C rem as t o s p er m a ( an d ot h er evolu t ion ar y digres s ion s ) Molecular phylogenetic, biogeographic, and taxonomic studies in Neotropical Annonaceae Cremastosperma (en andere evolutionaire dwalingen) Moleculair fylogenetische, biogeografische, en taxonomische studies in Neotropische Annonaceae (Met een samenvatting in het Nederlands) C r e m a s t o s p e r m a (y otras digresiones evolutivas) Filogenética molecular, biogeografía, y estudios taxonómicos en Annonaceae del Neotrópico (Con un resumen en Español) • 1 2 • C rem as t o s p er m a ( an d ot h er evolu t ion ar y digres s ion s ) Molecular phylogenetic, biogeographic, and taxonomic studies in Neotropical Annonaceae Cremastosperma (en andere evolutionaire dwalingen) Moleculair fylogenetische, biogeografische, en taxonomische studies in Neotropische Annonaceae (Met een samenvatting in het Nederlands) C r e m a s t o s p e r m a (y otras digresiones evolutivas) Filogenética molecular, biogeografía, y estudios taxonómicos en Annonaceae del Neotrópico (Con un resumen en Español) Proefschr ift Ter฀verkrijging฀van฀de฀graad฀van฀doctor฀aan฀de฀Universiteit฀Utrecht op฀gezag฀van฀de฀Rector฀Magnificus,฀Prof.dr.฀W.H.฀Gispen ingevolge฀het฀besluit฀van฀het฀College฀voor฀Promoties in฀het฀openbaar฀te฀verdedigen op฀vrijdag฀10฀juni฀2005 des฀middags฀te฀12.45฀uur door Michael฀ David฀ Pir ie geboren฀op฀18฀augustus฀1977฀te฀Londen฀(V.K.) • 3 Promotor:฀ ฀ ฀ ฀ ฀ Co-promotor:฀ ฀ ฀ ฀ ฀ 4 • Prof.฀Dr.฀P.J.M.฀Maas,฀hoogleraar฀ aan฀de฀Universiteit฀Utrecht,฀Faculteit฀Biologie Dr.฀L.W.฀Chatrou,฀Faculteit฀Biologie, Universiteit฀Utrecht • 5 ISBN฀nummer:฀90-393-39538 Cover:฀Flora฀Neotropica฀Basemap,฀prepared฀by฀Hendrik฀R.฀Rypkema.฀ Flower฀buds฀of฀Cremastosperma฀drawings฀by฀H.R.฀Rypkema;฀photos฀ (from฀top฀to฀bottom฀and฀left฀to฀right:฀C.฀brevipes,฀C.฀cauliflorum,฀฀ C.฀microcarpum,฀C.฀bullatum,฀C.฀yamayakatense,฀C.฀monospermum,฀฀ C.฀pendulum฀and฀C.฀leiophyllum)฀by฀P.J.M.฀Maas,฀L.W.฀Chatrou฀&฀฀ M.D.฀Pirie฀ Layout:฀Pieter฀van฀Dorp฀van฀Vliet,฀฀B&V฀Biologie Printing:฀Febodruk฀bv,฀฀Enschede/Utrecht 6 • Contents Chapter฀ 1฀ General฀introduction฀ 9฀ ฀ Chapter฀ 2฀ ฀ ฀ Phylogeny฀reconstruction฀and฀molecular฀dating฀ in฀four฀Neotropical฀genera฀of฀Annonaceae:฀the฀ effect฀of฀taxon฀sampling฀in฀age฀estimations 17 Chapter฀ 3฀ ฀ ฀ ฀ ‘Andean-centred’฀genera฀in฀the฀short฀branch฀ clade฀of฀Annonaceae:฀testing฀biogeographic฀ hypotheses฀using฀phylogeny฀reconstruction฀ and฀molecular฀dating 39฀ Chapter฀ 4฀ ฀ ฀ An฀ancient฀paralogue฀of฀the฀cpDNA฀trnL฀ (UAA)-trnF฀(GAA)฀region฀in฀Annonaceae and฀its฀application฀in฀phylogeny฀reconstruction 65 Chapter฀ 5฀ ฀ ฀ Summary฀ Revision฀and฀phylogeny฀of฀ Cremastosperma (Annonaceae) ฀ ฀ 87 ฀ ฀ 189 Samenvatting฀ ฀ Resumen฀ ฀ ฀ ฀ References฀ 193 ฀ 197 ฀ ฀ ฀ 201 • 7 ฀ Dankwoord฀ ฀ ฀ ฀ List฀of฀publications฀ ฀ ฀ Cur r iculum฀vitae฀ 211 ฀ ฀ 215 ฀ 217 Appendix฀A฀ ฀ 218 Appendix฀B฀ ฀ 249฀ ฀ 8 • ฀ Chapter 1 Introduction This฀thesis฀contains฀the฀results฀of฀molecular฀systematic฀studies฀of฀groups฀ in฀ the฀ flowering฀ plant฀ family฀Annonaceae฀ and฀ a฀ taxonomic฀ revision฀ of฀ the฀ genus฀Cremastosperma,฀which฀occurs฀in฀the฀Neotropics.฀Each฀Chapter฀includes฀ its฀ own฀ introduction,฀ but฀ it฀ is฀ my฀ intention฀ to฀ introduce฀ here฀ some฀ of฀ the฀ subjects,฀approaches,฀and฀more฀important฀concepts฀employed.฀An฀overview฀of฀ the฀contents฀of฀the฀rest฀of฀the฀thesis฀is฀also฀presented. Work฀in฀this฀thesis฀falls฀under฀what฀I฀believe฀to฀be฀two฀interrelated฀but฀ very฀ different฀ branches฀ of฀ biological฀ research.฀These฀ can฀ be฀ summarised฀ as฀ 1)฀ the฀ description฀ of฀ biological฀ diversity฀ ( α-taxonomy)฀ and฀ 2)฀ the฀ increase฀ of฀our฀understanding฀of฀patterns฀and฀processes฀underlying฀that฀diversity.฀The฀ biological฀ diversity฀ under฀ study฀ all฀ belongs฀ to฀ a฀ family฀ of฀ flowering฀ plants฀ known฀as฀the฀Annonaceae.฀Annonaceae,฀first฀described฀by฀Jussieu฀in฀1789,฀grow฀ in฀the฀forms฀of฀trees,฀shrubs,฀and฀woody฀vines,฀and฀are฀almost฀entirely฀limited฀ to฀tropical฀climes,฀particularly฀tropical฀Asia฀and฀Africa฀(the฀old฀world฀tropics),฀ and฀Latin฀America฀(the฀new฀world,฀or฀Neotropics).฀Those฀more฀familiar฀with฀ temperate฀plants฀might฀still฀have฀encountered฀Annonaceae฀in฀the฀form฀of฀the฀ juice฀of฀the฀cherimoya฀fruit฀(Annona฀ c her imola฀Mill.)฀or฀in฀perfumes฀made฀ with฀ylang-ylang฀(Cananga฀odorata฀(Lam.)฀Hook.฀f.฀&฀Thoms.). Annonaceae, taxonomy and systematics There฀are฀around฀2,500฀species฀of฀Annonaceae,฀900฀of฀which฀are฀found฀in฀ the฀Neotropics฀(Chatrou฀et฀al.,฀2004).฀These฀numbers฀are฀an฀estimation฀on฀at฀ least฀two฀counts:฀firstly,฀the฀biological฀diversity฀of฀much฀of฀the฀world’s฀tropics฀ is฀very฀poorly฀known.฀New฀species฀of฀plants฀are฀described฀every฀day฀(13฀new฀ species฀of฀Cremastosperma฀have฀been฀described฀in฀the฀preparation฀of฀this฀thesis฀ alone).฀Secondly,฀the฀species฀as฀a฀concept฀is฀difficult,฀if฀not฀impossible,฀to฀define.