamorph absent. Spores crowded, densely packed, globose, subglobose, ovoid, subpolyhedrally irregular, 15 × (7)9–18(21) µm diam. Spore wall two-layered, uneven, 1–2(4) µm thick. Outer layer yellowish, brown, smooth. Host range: Eryngium macrocalyx section Halobia. Notes: This smut has been reported as E. eryngii from Kazakhstan and Uzbekistan (Ramazanova et al. 1987). We were not able to obtain specimens for microscopic examination, but according to the original description it possesses some clear differences from all known species of Entyloma on Eryngium: It differs from E. eryngii by the morphology of sori and thinner spore walls, has larger spores than any of the remaining species of Entyloma on Eryngium and has a unique sorus color. Therefore it is treated here as a separate taxon without formal description. DISCUSSION Macro- and micromorphological study, combined with phylogenetic analyses, revealed a greater amount of diversity among Entyloma species on Eryngium than was reflected in the existing treatments (Vánky 2009, 2011). Eleven species now are recognized: E. argentinense and E. ho-chunkii from the Americas; E. carmeli, E. eryngii, E. eryngiialpini, E. eryngii-cretici, E. eryngii-maritimi and E. eryngii-plani from Eurasia; E. eryngiidichotomi, E. eryngii-tricuspidati and E. maroccanum from Africa. The speciation of Entyloma on Eryngium is clearly related to the evolutionary relationships of the hosts. Indeed, high speciation is not rare among smut fungi and was reported for Microbotryum Lév. (Freeman et al. 2002; Lutz et al. 2005, 2008; le Gac et al. 2007; Denchev et al. 2008; Devier et al. 2010), Antherospora R. Bauer et al. (Bauer et al. 2008, Piątek et al. 2013), Tilletia Syd. & P. Syd. (Castlebury and Carris 1999; Carris et al. 2007) and Anthracoidea Bref. (Savchenko et al. 2013). It might seem surprising that more than half of the Entyloma species recognized occur on members of section Plana, but it is noteworthy that this section was found to be heterogeneous (Calviño et al. 2008, 2010). Our phylogenetic analyses demonstrate that all Entyloma species on Apiaceae clustered within the Apiaceae clade, sharing a common ancestor with other groups of Entyloma on Asterids. The distribution in the phylogenetic topology does not suggest monophyly for the species on Eryngium, in that E. magocsyanum from Tordylium Tourn. ex L., another genus of Apiaceae, also occurred within this clade. Eryngium and Todylium belong to different subfamilies and it was proved by molecular phylogeny (Downie et al. 2000). Therefore, the possible close relationship between E. magocsyanum and Entyloma on Eryngium may be explained by a recent host shift. Moreover, E. calceolariae clusters in the same subgroup. In this case, the host is not a member of Apiaceae or a closely related family and instead belongs to Lamiales. Such a distribution could be explained by host jumps, which are not rare in smut fungi (Begerow et al. 2002, 2004). Two species from Israel, E. carmeli and E. eryngii-cretici, are not closely related, and there is a 12 bp difference in the ITS nucleotide sequences between the specimens infecting Eryngium falcatum and Eryngium creticum. Furthermore, according to our data, E. carmeli is possibly more closely related to E. magocsyanum and E. eryngii-plani than to E. eryngiicretici. Even