฀ In฀this฀thesis,฀species฀are฀distinguished฀following฀the฀taxonomic฀species฀concept฀ of฀Davis฀&฀Heywood฀(1963),฀in฀which฀they฀are฀described฀as฀“assemblages฀of฀ �eneral introduction • 9 individuals฀with฀morphological฀features฀in฀common฀and฀separable฀from฀other฀ such฀assemblages฀by฀correlated฀morphological฀discontinuities฀in฀a฀number฀of฀ features.”฀This฀approach฀represents฀not฀so฀much฀a฀concept฀of฀what฀a฀species฀ actually฀is,฀but฀rather฀a฀practical฀solution฀for฀living฀in฀a฀world฀where฀species฀very฀ clearly฀exist,฀but฀where฀it฀is฀not฀always฀easy,฀or฀even฀possible,฀to฀identify฀them.฀ Exactly฀how฀different฀two฀such฀assemblages฀have฀to฀be฀in฀order฀to฀constitute฀ separate฀species฀remains฀a฀subjective฀decision.฀This฀subjectivity฀cannot,฀in฀my฀ opinion,฀be฀avoided.฀This฀is฀due,฀firstly,฀to฀the฀variety฀of฀unique฀circumstances฀ in฀which฀speciation฀events฀occur.฀Secondly,฀should฀you฀find฀grounds฀on฀which฀ to฀ accept฀ one฀ particular฀ species฀ concept฀ from฀ the฀ large฀ available฀ selection฀ (the฀existence฀of฀which฀probably฀owes฀a฀great฀deal฀to฀those฀various฀unique฀ circumstances),฀the฀demands฀for฀data฀necessary฀to฀apply฀it฀are฀often฀unrealistic.฀ Increases฀in฀collections฀will฀continue฀to฀unearth฀new฀variation,฀which฀in฀some฀ cases฀may฀challenge฀views฀of฀species฀delimitations.฀Species฀defined฀following฀ the฀ taxonomic฀ species฀ concept฀ can฀ only฀ be฀ regarded฀ as฀ hypotheses฀ of฀ the฀ underlying฀biological฀reality฀(however฀that฀is฀to฀be฀defined).฀These฀hypotheses฀ are,฀ however,฀ a฀ vitally฀ important฀ tool฀ for฀ those฀ studying฀ organisms฀ in฀ any฀ branch฀of฀biology.฀ Biological฀ diversity฀ is฀ immense.฀There฀ are฀ estimated฀ to฀ be฀ more฀ than฀ 78,800฀flowering฀plant฀species฀in฀the฀Neotropics฀alone฀(Smith฀et฀ al.,฀2004).฀ Not฀only฀do฀these฀species฀need฀to฀be฀described,฀but฀this฀information฀needs฀to฀ be฀organised฀in฀such฀a฀way฀that฀it฀can฀be฀retrieved฀quickly฀and฀easily.฀This฀end฀ is฀served฀through฀classifying฀species฀into฀groups,฀and฀those฀groups฀into฀larger฀ groups.฀For฀example,฀the฀species฀Cremastosperma฀cauliflorum,฀as฀described฀by฀ Robert฀E.฀Fries฀(1931)฀belongs฀to฀the฀genus฀Cremastosperma.฀Cremastosperma,฀ along฀with฀genera฀such฀as฀Annona฀and฀Cananga,฀is฀classified฀within฀the฀family฀ Annonaceae,฀which฀in฀turn฀is฀grouped฀with฀families฀including฀Magnoliaceae฀ and฀Myristicaceae฀to฀comprise฀the฀order฀Magnoliales.฀ A฀number฀of฀studies฀have฀presented฀groupings฀of฀genera฀within฀Annonaceae฀ following฀different฀principles฀and฀using฀different฀sources฀of฀data.฀Cremastosperma฀ has฀been฀associated฀with฀a฀number฀of฀other฀mostly฀Neotropical฀genera,฀under฀ the฀names฀‘Malmea-tribe/group’฀(on฀the฀basis฀of฀pollen฀morphology;฀Walker,฀ 1971)฀and฀‘Cremastosperma-group’฀(flower฀morphology;฀Van฀Heusden,฀1992).฀ These฀ groupings฀ were฀ contradicted฀ to฀ a฀ greater฀ or฀ lesser฀ extent฀ by฀ works฀ based฀on฀gross฀morphology฀(Fries,฀1959)฀and฀fruit฀and฀seed฀morphology฀(Van฀ Setten฀&฀Koek-Noorman,฀1992),฀and฀by฀phenetic฀analyses฀of฀floral฀and฀fruit฀ morphological฀data฀(Koek-Noorman฀et฀ al.,฀1997).฀Any฀assumption฀that฀one฀ subset฀of฀characters฀(in฀itself฀an฀effectively฀arbitrary฀subdivision฀of฀the฀total฀ variation)฀is฀in฀some฀way฀a฀superior฀foundation฀for฀classification฀than฀others฀ requires฀ explicit฀ justification.฀This฀ is฀ particularly฀ the฀ case฀ where฀ different฀ partitions฀of฀characters฀appear฀to฀support฀different฀conclusions.฀However,฀the฀ same฀is฀true฀for฀the฀assumptions฀made฀in฀the฀interpretation฀of฀those฀characters.฀ 10 • Chapter 1 Conflicting฀ results฀ can฀ also฀ be฀ derived฀ from฀ exactly฀ the฀ same฀ data,฀ if฀ the฀ groupings฀are฀made฀following฀different฀guiding฀principles.฀ Monophyly and phylogeny reconstruction Whilst฀the฀only฀effective฀way฀of฀delimiting฀species฀may฀rest฀on฀the฀opinion฀ of฀an฀expert฀in฀a฀particular฀group,฀cladistic฀methodology฀(Hennig,฀1966)฀has฀ effectively฀ excluded฀ such฀ subjectivity฀ from฀ higher฀ level฀ classification.฀The฀ structure฀provided฀by฀evolutionary฀history฀can฀be฀used฀to฀make฀a฀system฀of฀ classification฀which฀not฀only฀best฀reflects฀the฀information฀we฀have฀for฀particular฀ species,฀but฀also฀allows฀us฀to฀predict฀characteristics฀that฀have฀yet฀to฀be฀observed.฀ Groups฀ are฀ only฀ recognised฀ where฀ they฀ include฀ all฀ of฀ the฀ descendents฀ of฀ a฀ common฀ ancestor,฀ and฀ no฀ others.฀These฀ are฀ identified฀ by฀ the฀ possession฀ of฀ shared฀derived฀character฀states.฀Such฀a฀group฀is฀described฀as฀monophyletic฀and฀ otherwise฀known฀as฀a฀clade.฀Characteristics฀can฀be฀lost฀by฀individual฀lineages฀ within฀a฀clade,฀and฀new฀characteristics฀can฀also฀be฀gained.฀Groups฀defined฀by฀ the฀absence฀of฀characters฀which฀are฀present฀in฀other฀groups,฀rather฀than฀on฀the฀ presence฀of฀characters฀inferred฀to฀have฀been฀acquired฀in฀a฀common฀ancestor,฀ can฀have฀more฀other฀characteristics฀in฀common฀with฀other฀groups฀than฀with฀ each฀other.฀The฀predictive฀value฀of฀such฀groups฀is฀correspondingly฀lower. In฀Annonaceae,฀the฀principle฀of฀monophyly฀has฀been฀applied฀on฀a฀limited฀ scale฀ and฀ led฀ to฀ some฀ re-classification฀ at฀ the฀ generic฀ level฀ (Chatrou฀ et฀ al.,฀ 2000;฀Mols,฀2004).฀The฀assumption฀behind฀all฀phylogenetic฀inference฀methods฀ is฀that฀the฀input฀data฀consist฀of฀a฀set฀of฀homologous฀characters฀(Sanderson฀&฀ Shaffer,฀2002).฀However,฀difficulties฀in฀homology฀assessment฀and฀high฀levels฀of฀ homoplasy฀in฀cladistic฀analyses฀using฀morphological฀data฀at฀higher฀taxonomic฀ levels฀in฀Annonaceae฀has฀yielded฀equivocal฀results฀(Doyle฀&฀Le฀Thomas,฀1996).฀ This฀has฀led฀to฀recent฀studies฀relying฀on฀DNA฀sequence฀data฀alone฀(Mols฀et฀al.,฀ 2004;฀Richardson฀et฀al.,฀2004)฀or฀in฀combination฀with฀morphology฀(Doyle฀et฀ al.,฀2000;฀Mols,฀2004).฀A฀number฀of฀chloroplast฀DNA฀(cpDNA)฀regions฀have฀ been฀exploited฀as฀sources฀of฀phylogenetically฀informative฀characters:฀in฀particular฀ the฀ widely฀ used฀ rbcL฀ gene฀ and฀ tr nL-F฀ region.฀ Uniparental฀ inheritance฀ and฀ effective฀lack฀of฀recombination฀means฀that฀although฀phylogenies฀inferred฀using฀ cpDNA฀sequence฀data฀may฀or฀may฀not฀reflect฀species฀phylogenies฀(Nichols,฀ 2001),฀they฀can฀in฀any฀case฀be฀regarded฀as฀gene฀trees฀for฀which฀the฀congruence฀ with฀organismal฀history฀becomes฀greater฀with฀increasing฀time฀scale฀(Clegg฀&฀ Zurawski,฀1992)฀(i.e.฀assuming฀that฀reticulation฀is฀less฀frequent฀between฀more฀ distantly฀related฀plant฀species).฀Given฀the฀levels฀of฀uncertainty฀and฀disagreement฀ in฀the฀interpretation฀of฀morphological฀variation฀in฀Annonaceae,฀it฀is฀not฀to฀ be฀expected฀that฀the฀reconstruction฀of฀phylogeny฀with฀DNA฀sequences฀will฀ �eneral introduction • 11 necessarily฀identify฀clades฀that฀are฀supported฀by฀all฀or฀large฀parts฀of฀the฀previous฀ evidence.฀They฀may฀not฀even฀be฀straightforward฀to฀diagnose.฀Without฀knowledge฀ of฀the฀organisms฀themselves,฀a฀phylogenetic฀tree฀based฀on฀DNA฀sequence฀data฀ serves฀little฀purpose.฀However,฀in฀circumstances฀such฀as฀these฀it฀is฀a฀particularly฀ powerful฀tool,฀providing฀a฀greater฀weight฀of฀evidence฀which฀can฀be฀used฀to฀ assess฀the฀homology฀and฀phylogenetic฀signal฀of฀morphological฀characters฀with฀ which฀no฀single฀supported฀result฀could฀otherwise฀be฀obtained. Phylogeny฀reconstruction฀in฀this฀thesis฀is฀performed฀using฀two฀different฀ inference฀ methods:฀ maximum฀ parsimony฀ (MP),฀ and฀ Bayesian฀ inference฀ (see฀ Chapter฀3).฀Support฀for฀clades฀is฀represented฀mainly฀by฀bootstrap฀percentages฀ (Felsenstein,฀ 1985)฀ under฀ MP,฀ and฀ posterior฀ probabilities฀ (Huelsenbeck฀ et฀ al.,฀ 2001)฀ under฀ Bayesian฀ analysis.฀ Bootstrap฀ percentages฀ are฀ regarded฀ as฀ conservative฀ estimations฀ of฀ support฀ (Hillis฀ &฀ Bull,฀ 1993).฀ Relatively฀ higher฀ posterior฀probabilities฀have฀been฀reported฀for฀clades฀recovered฀using฀Bayesian฀ inference฀ (e.g.฀ compared฀ with฀ maximum฀ likelihood฀ analysis;฀ Suzuki฀ et฀ al.,฀ 2002;฀and฀with฀parsimony฀analyses;฀Simmons฀&฀Miya,฀2004).฀Although฀in฀some฀ cases฀this฀has฀been฀shown฀to฀represent฀an฀efficient฀means฀of฀correctly฀resolving฀ relationships฀(Simmons฀&฀Miya,฀2004),฀it฀has฀led฀some฀workers฀to฀regard฀these฀ results฀ with฀ scepticism฀ (Suzuki฀ et฀ al.,฀ 2002).฀ Support฀ values฀ derived฀ under฀ different฀methods฀are฀impossible฀to฀compare฀directly.฀In฀the฀rare฀cases฀where฀ application฀of฀different฀methods฀appears฀to฀give฀strongly฀conflicting฀results฀(e.g.฀ Huelsenbeck,฀1997)฀this฀may฀or฀may฀not฀reflect฀a฀shortcoming฀in฀ones฀of฀the฀ techniques฀(see฀Siddall฀&฀Whiting,฀1999).฀Use฀of฀only฀one฀such฀method฀would฀ of฀course฀fail฀to฀reveal฀such฀a฀problem.฀However,฀in฀general฀I฀consider฀time฀to฀ be฀better฀invested฀in฀improving฀the฀data฀than฀spent฀interpreting฀phylogenies฀ that฀are฀either฀unsupported฀or฀not฀robust฀to฀the฀method฀applied. Biogeography The฀overall฀objective฀of฀this฀thesis฀could฀be฀summarised฀as฀to฀describe฀ the฀diversity฀of฀species฀found฀in฀the฀genus฀Cremastosper ma,฀and฀investigate฀ how฀that฀diversity฀might฀have฀originated.฀The฀distribution฀of฀the฀species฀of฀ Cremastosper ma฀ is฀ of฀ particular฀ interest:฀ the฀ highest฀ diversity฀ of฀ species฀ of฀ Cremastosper ma฀is฀found฀in฀areas฀surrounding฀the฀Andean฀mountain฀chain,฀ and฀ they฀ are฀ all฀ but฀ absent฀ from฀ central฀ Amazonia.฀This฀‘Andean-centred’฀ distribution฀is฀also฀found฀in฀a฀small฀number฀of฀other฀genera฀of฀Annonaceae.฀ That฀the฀constituent฀species฀of฀multiple฀disparate฀taxa฀have฀been฀collected฀in฀ particular฀areas฀has฀often฀invited฀the฀explanation฀of฀a฀common฀biogeographic฀ history.฀In฀the฀case฀of฀Andean-centred฀taxa฀the฀orogeny฀of฀the฀Andean฀mountain฀ chain฀has฀been฀suggested฀to฀be฀a฀factor฀driving฀their฀diversification฀(Gentry,฀ 12 • Chapter 1 1982).฀Most฀genera฀of฀Annonaceae฀are฀relatively฀more฀diverse฀in฀Amazonia.฀ Confirming฀monophyly฀in฀Cremastosperma฀is฀of฀primary฀importance฀in฀testing฀ this฀hypothesis.฀Amazonian-centred,฀rather฀than฀Andean-centred,฀distribution฀ patterns฀could฀be฀the฀result฀of฀some฀possible฀poly-฀or฀paraphyletic฀groupings฀ of฀species฀of฀Cremastosperma฀and฀related฀genera. A฀robust฀phylogeny฀makes฀a฀number฀of฀approaches฀to฀reconstructing฀the฀ biogeographic฀history฀of฀organisms฀possible.฀Ancestral฀distribution฀areas฀can฀ be฀inferred฀(e.g.