so, their hosts belong to the same section and both species grow in the same region. Hence, it does not appear that geographical isolation played a role in the evolutionary history of these species. However, our observations revealed that they grow in different habitats and therefore ecological sympatric speciation may have occurred in this case. In conclusion, our data reveal a strong correlation of these parasites with their respective hosts. We accept seven described species of Entyloma on Eryngium and describe four new species based on molecular and/or morphological data and hosts from different sections. Our results provide an additional resource for accurate species identification. The fact that there are only two species from the Americas where a second center of speciation of Eryngium occured (Calviño et al. 2008, 2010) makes it clear that additional collections in that region might lead to recognition of greater diversity within the genus. KEY TO SPECIES OF ENTYLOMA ON ERYNGIUM 1. Sori pustulate, cracked when mature, spore wall 2.5–7 µm thick. On Eryngium campestre………………………………………………………………………… E. eryngii 1a. Sori not so, spore wall thinner. On other hosts …………………………………………2 2. On American species of Eryngium ………………………………………………………3 2a. On Eurasian or African species of Eryngium………………..………………………… 4 3. Spore wall 1.5–5 µm thick, uneven, spores yellowish brown, yellowish. On Eryngium nudicaule …………………………………………………………………..E. argentinense 3a. Spore wall 1.5–2 µm thick, even, spores hyaline. On Eryngium yuccifolium …………………………………………………………………………….....E. ho-chunkii 4. Spore wall up to 4 µm thick. On Eryngium falcatum …………………………E. carmeli 4a. Spore wall thinner, up to 3 µm thick. On other hosts…………………………………...5 5. Spore wall 0.8–1.5 µm thick, on Eryngium maroccanum ………………E. maroccanum 5a. Spore wall thicker, on other species ……………………………………………………6 6. Spores small, 7–10 µm (av. 8.3 µm) diam, sori greenish. On Eryngium maritimum………………………………………………………………E. eryngii-maritimi 6a. Spores larger, sori not so. On other hosts………………………………………………7 7. Anamorph abundant, amphigenous, on Eryngium dichotomum ………………………………………………………………………..E. eryngii-dichotomi 7a. Anamorph absent, or if present, not abundant, on one side of the leaves. On other species………………………………………………………………………………………8 8. Sori scattered, well delimited, cryptic, coffee brown to dark brown. On Eryngium tricuspidatum ……………………………………………………….E. eryngii-tricuspidati 8a. Sori yellowish, yellowish brown, fawn, confluent. On other species ………………….9 9. Anamorph absent, spores hyaline-yellowish, gold, on Eryngium creticum …………………………………………………………………………….E. eryngii-cretici 9a. Anamorph adaxial, spores hyaline, on Eryngium planum …………….E. eryngii-plani 9b. Anamorph abaxial, spores hyaline, on Eryngium alpinum ……………E. eryngii-alpini ACKNOWLEDGMENTS We thank the curators of herbaria BPI, DAOM, FT, G, K, KRAM, MICH, NY, TUR, WIS and WSP for the loan of specimens, Dominik Begerow and Matthias Lutz for useful advice on DNA extraction, Marcin Piątek for providing the specimen of E. eryngii-plani and Carolina Calviño for the clarification on the current systematics of Eryngium. LITERATURE CITED Ainsworth GC, Sampson K. 1950. The British smut fungi (Ustilaginales). Kew, UK: C.M.I. 137 p. Anonymous. 