฀Doyle฀&฀Le฀Thomas,฀1997;฀Doyle฀et฀al.,฀2004)฀and฀a฀cladistic฀ biogeographic฀ approach฀ used฀ to฀ identify฀ congruent฀ area฀ cladograms฀ (e.g.฀ Pennington฀et฀al.,฀2004).฀Recent฀developments฀in฀molecular฀dating฀have฀begun฀ to฀offer฀the฀possibility฀of฀being฀able฀to฀demonstrate฀when฀diversifications฀in฀ such฀ groups฀ actually฀ occurred฀ within฀ the฀ time฀ slices฀ of฀ such฀ biogeographic฀ hypotheses.฀For฀example,฀hypotheses฀corresponding฀to฀refuge฀theory฀(Haffer,฀ 1969),฀taking฀place฀within฀the฀time฀frame฀of฀Pleistocene฀climatic฀changes,฀can฀ be฀compared฀to฀those฀suggesting฀speciation฀to฀be฀concentrated฀earlier฀in฀the฀ Tertiary฀as฀suggested฀by฀other฀authors฀(e.g.฀Hooghiemstra฀&฀van฀der฀Hammen,฀ 1998). Molecular dating Molecular฀dating฀is฀subject฀to฀a฀number฀of฀sources฀of฀error฀which฀must฀ be฀taken฀into฀account฀when฀assessing฀the฀significance฀of฀the฀result฀(Sanderson฀ &฀ Doyle,฀ 2001).฀ Chloroplast฀ sequences฀ do฀ not฀ adhere฀ to฀ a฀ time฀ calibrated฀ molecular฀ clock฀ (Gaut,฀ 1998).฀ Rate฀ heterogeneity฀ can฀ be฀ addressed฀ using฀ techniques฀such฀as฀nonparametric฀rate฀smoothing฀(NPRS:฀Sanderson,฀1997),฀ penalized฀ likelihood฀ (PL:฀ Sanderson,฀ 2002),฀ or฀ a฀ Bayesian฀ approach฀ (e.g.฀ Thorne฀ et฀ al.,฀ 1998).฀These฀ methods฀ assume฀ autocorrelation฀ of฀ rates฀ across฀ a฀phylogeny฀(Sanderson฀et฀ al.,฀2004).฀Uncertainty฀can฀be฀associated฀with฀the฀ topology฀of฀the฀phylogeny฀in฀question฀(phylogenetic฀uncertainty),฀and฀in฀the฀ estimation฀of฀branch฀lengths฀optimised฀onto฀it฀given฀the฀character฀sampling฀ used฀(‘substitutional฀noise’)฀(Sanderson฀&฀Doyle,฀2001).฀Assessing฀the฀effects฀ of฀ the฀ latter฀ source฀ of฀ error฀ when฀ using฀ NPRS฀ and฀ PL฀ is฀ possible฀ through฀ bootstrap฀resampling฀techniques,฀and฀the฀effects฀of฀phylogenetic฀uncertainty฀ can฀be฀avoided฀by฀limiting฀dating฀to฀significantly฀supported฀nodes. Absolute,฀as฀opposed฀to฀relative,฀ages฀can฀only฀be฀estimated฀by฀calibration฀ using฀fossil฀or฀geological฀data฀to฀set฀or฀limit฀the฀age฀of฀one฀or฀more฀nodes.฀ This฀process฀is฀controversial฀(see฀Shields,฀2004),฀often฀afflicted฀by฀considerable฀ uncertainty฀(Bremer,฀2000;฀Smith฀&฀Peterson,฀2002;฀Hedges฀&฀Kumar,฀2004).฀ Due฀to฀the฀incompleteness฀of฀the฀fossil฀record,฀age฀estimates฀produced฀using฀ fossil฀calibrations฀will฀arguably฀always฀be฀underestimations฀of฀true฀ages฀(Reisz฀ �eneral introduction • 13 &฀Muller,฀2004).฀More฀accurate฀results฀might฀be฀achieved฀if฀fossils฀could฀be฀ used฀ to฀ constrain฀ multiple฀ nodes฀ within฀ the฀ phylogeny฀ (Soltis฀ et฀ al.,฀ 2002;฀ Sanderson฀et฀al.,฀2004),฀the฀placement฀of฀which฀can฀be฀facilitated฀by฀assessments฀ of฀homology฀made฀possible฀by฀robust฀phylogenies.฀The฀oldest฀unambiguously฀ identifiable฀fossil฀Annonaceae฀remains฀have฀been฀found฀in฀the฀Maastrichtian฀of฀ Nigeria฀(seeds฀with฀perichalazal฀ring฀and฀ruminate฀endosperm;฀Chesters,฀1955)฀ and฀Colombia฀(reticulate฀monosulcate฀pollen;฀Sole฀de฀Porta,฀1971).฀The฀ages฀ of฀these฀fossils฀are฀not฀suitable฀to฀address฀the฀question฀of฀the฀age฀of฀the฀family฀ itself,฀but฀a฀number฀of฀recent฀studies฀have฀used฀molecular฀dating฀techniques,฀ deriving฀consistent฀age฀estimations฀for฀nodes฀within฀the฀Magnoliales฀(Doyle฀ et฀ al.,฀ 2004;฀ Richardson฀ et฀ al.,฀ 2004).฀ Richardson฀ et฀ al.฀ (2004),฀ largely฀ in฀ agreement฀with฀Doyle฀&฀Le฀Thomas฀(1997),฀suggested฀pantropical฀distributions฀ of฀ clades฀ within฀Annonaceae฀ to฀ be฀ due฀ to฀ interchange฀ between฀Africa฀ and฀ South฀America฀across฀the฀opening฀Atlantic฀ocean฀followed฀by฀dispersal฀across฀ Bering฀and/or฀North฀Atlantic฀land฀bridges,฀rather฀than฀the฀older฀explanation฀ of฀vicariance฀of฀Gondwana.฀ Overview The฀aims฀of฀Chapter฀2฀are฀to฀test฀monophyly฀of฀the฀genera฀Cremastosperma,฀ Mosannona,฀Duguetia,฀and฀Guatteria฀by฀phylogenetic฀analysis฀of฀rbcL,฀tr nLF฀ and฀ matK฀ cpDNA฀ data.฀ Comparisons฀ of฀ relative฀ timing฀ and฀ preliminary฀ estimates฀for฀the฀absolute฀ages฀of฀their฀most฀recent฀common฀ancestors฀(MRCAs)฀ are฀made,฀facilitated฀by฀the฀inclusion฀of฀all฀four฀groups฀in฀a฀single฀analysis฀using฀ NPRS฀(Sanderson,฀1997).฀These฀estimations฀are฀used฀to฀investigate฀whether฀ diversification฀within฀these฀contrasting฀genera฀can฀be฀shown฀to฀have฀occurred฀ within฀the฀same฀or฀significantly฀different฀time฀slices.฀An฀important฀consideration฀ in฀working฀with฀Cremastosperma,฀Mosannona,฀Duguetia,฀and฀Guatteria฀is฀the฀ possible฀source฀of฀error฀represented฀by฀the฀large฀disparity฀in฀numbers฀of฀taxa฀ available฀for฀analysis.฀In฀order฀to฀assess฀whether฀and฀to฀what฀extent฀the฀outcome฀ of฀age฀estimations฀might฀be฀influenced฀not฀only฀by฀character฀(as฀above)฀but฀also฀ by฀taxon฀sampling,฀re-sampling฀approaches฀are฀used฀and฀the฀results฀compared. In฀Chapter฀3฀Gentry’s฀(1982)฀explanation฀for฀the฀distribution฀of฀Andeancentred฀taxa฀is฀tested.฀That฀was฀that฀they฀1)฀originated฀in฀South฀America฀(i.e.฀by฀ vicariance฀of฀Gondwana)฀and฀2)฀were฀subject฀to฀a฀recent฀burst฀of฀speciation฀as฀a฀ result฀of฀the฀orogeny฀of฀the฀Andean฀mountain฀chain.฀Phylogeny฀reconstruction฀ using฀eight฀cpDNA฀markers฀(psbA-trnH,฀ndhF,฀trnT-L,฀trnS-G฀and฀atpb-rbcL,฀ as฀well฀as฀tr nL-F,฀rbcL฀and฀matK)฀and฀different฀molecular฀dating฀techniques฀ (NPRS,฀PL,฀Bayesian)฀are฀applied฀in฀order฀to: 1)฀Determine฀ the฀ geographic฀ origin฀ of฀ the฀ Andean-centred฀ genera฀ 14 • Chapter 1 Cremastosper ma,฀ Klarobelia,฀ Malmea,฀ and฀ Mosannona:฀ identify฀ the฀ sister฀ groups฀ of฀ the฀ four฀ genera฀ (whether฀ these฀ are฀ other฀ Neotropical,฀ or฀ Asian฀ clades),฀and฀the฀age฀of฀these฀sister฀groups฀(whether฀their฀distribution฀could฀ have฀originated฀by฀vicariance฀of฀Gondwana,฀dispersal฀from฀Africa฀or฀dispersal฀ across฀the฀Boreotropics).฀฀ 2)฀Determine฀whether฀the฀four฀genera฀may฀have฀congruent฀evolutionary฀ histories฀that฀may฀all฀have฀been฀affected฀by฀the฀Andean฀orogeny:฀testing฀the฀ monophyly฀ of฀ each฀ of฀ these฀ genera฀ and฀ determining฀ the฀ age฀ of฀ the฀ crown฀ group฀of฀each฀genus.฀ In฀ Chapter฀ 4฀ conflicting฀ signal฀ is฀ investigated฀ and฀ the฀ hypothesis฀ of฀ a฀ paralogy฀in฀the฀trnL-F฀regions฀in฀Annonaceae฀is฀tested,฀using฀both฀PCR-based฀ and฀ phylogenetic฀ analysis฀ techniques.฀ Further฀ conclusions฀ are฀ drawn฀ with฀ respect฀to฀the฀timing฀of฀duplication฀and฀to฀the฀phylogenetic฀signal฀contained฀ in฀ both฀ copies.฀To฀ address฀ the฀ question฀ of฀ functional฀ homology฀ in฀ the฀ two฀ copies,฀comparisons฀are฀drawn฀between฀sequences฀obtained฀in฀this฀study฀and฀ proposed฀secondary฀structures฀and฀corresponding฀functional฀constraints฀in฀the฀ tr nL฀gene฀and฀Group฀I฀intron฀in฀land฀plants฀(Borsch฀et฀ al.,฀2003;฀Quandt฀et฀ al.,฀2004).฀Positional฀homology฀(i.e.฀within฀or฀possibly฀without฀the฀chloroplast฀ genome)฀is฀less฀straightforward฀to฀determine฀from฀sequences฀alone.฀A฀number฀ of฀potential฀directions฀for฀future฀research฀are฀therefore฀also฀discussed. In฀Chapter฀5฀a฀taxonomic฀revision฀of฀the฀genus฀Cremastosperma฀is฀presented.฀ All฀ 29฀ recognised฀ species฀ are฀ treated.฀ ฀ An฀ identification฀ key,฀ descriptions,฀ illustrations,฀and฀a฀list฀of฀exsicatae฀is฀provided.฀Results฀of฀phylogenetic฀analysis฀ using฀ multiple฀ chloroplast฀ markers฀ plus฀ the฀ paralogue฀ of฀ the฀ tr nL-F฀ region฀ discovered฀in฀Chapter฀4฀are฀briefly฀discussed.฀ �eneral introduction • 15 16 • Chapter 1 Chapter 2 Phylogeny reconstruction and molecular dating in four Neotropical genera of Annonaceae: the effect of taxon sampling in age estimations M i c h a e l D . P i r i e 1 , L a r s W. C h a t r o u 1 , R o y H . J . E r k e n s 1 , J a n W. M a a s 1 , T i m o t h e ü s v a n d e r N i e t 2 , J o h a n B . M o l s 3 a n d James E. Richardson4 Published in Regnum Vegetabile 143, A.R.G. Gantner Verlag, Liechtenstein, pp.149-174 1 Nat i onaal Her bar i um Neder l and, Uni ver s i t ei t Ut r ec h t br anc h, Hei del ber gl aan 2, 2 I ns t i t ut e of Sys t em at i c Bot any, Zol l i ker s t r as s e 107, CH- 8008 Zür i c h, Sw i t zer l and 3 Nationaal Herbarium Nederland, Universiteit Leiden branch 4 Nat i onaal Her bar i um Neder l and, Wageni ngen Uni ver s i t ei t br anc h, Gener aal 3584 CS Ut r ec ht , The Net her l ands P. � . B o x 9 5 1 4 , 2 3 0 0 R A L e i d e n , T h e N e t h e r l a n d s Foul kes w eg 37, 6703 BL Wageni ngen, The Net her l ands �olecular dating in Annonaceae • 17 Abstract Preliminary฀results฀are฀presented฀comparing฀the฀timing฀of฀diversification฀ in฀four฀predominantly฀Neotropical฀genera฀of฀Annonaceae;฀Cremastosper ma,฀ Duguetia,฀ Guatter ia,฀ and฀ Mosannona.฀With฀ the฀ exception฀ of฀ a฀ few฀ basal฀ lineages,฀the฀majority฀of฀the฀ca.฀2500฀species฀of฀the฀Annonaceae฀diverge฀into฀ two฀major฀sister฀clades.฀One฀comprises฀relatively฀few,฀larger,฀genera฀(including฀ Duguetia฀and฀Guatter ia)฀representing฀roughly฀twice฀as฀many฀species฀in฀total฀ and฀with฀an฀apparent฀rate฀of฀molecular฀divergence฀(revealed฀by฀branch฀lengths฀ in฀most฀parsimonious฀trees)฀around฀three฀times฀as฀high฀as฀the฀other฀(including฀ Cremastosperma฀and฀Mosannona),฀in฀which฀more฀genera฀each฀comprise฀fewer฀ species.฀ Explanations฀ for฀ the฀ disparity฀ in฀ numbers฀ of฀ species฀ in฀ these฀ four฀ genera฀are฀sought฀by฀use฀of฀phylogeny฀reconstruction฀and฀molecular฀dating฀ techniques฀(using฀nonparametric฀rate฀smoothing)฀aiming฀to฀assess฀monophyly฀ and฀ arrive฀ at฀ preliminary฀ estimates฀ of฀ the฀ relative฀ ages฀ of฀ their฀ most฀ recent฀ common฀ancestors฀(MRCAs).฀The฀effects,฀in฀particular,฀of฀taxon฀and฀character฀ sampling฀on฀date฀estimates฀in฀these฀genera฀is฀assessed฀and฀compared.฀Results฀ in฀the฀species฀rich฀genus฀Guatteria฀show฀higher฀sampling฀of฀crown฀group฀taxa฀ resulting฀in฀significantly฀older฀age฀estimation฀for฀the฀MRCA. Keywords:฀ Annonaceae,฀ Cremastosperma,฀ Mosannona,฀ Duguetia,฀ Guatteria,฀ phylogeny฀ reconstruction,฀molecular฀dating,฀taxon฀sampling. Abbreviations:฀BS฀(bootstrap฀support);฀LBC฀(long฀branch฀clade);฀LTT฀(lineages฀through฀ time);฀ ML฀ (maximum฀ likelihood);฀ MP฀ (maximum฀ parsimony);฀ MRCA฀ (most฀ recent฀ common฀ancestor);฀NPRS฀(nonparametric฀rate฀smoothing);฀SBC฀(short฀branch฀clade). Introduction Annonaceae in the Neotropics Annonaceae฀comprises฀around฀2500฀species฀of฀trees฀and฀lianas฀in฀ca.130฀ genera฀distributed฀pan-tropically,฀predominantly฀in฀tropical฀rain฀forests.