1960. Index of plant diseases in the United States. USDA Agricultural Handbook 165:1–531. Bauer R, Lutz M, Begerow D, Piątek M, Vánky K, Bacigálová K, Oberwibkler F. 2008. Anther smut fungi on monocots. Mycol Res 112:1297–1306. ———, Oberwinkler F, Vanky K. 1997. Ultrastructural markers and systematics in smut fungi and allied taxa. Can J Bot 75:1273–1314. Begerow D, Göker M, Lutz M, Stoll M. 2004. On the evolution of smut fungi on their hosts. In: Agerer R, Peiepenbring M, Blanz P, eds. Frontiers in Basidiomycete mycology. Eching, Germany: IHW-Verlag. p 81– 98. ———, Lutz M, Oberwinkler F. 2002. Implications of molecular characters for the phylogeny of the genus Entyloma. Mycol Res 106:1392–1399. Calviño CI, Martinez SG, Downie SR. 2008. The evolutionary history of Eryngium (Apiaceae, Saniculoideae): rapid radiations, long distance dispersals and hybridizations. Mol Phyl Evol 46:1129–1150. ———, Martinez SG, Downie SR. 2010. Unraveling the taxonomic complexity of Eryngium L. (Apiaceae, Saniculoideae): Phylogenetic analysis of 11 noncoding DNA loci corroborates rapid radiations. Plant Divers Evol 128:137–149. Carris LM, Castlebury LA, Huang G, Alderman SC, Luo J, Bao X. 2007. Tilletia vankyi, a new species of reticulate-spored bunt fungus with non-conjugating basidiospores infecting species of Festuca and Lolium. Mycol Res 111:1386–1398. Castlebury LA, Carris LM. 1999. Tilletia walkeri, a new species on Lolium multiflorum and L. perenne. Mycologia 71:449–455. Castresana J. 2000. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Biol Evol 17:540–552. Chakrabarty P. 2010. Genetypes: a concept to help integrate molecular phylogenetics and taxonomy. Zootaxa 2632:67–68. Ciferri R. 1924. Prima contribuzione allo studio degli Ustilaginales (Nos. 1–22). Bull Soc Bot Italy 2:46–59. Clinton GP. 1902. North American Ustilagineae. J Mycol 8:128–156. Corda ACI. 1839. Icones fungorum hucusque cognitorum. Vol. 3. Prague. Crous PW, Gams W, Stalpers JA, Robert V, Stegehuis G. 2004. MycoBank: an online initiative to launch mycology into the 21st century. Stud Mycol 50:19–22. de Bary A. 1874. Protomyces microsporus und seine Verwandten. Bot Zeitung 32:97–108. Denchev CM, Giraud T, Hood ME. 2009. Three new species of anthericolous smut fungi on Caryophyllaceae. Mycol Balcan 6:79–84. Devier B, Aguileta G, Hood ME, Giraud T. 2010. Using phylogenies of pheromone receptor genes in the Microbotryum violaceum species complex to investigate possible events of speciation by hybridization. Mycologia 102:689–696 Downie SR, Katz-Downie DS, Watson MF. 2000. A phylogeny of the flowering plant family Apiaceae based on chloroplast DNA rpl16 and rpoC1 intron sequences: toward a suprageneric classification of subfamily Apioideae. Am J Bot 87:273–292. Ellett CW. 1959. The Ustilaginales (smut fungi) of Ohio. Ohio J Sci 59:313–318. Fischer GW. 1953. Manual of the North American smut fungi. New York: Ronald Press Co. 343 p. Freeman FB, Doung KK, Shi T, Hughes CF, Perlin MH. 2002. Isolates of Microbotryum violaceum from North American host species are phylogenetically distinct from their European host-derived counterparts. Mol Phylogenet Evol 23:158–170. Gilman JC, Archer WA. 1929. The fungi of Iowa parasitic on plants. Iowa State J Sci 3:299–507. Greene HC. 1942. Notes on Wisconsin parasitic fungi II. Trans Wisconsin Acad Sci 34:83–98. Hill DM, Bull JJ. 1943. An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis. Syst Biol 42:182–192. Huelsenbek JP, Ronquist F. 2001. MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics 817:754–755. Katoh K, Misawa K, Kuma K, Miyata T. 2002. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Ac Res 30:3059–3066. ———, Toh H. 2008. Recent developments in the MAFFT multiple sequence alignment program (outlines 6). Brief Bioinform 9:286–298. le Gac M, Hood ME, Fournier E, Giraud T. 2007. Phylogenetic evidence of host-specific cryptic species in the anther smut fungus. Evolution 61:15–26. Legon N, Henrici A, Roberts PJ, Spooner BM, Watling R. 2005. Checklist of the British and Irish Basidiomycota. Kew, UK: Royal Botanic Gardens. 156 p. Lutz M, Göker M, Piątek M, Kemler M, Begerow D, Oberwinkler F. 2005. Anther smuts of Caryophyllaceae: molecular characters indicate host-dependent species delimitation. Mycol Prog 4:225–238. ———, Piątek M, Kemler M, Chlebicki A, Oberwinkler F. 2008. Anther smuts of Caryophyllaceae: molecular analyses reveal further new species. Mycol Res 112:1280–1296. Maire R. 1917. Champignons nord-africans nouveaux ou peu connus. Bull Soc Hist Nat Afrique N 8:134– 200. ———. 1919. Schedae ad Mycothecam Boreali-Africanam. Bull Soc Hist Nat Afrique N 10:130–151. ———, Werner RG. 1937. Fungi Maroccani. Mem Soc Sci Nat Maroc 45:1–147. Pantidou ME. 1973. Fungus-host index for Greece. Athens: Benaki Phytopathol Inst, Kiphissia. 382 p. Piątek M, Lutz M, Chater AO. 2013. Cryptic diversity in the Antherospora vaillantii complex on Muscari species. IMA Fungus 4:5–19. Ramazanova SS, Ahmedova FG, Sagdullaeva MS, Kirgizbaeva HM, Gaponenko NI. 1987. Flora gribov Uzbekistana. Tom IV. Fan, Tashkent, Uzbeckistan: Golovnevye gribi. 148 p. Rannala B, Yang Z. 1996. Probability distribution of molecular evolutionary trees: a new method of phylogenetic inference. J Mol Evol 43:304–311. Rayss T, Zwirn E. 1944. Some interesting Ustilaginales new to Palestine. Palest J Bot Jerusalem Ser 3:114– 116. Ronquist F, Huelsenbeck JP. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19:1572–1574. Savchenko KG, Heluta VP, Wasser SP, Nevo E. 2010. Smut fungi of Israel: a preliminary check list. Mycol Balcan 7:111–116. ———, Lutz M, Piątek M, Heluta VP, Nevo E. 2013. Anthracoidea caricis-meadii is a new North American smut fungus on Carex sect. Paniceae. Mycologia 105:181–193. Servazzi O. 1950. Sur un Entyloma (sp. n.?) di Eryngium alpinum L. Nuovo Giorn Bot Italy 57:306–308. Spegazzini C. 1925. Ustilagineas Argentinas nuevas o criticas. Rev Argent Bot 1:145–156. Swofford DL. 2002. PAUP*: phylogenetic analysis using parsimony (*and other methods). 4.0b8. Sunderland, Massachusetts: Sinauer Associates. Vánky K. 2009. Taxonomic studies on Ustilaginomycetes 29. Mycotaxon 110:289–324. ———. 2011. Smut fungi of the world. St. Paul, Minnesota: APS Press. 1458 p. Viennot-Bourgin G. 1956. Mildious, oidiums, caries, charbons, rouilles des plantes de France I–II. Encyclopédie mycologique 26. Paris : Paul Lechevalier. 317 p. Voss W. 1889–1892. Mycologia Carniolica. Ein beitrag zur Pilzkunde des Alpenlanders. Berlin: R. Friedländer & Sohn. 330 p. White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification and direct sequencing of fungal ribosomal RNA sequences for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ, eds. PCR protocols: a guide to methods and applications. San Diego, California: Academic Press. p 315–322. Wolff H. 1913. Umbelliferae-Saniculoideae. In: Engler A, ed. Das Pflanzenreich. Vol. IV 228 p. Leipzig, Berlin: Wilhelm Engelmann. 