฀Over฀ 900฀species฀are฀found฀in฀the฀Neotropics฀(Chatrou฀et฀ al.,฀2004),฀where฀they฀ represent฀ a฀ significant฀ part฀ of฀ plant฀ diversity,฀ in฀ terms฀ both฀ of฀ numbers฀ of฀ species฀and฀of฀individuals.฀Studies฀on฀ α-diversity฀of฀Amazonian฀forests฀rank฀ Annonaceae฀among฀the฀most฀abundant฀families฀(e.g.฀Valencia฀et฀ al.,฀1994;฀Ter฀ Steege฀et฀ al.,฀2000).฀Relative฀abundance฀of฀Annonaceae฀is฀high฀in฀areas฀that฀ have฀ been฀ designated฀ biodiversity฀ hotspots฀ by฀ Myers฀ et฀ al.฀ (2000)฀ such฀ as฀ the฀Chocó/Darién/Western฀Ecuador฀region,฀the฀tropical฀Andes,฀and฀Brazil’s฀ Atlantic฀forest฀(Davis฀et฀al.,฀1997). Neotropical฀ representatives฀ of฀Annonaceae฀ have฀ been฀ a฀ major฀ focus฀ of฀ systematic฀research฀at฀the฀Utrecht฀branch฀of฀the฀National฀Herbarium฀of฀the฀ 18 • Chapter 2 Netherlands฀ since฀ the฀ founding฀ of฀ the฀ international฀Annonaceae฀ project฀ in฀ 1983.฀This฀ has฀ resulted฀ in฀ taxonomic฀ revisions฀ of฀ a฀ large฀ proportion฀ of฀ the฀ Neotropical฀ genera฀ (e.g.฀ Maas฀ &฀Westra,฀ 1984;฀ 1985;฀ 1992;฀ Murray,฀ 1993;฀ Johnson฀&฀Murray,฀1995;฀Chatrou,฀1998;฀Maas฀&฀Westra,฀2003;฀Maas฀ et฀ al.,฀ 2003),฀with฀the฀remaining฀unrevised฀Neotropical฀genera฀almost฀all฀the฀subjects฀ of฀ current฀ work.฀ In฀ parallel฀ to฀ alpha฀ taxonomic฀ treatment฀ of฀Annonaceae,฀ efforts฀to฀reconstruct฀the฀phylogeny฀of฀the฀family฀based฀on฀morphology฀(Doyle฀ &฀Le฀Thomas,฀1996)฀and฀DNA฀sequence฀data฀(Bygrave,฀2000;฀Doyle฀ et฀ al.,฀ 2000;฀Mols฀et฀al.,฀2004;฀Richardson฀et฀al.,฀2004)฀are฀providing฀an฀increasingly฀ robust฀framework฀for฀the฀interpretation฀of฀the฀diversity฀being฀described.฀The฀ most฀ recent฀ data฀ gathered฀ to฀ reconstruct฀ the฀ family฀ phylogeny฀ (Chatrou฀ et฀ al.,฀ unpublished฀ data)฀ include฀ rbcL฀ and฀ tr nL-F฀ sequences฀ for฀ 220฀ species,฀ representing฀over฀80฀genera,฀sampled฀from฀all฀continents. Phylogeny reconstruction in Annonaceae Annonaceae฀ is฀ classified฀ within฀ the฀ order฀ Magnoliales฀ (APG฀ II,฀ 2003),฀ wherein฀ it฀ is฀ highly฀ supported฀ as฀ sister฀ group฀ to฀ Eupomatiaceae,฀ based฀ on฀ cladistic฀analysis฀of฀morphological฀and฀molecular฀data฀(Sauquet฀et฀ al.,฀2003),฀ with฀ Anaxagorea฀ resolved฀ as฀ sister฀ to฀ the฀ rest฀ of฀ the฀ family฀ (Doyle฀ et฀ al.,฀ 2000;฀ Sauquet฀ et฀ al.,฀ 2003).฀The฀ further฀ use฀ of฀ morphology฀ for฀ phylogeny฀ reconstruction฀ in฀ Annonaceae฀ has฀ been฀ problematic฀ due฀ to฀ difficulties฀ in฀ homology฀assessment฀and฀high฀levels฀of฀homoplasy฀(Doyle฀&฀Le฀Thomas,฀1996).฀ Results฀of฀phylogenetic฀research฀based฀on฀rbcL฀and฀trnL-F฀sequence฀data฀(Mols฀ et฀al.,฀2004;฀Richardson฀et฀al.,฀2004)฀support฀the฀position฀of฀Anaxagorea฀and฀ further฀divide฀the฀rest฀of฀Annonaceae฀between฀a฀small฀clade฀including฀Cananga฀ and฀Cleistopholis,฀sister฀group฀to฀a฀large฀clade฀including฀the฀majority฀of฀species฀ of฀the฀family฀(see฀Fig.฀1). This฀large฀clade฀is฀further฀divided฀into฀two฀sub-clades.฀Clade฀A฀includes฀ roughly฀1,500฀species,฀all฀characterised฀by฀an฀inaperturate฀pollen฀condition,฀ almost฀a฀thousand฀of฀which฀are฀divided฀between฀7฀species-rich฀genera:฀Annona,฀ Ar tabotrys,฀ Duguetia,฀ Goniothalamus,฀ Guatter ia,฀ Uvar ia,฀ and฀ Xylopia.฀The฀ remaining฀ 500฀ species฀ are฀ currently฀ assigned฀ to฀ around฀ 40฀ smaller฀ genera.฀ Two฀ specialised฀ morphological฀ syndromes฀ within฀ the฀ family฀ are฀ confined฀ exclusively฀to฀this฀clade:฀fused฀fruiting฀carpels,฀and฀climbing฀habit.฀The฀second฀ sub-clade฀(B)฀is฀predominantly฀Asian,฀but฀also฀including฀all฀the฀Neotropical฀ genera฀ with฀ imbricate฀ sepals,฀ one฀ basal฀ ovule,฀ and฀ apocarpous฀ fruits฀ (such฀ as฀ Malmea,฀ Oxandra,฀ Ephedranthus,฀ Cremastosper ma,฀ and฀ Mosannona).฀ It฀ comprises฀around฀700฀species฀(less฀than฀half฀the฀total฀number฀estimated฀for฀ clade฀A),฀divided฀between฀around฀50฀genera,฀few฀of฀which฀represent฀more฀than฀ 40฀species.฀Pollen฀types฀in฀this฀clade฀have฀been฀characterised฀as฀monosulcate฀or฀ disulculate,฀however,฀recent฀results฀suggest฀this฀pattern฀to฀be฀less฀straightforward,฀ i.e.฀with฀cryptic฀apertures฀and฀inaperturate฀compound฀pollen฀forms฀observed฀ �olecular dating in Annonaceae • 19 ��aro�e��a 62 88 70 �osa��o�a 98 B 95 Cremastosperma 89 100 87 100 100 90 A 100 100 ����et�a F�saea ��atter�a 100 Ca�a��a��C�e�stop�o��s etc� 100 ��a�a�orea �utgroups F i g . 1 . 50% majority rule bootstrap consensus of Annonaceae (with selected bootstrap frequencies indicated) using Magnoliales and Lauraceae representatives as outgroups, based on phylogenetic analysis of rbcL and trnL-F DNA sequences (adapted from Richardson et al., 2004). Genera emphasised in this study are indicated, A and B denote long and short branch clades, respectively. 20 • Chapter 2 in฀some฀groups฀(Mols฀et฀ al.,฀submitted).฀Phylogenetic฀analysis฀of฀both฀ rbcL฀ and฀trnL-F฀plastid฀DNA฀sequences฀reveals฀a฀large฀difference฀in฀branch฀lengths฀ between฀clades฀A฀and฀B.฀The฀mean฀branch฀length฀from฀the฀common฀ancestor฀ to฀the฀terminals฀is฀twice฀as฀long฀in฀the฀former฀than฀in฀the฀latter฀(Richardson฀et฀ al.,฀2004),฀hence฀the฀informal฀names฀‘long฀branch฀clade’฀(A)฀and฀‘short฀branch฀ clade’฀(B)฀used฀here.฀Though฀such฀a฀phenomenon฀might฀be฀symptomatic฀of฀ inappropriate฀rooting,฀the฀results฀of฀multiple฀independent฀analyses฀(see฀above)฀ which฀agree฀on฀Anaxagorea฀as฀sister฀to฀the฀rest฀of฀Annonaceae฀appear฀to฀refute฀ this฀explanation.฀ The฀availability฀of฀a฀robust฀and฀well-sampled฀phylogeny฀facilitates฀many฀ approaches฀to฀the฀study฀of฀plant฀evolution.฀The฀pan-tropical฀distribution฀of฀ the฀Annonaceae฀enables฀the฀study฀of฀biogeographic฀history฀on฀a฀global฀scale฀ (Richardson฀ et฀ al.,฀ 2004).฀ In฀ addition,฀ increasingly฀ dense฀ sampling฀ of฀ taxa฀ allows฀historical฀inference฀at฀more฀local฀scales฀and฀recent฀evolutionary฀time฀ frames,฀such฀as฀the฀biogeography฀of฀the฀Central฀and฀South฀American฀tropics.฀It฀ further฀enables฀one฀to฀address฀other฀biological฀questions฀such฀as฀which฀factors฀ may฀be฀important฀in฀driving฀speciation฀and฀the฀evolution฀of฀morphological฀ traits.฀In฀this฀chapter฀we฀concentrate฀on฀four฀Neotropical฀genera฀currently฀the฀ subject฀of฀species-level฀phylogeny฀reconstruction฀projects,฀Cremastosperma฀and฀ Mosannona฀of฀the฀short฀branch฀clade,฀and฀Duguetia฀and฀Guatteria฀of฀the฀long฀ branch฀clade฀(Fig.฀2,฀see฀Appendix฀B).฀We฀aim฀both฀to฀test฀monophyly฀of฀these฀ genera฀and฀to฀arrive฀at฀preliminary฀estimates฀for฀the฀ages฀of฀their฀most฀recent฀ common฀ancestors฀(MRCAs).฀This฀will฀enable฀more฀meaningful฀comparisons฀ to฀be฀made฀of฀the฀contrasting฀species฀richness฀and฀apparent฀rates฀of฀molecular฀ evolution฀in฀these฀clades,฀which฀may฀in฀turn฀contribute฀to฀our฀understanding฀ of฀ the฀ contrasting฀ evolutionary฀ trends฀ that฀ they฀ appear฀ to฀ represent฀ within฀ Annonaceae. Species level systematics in four Neotropical genera of Annonaceae Cremastosperma฀can฀be฀distinguished฀from฀other฀Neotropical฀Annonaceae฀ by฀the฀presence฀of฀a฀distinctive฀groove฀running฀down฀the฀raised฀midrib,฀the฀ parallel฀ organisation฀ of฀ the฀ tertiary฀ veins,฀ and฀ the฀ roof-like฀ structure฀ of฀ the฀ expanded฀ connective฀ above฀ the฀ thecae.฀The฀ genus฀ currently฀ comprises฀ 17฀ recognised฀species,฀though฀the฀completion฀of฀taxonomic฀revision฀will฀result฀ in฀ approximately฀ 35฀ species฀ in฀ total฀ of฀‘Andean฀ centred’฀ distribution,฀ sensu฀ Gentry฀ (1982)฀ (this฀ thesis,฀ Chapter฀ 5 * ).฀The฀ bulk฀ of฀ collections฀ have฀ been฀ made฀in฀lowland฀to฀premontane฀forest฀along฀the฀Andes฀in฀Peru฀and฀Ecuador,฀ with฀significant฀diversity฀extending฀north฀through฀Colombia฀into฀Panama฀and฀ * An overestimation of the number of species of Cremastosperma was made at the time this paper was accepted for publication. Twenty-nine, plus two insufficiently known, are treated in Chapter 5. �olecular dating in Annonaceae • 21 a b F i g . 3 a - d . Distribution maps: Cremastosperma (a), Mosannona (b), Duguetia (c), and Guatteria (d) (collections denoted by grey dots, collections sampled for DNA sequence data denoted by black stars) Costa฀Rica฀(see฀Fig.฀3a).฀Only฀three฀species*฀have฀been฀collected฀from฀further฀ east:฀C.฀ macrocar pum฀ in฀coastal฀Venezuela,฀C.฀ brevipes฀in฀French฀Guiana฀and฀ C.฀ monosper mum฀widespread฀from฀Peru฀across฀northern฀Bolivia฀and฀central฀ * A fourth, C. venezuelanum Pirie, has since been published (Chatrou & Pirie, 2005) 22 • Chapter 2 c d Brazil.฀Cladistic฀analysis฀of฀rbcL฀and฀tr nL-F฀sequences฀does฀not฀resolve฀the฀ sister฀group฀of฀Cremastosperma฀(Richardson฀et฀al.,฀2004),฀for฀which฀a฀number฀ of฀clades฀of฀Neotropical฀Annonaceae฀may฀be฀implicated,฀notably฀the฀isolated฀ genera฀Malmea฀and฀Pseudoxandra. Mosannona฀ comprises฀ 15฀ species฀ excluded฀ from฀ the฀ genus฀ Malmea฀ (Chatrou,฀ 1998),฀ differing฀ from฀ those฀ of฀ other฀ Annonaceae฀ genera฀ by฀ the฀ combination฀of฀a฀raised฀midrib฀and฀terminal฀inflorescences฀(Chatrou฀et฀ al.,฀ 2004).฀The฀distribution฀areas฀of฀species฀of฀Mosannona฀(see฀Fig.฀3b)฀are฀small,฀ �olecular dating in Annonaceae • 23 with฀little฀overlap฀between฀areas,฀the฀pattern฀of฀distribution฀closely฀resembling฀ that฀ of฀ the฀ species฀ of฀ Cremastosper ma.฀The฀ majority฀ of฀ Mosannona฀ species฀ are฀found฀in฀the฀upper฀Amazonian฀forests฀of฀Ecuador฀and฀Peru,฀several฀species฀ occur฀in฀Mesoamerica,฀distributed฀as฀far฀north-west฀as฀the฀Mexican฀states฀of฀ Veracruz฀and฀Nayarit,฀and฀M.฀discolor฀has฀been฀collected฀in฀two฀disjunct฀areas฀in฀ Guyana฀and฀Suriname.฀The฀Mosannona฀clade฀receives฀significant฀support฀(88%฀ bootstrap,฀see฀Fig.฀2)฀on฀the฀basis฀of฀rbcL฀and฀tr nL-F฀sequence฀comparison฀ as฀ sister฀ group฀ to฀ a฀ larger฀ clade฀ containing฀ genera฀ including฀ Oxandra฀ and฀ Ephedranthus฀and฀recently฀described฀genera฀formerly฀representing฀species฀of฀ Malmea:฀Klarobelia฀and฀Pseudomalmea฀(Richardson฀et฀al.,฀2004).฀ Duguetia฀occurs฀in฀tropical฀West฀and฀Central฀Africa฀as฀well฀as฀in฀the฀New฀ World,฀though฀92฀out฀of฀96฀species฀are฀Neotropical,฀distributed฀evenly฀across฀ the฀South฀American฀tropics฀and฀in฀Central฀America฀north฀into฀Costa฀Rica฀(see฀ Fig.