320 p. Wörz A. 2005. A new subgeneric classification of the genus Eryngium L. (Apiaceae, Saniculoideae). Bot Jahrb Syst 126:253–259. Zundel GL. 1953. The Ustilaginales of the world. Contr Dept Bot Pennsylvania State Coll School Agric 176:1–410. LEGENDS FIG 1. The phylogram inferred from the Bayesian inference of rDNA ITS sequences of Entyloma. The posterior probabilities from Bayesian inference above 80% are shown above the branches, and the MP bootstrap support values >70% from 1000 replicates below the branches. Sequences of Entyloma on Eryngium are in bold. Scale bar represents 0.05 substitutions per site. The tree was rooted to E. atlanticum. FIG 2. Sori and spores of Entyloma species on Eryngium. A–C E. carmeli HAI 4521; D–F E. eryngii-alpini G 00111412; G–I E. eryngii-cretici HAI 4525; J, K E. eryngii FT 5322/2; L, M E. eryngii-dichotomi NY 00738484. – Scale bars: A, B, D, E, G, H, J, L = 5 mm; C, F, I, K, M = 10 µm. FIG 3. Sori and spores of Entyloma species on Eryngium. A–C E. eryngii-plani BPI 175291; D, E E. eryngii-maritimi BPI 175289; F, G E. eryngii-tricuspidati NY 0738486b; H–J E. ho-chunkii BPI 175298. Scale bars: A, B, D, F, H, I = 5 mm; C, E, G, J = 10 µm. FOOTNOTES Submitted 30 Sep 2013; accepted for publication 13 Feb 2014. 1 Corresponding author. E-mail: savchenko.kyryll@gmail.com TABLE I. Specimens used in this study Species Voucher Year E. carmeli HAI 4521 2011 E. carmeli HAI 4522 2011 E. eryngii WSP 70820 WSP 70822 FT 5322/2 1997 E. eryngii E. eryngii E. eryngii E. eryngii WSP 18106 WSP 2091 1999 1879 1930 1925 E. eryngii DAOM 70831 FT 5322/1 1974 E. eryngii FT 5322/3 1883 E. eryngii FT 5322/3a 1906 E. eryngii BPI 1919 175263 BPI 1931 175280 G00111412 1987 E. eryngii E. eryngii E. eryngiialpini E. eryngiialpini E. eryngiialpini E. eryngiialpini E. eryngiialpini E. eryngiialpini E. eryngiicretici E. eryngiicretici 1906 G00111416 1987 G00111417 1987 G00111413 1987 G00111414 1987 G00111418 1987 HAI 4525 2011 HAI 4526 2011 Host Eryngium falcatum Eryngium falcatum Eryngium campestre Eryngium campestre Eryngium campestre Eryngium campestre Eryngium campestre Eryngium campestre Eryngium campestre Eryngium campestre Eryngium campestre Eryngium campestre Eryngium campestre Eryngium alpinum Eryngium alpinum Eryngium alpinum Eryngium alpinum Eryngium alpinum Eryngium alpinum Eryngium creticum Eryngium creticum Country of origin Israel Collector Austria K.G. Savchenko K.G. Savchenko Th. Barta Austria Th. Barta Czech Republic Czech Republic Germany Prof. Niessl A. Ludwig Germany G. Oertel Hungary K. Vánky Italy G. Passerini Italy Spain C. Massalongo P. Unamuno Spain P. Unamuno Israel GenBank number KF310892 KF310893 KF310896 KF310897 J. Hruby Switzerland A. Bolay Switzerland A. Bolay Switzerland A. Bolay Switzerland A. Bolay Switzerland A. Bolay Switzerland A. Bolay Israel Israel K.G. Savchenko K.G. Savchenko KF310894 KF310895 E. eryngiicretici E. eryngiidichotomi E. eryngiimaritimi E. eryngiimaritimi E. eryngiiplani E. eryngiiplani E. eryngiiplani E. eryngiiplani E. eryngiiplani E. eryngiiplani E. eryngiiplani E. eryngiitricuspidati E. eryngiitricuspidati E. hochunkii E. hochunkii E. hochunkii E. hochunkii E. hochunkii E. hochunkii E. hochunkii HAI 4527 2001 NY 00738484 BPI 175289 K(M) 106003 KRAM 1915 1935 1996 2006 1935 TUR 060378 NY 1958 NY 1983 BPI 175291 BPI 175290 BPI 175293 NY 0738486b NY 0738486a BPI 175298 MICH 00073560 BPI 175295 BPI 198976 WIS 1922 1927 1914 1915 1917 1940 1940 1885 1940 1944 WIS 1948 WIS 1940 Eryngium creticum Eryngium dichotomum Eryngium maritimum Eryngium maritimum Eryngium planum Eryngium planum Eryngium planum Eryngium planum Eryngium planum Eryngium planum Eryngium planum Eryngium tricuspidatum Eryngium tricuspidatum Eryngium yuccifolium Eryngium yuccifolium Eryngium yuccifolium Eryngium yuccifolium Eryngium yuccifolium Eryngium yuccifolium Eryngium yuccifolium Israel T. Pavlichek Algeria R. Maire France Anonimous UK, Wales N. Stringer Poland Poland J. & M. Piątek J. Kochman Poland J. Kochman Romania T., U., & K. AY081034 Vánky Murashkinsky Russian Federation Russian Federation Ukraine Ziliny Algeria A. Wroblewski R. Maire Algeria R. Maire USA, Wisconsin USA, Wisconsin USA, Iowa H.C. Greene USA, Wisconsin USA, Wisconsin USA, Wisconsin USA, Wisconsin H.C. Greene Holway H.C. Greene H.C. Greene H.C. Greene H.C. Greene TABLE II. Morphological characters of Entyloma species on Eryngium Species E. argentinense E. carmeli E. eryngii E. eryngiialpini Character of spots Confl uence of sori Diam of sori, mm flat, slightly swollen flat not specif ied no not specif ied 0.2–4 flat, pustulate, cracked flat yes Color of sori dark brown Medi um spore lengt h, µm 9–13 Medi um spore width , µm 10– 17 brown, dark brown 12.1 13.5 0.3– 1.5 pale brown, brown 12.9 yes 0.2–1 yellowish white, yellowish brown Wall, µm Outer layer, µm Inner layer, µm Charact er of spore wall Color of spore matrix Color of outer layer Position of anamorph 1.5–5 not specifie d 0.5–1 uneven yellow, yellowis h brown hyalinegold absent? uneven amphigen whitish yellowish 14.8 2.5–7 2–6 0.5–1 uneven not seen not seen 10.8 12.4 1.5– 2.5 1.5–2 0.5 even to uneven yellow, yellowish brown yellowish brown, pale golden subhyaline, pale yellowish hyaline, pale yellowish absent? 1.5–4 not specif ied 1.5–3 abaxial White 10.6 12.0 1.5– 2.5 (3) 1.5–2 (2.5) 0.5 even to uneven absent Absent amphigen, abundant white, yellowish white absent Absent hyalineyellowis h subhyali ne adaxial Whitish absent Absent hyaline not specified absent present Absent not specified E. eryngiicretici flat, slightly swollen yes 0.2–4 yellowish brown, fawn E. eryngiidichotomi flat, thickened yes 0.5– 2.5 pale brown 9.5 12.0 1–2 1–1.5 0.5 uneven E. eryngiimaritimi flat, slightly thickened flat, slightly thickened flat no 0.1–3 greenish gray 7.6 8.3 1–2 1–1.5 0.5 uneven hyaline yes 0.5–3 yellowish green, yellowish brown 10.9 12.4 1–2.5 1–2 0.5 even to uneven hyaline no 0.5–1 Coffee brown 8.8 11.6 2–2.5 1.5–2 0.5 uneven flat stripes flat no no 1–20 1–1.5 pale yellowish pale yellowish brown 9.6 7–11 10.5 9– 13.5 1.5–2 0.8– 1.5 1–1.5 not specif ied 0.5 not specifie d even even, slightly uneven subhyaline, pale yellowish hyaline subhyaline, pale yellow E. eryngiiplani E. eryngiitricuspidati E. ho-chunkii E. maroccanum pale yellowish, pale golden subhyaline, pale yellowish pale yellowis h hyaline,p ale yellowis h pale yellowis h subhyali ne, pale yellowis h hyaline Color of anamorph