฀3c).฀Duguetia฀is฀clearly฀distinguishable฀from฀other฀genera฀by฀the฀covering฀ of฀stellate฀and/or฀lepidote฀hairs,฀and฀the฀so-called฀pseudosyncarpous฀fruits:฀the฀ stipeless฀carpels฀are฀held฀by฀shallow฀depressions฀in฀the฀fruiting฀receptacle,฀and฀ in฀addition฀show฀various฀degrees฀of฀fusion฀of฀the฀carpels฀among฀the฀different฀ species.฀A฀thorough฀treatment฀of฀the฀genus฀was฀assembled฀in฀the฀monograph฀ by฀ Maas฀ et฀ al.฀ (2003),฀ and฀ phylogenetic฀ analysis฀ using฀ morphological฀ and฀ anatomical฀ characters฀ revealed฀ a฀ clade฀ comprising฀ the฀ Neotropical฀ genera฀ Fusaea฀ and฀ Duc keanthus฀ and฀African฀ Letestudoxa฀ and฀ Pseudar tabotrys฀ as฀ sister฀group฀to฀Duguetia฀(Chatrou฀et฀al.,฀2000).฀DNA฀sequence฀data,฀however,฀ indicate฀Fusaea฀alone฀is฀sister฀to฀Duguetia฀(Richardson฀et฀al.,฀2004).฀ Comprising฀ more฀ than฀ 250฀ species,฀ Guatter ia฀ is฀ the฀ most฀ species฀ rich฀ genus฀ in฀ Annonaceae,฀ with฀ the฀ only฀ exception฀ of฀ the฀ highly฀ polyphyletic฀ old฀world฀genus฀Polyalthia฀(Mols฀et฀ al.,฀2004).฀It฀is฀distributed฀throughout฀ the฀Neotropics฀(see฀Fig.฀3d)฀where฀it฀is฀common฀in฀wet฀forest฀areas฀across฀a฀ relatively฀wide฀range฀of฀elevation฀from฀sea฀level฀to฀around฀2,000฀m.฀Guatteria฀ resembles฀Neotropical฀short฀branch฀clade฀genera฀such฀as฀Cremastosper ma,฀in฀ that฀it฀also฀produces฀single฀seeded฀apocarpous฀fruits฀from฀axillary฀flowers.฀The฀ Guatter ia฀ clade฀ is฀ subtended฀ by฀ a฀ long฀ branch฀ leading฀ from฀ a฀ polytomy฀ in฀ the฀ basal฀ portion฀ of฀ the฀ long฀ branch฀ clade฀ (see฀ Fig.฀ 1),฀ leaving฀ the฀ identity฀ of฀ the฀ sister฀ group฀ to฀ Guatter ia฀ ambiguous.฀The฀ last฀ revision฀ of฀ the฀ genus฀ dates฀back฀to฀Fries’฀treatment฀(1939)฀and฀to฀date฀Guatteria฀is฀the฀only฀major฀ genus฀ of฀ Neotropical฀Annonaceae฀ awaiting฀ revision,฀ a฀ task฀ impeded฀ by฀ low฀ morphological฀and฀anatomical฀diversity฀within฀the฀genus. Preliminary฀results฀of฀phylogenetic฀analyses฀with฀higher฀levels฀of฀taxon฀ sampling฀in฀the฀four฀genera฀showed฀that฀the฀combination฀of฀rbcL฀and฀tr nLF฀ DNA฀ sequences฀ did฀ not฀ yield฀ sufficient฀ resolution.฀The฀ large฀ number฀ of฀ species฀required฀in฀order฀to฀meaningfully฀represent฀Duguetia฀and฀Guatter ia฀ in฀ a฀ phylogenetic฀ analysis฀ results฀ in฀ an฀ increased฀ demand฀ for฀ informative฀ characters฀ (Bremer฀ et฀ al.,฀ 1999).฀ For฀ Cremastosper ma฀ and฀ Mosannona,฀ 24 • Chapter 2 although฀comprising฀fewer฀species,฀the฀proportion฀of฀informative฀characters฀ to฀taxa฀in฀rbcL฀/฀tr nL-F฀matrices฀produced฀was฀nevertheless฀low.฀In฀general,฀ complete฀sampling฀of฀taxa฀for฀plant฀species฀level฀molecular฀phylogenies฀remains฀ challenging฀when฀working฀with฀tropical฀groups.฀For฀instance,฀in฀Annonaceae฀ relatively฀few฀of฀the฀often฀large฀numbers฀of฀species฀are฀in฀cultivation,฀and฀of฀ the฀ rest,฀ many฀ are฀ rarely฀ collected฀ in฀ the฀ wild.฀ Extraction฀ from฀ herbarium฀ material,฀though฀often฀successful,฀presents฀challenges฀for฀the฀amplification฀of฀ target฀regions฀(Savolainen฀et฀ al.,฀1995).฀Low฀quality฀of฀many฀of฀the฀available฀ DNA฀samples฀prevented฀large-scale฀generation฀of฀AFLP TM฀data.฀Currently฀no฀ protocol฀is฀available฀for฀amplifying฀nuclear฀encoded฀marker฀regions฀across฀the฀ Annonaceae.฀We฀therefore฀decided฀to฀sample฀additional฀characters฀from฀the฀ plastid฀genome,฀i.e.฀the฀gene฀matK฀(encoding฀a฀maturase).฀This฀region฀has฀been฀ shown฀to฀be฀highly฀effective฀in฀reconstructing฀phylogeny฀in฀angiosperms฀(Hilu฀ et฀al.,฀2003)฀and฀universal฀primers฀are฀available฀(Cuénoud฀et฀al.,฀2002). Molecular dating: a tool for testing hypotheses in Neotropical Annonaceae Comparison฀of฀taxa฀classified฀at฀the฀generic฀level฀becomes฀meaningful฀if฀we฀ have฀an฀idea฀of฀the฀time฀frame฀within฀which฀they฀diversified.฀The฀larger฀number฀ of฀species฀in฀the฀genus฀Guatteria฀might฀simply฀indicate฀diversification฀over฀a฀ longer฀period฀of฀time.฀Molecular฀dating฀provides฀a฀possibility฀to฀test฀hypotheses฀ for฀which฀there฀is฀an฀implicit฀time฀frame.฀For฀instance,฀biogeographic฀hypotheses฀ linking฀diversifications฀in฀groups฀such฀as฀Cremastosper ma฀in฀the฀Neotropics฀ to฀relatively฀recent฀events฀such฀as฀the฀Andean฀orogeny฀(Gentry,฀1982)฀can฀be฀ tested,฀as฀can฀those฀seeking฀to฀explain฀cross-continentally฀distributed฀tropical฀ groups฀such฀as฀Duguetia฀according฀to฀either฀long฀distance฀dispersal,฀Tertiary฀ migration฀or฀older฀Gondwanan฀distributions฀(e.g.,฀Renner฀et฀ al.,฀2001;฀Davis฀ et฀al.,฀2002).฀Even฀the฀relative฀contributions฀of฀the฀processes฀of฀speciation฀and฀ extinction฀in฀producing฀such฀patterns฀can฀be฀assessed฀(Barraclough฀&฀Reeves,฀ 2005). Molecular฀dating฀techniques฀are฀subject฀to฀a฀number฀of฀sources฀of฀error฀ which฀ must฀ be฀ taken฀ into฀ account฀ when฀ assessing฀ the฀ significance฀ of฀ the฀ result฀(Sanderson฀&฀Doyle,฀2001).฀Rate฀heterogeneity฀can฀be฀addressed฀using฀ techniques฀such฀as฀nonparametric฀rate฀smoothing฀(Sanderson,฀1997),฀which฀ assumes฀autocorrelation฀of฀rates฀across฀a฀phylogeny,฀and฀which฀can฀incorporate฀ assumption฀ of฀ a฀ gamma฀ distribution฀ of฀ rates฀ across฀ characters.฀ Absolute,฀ as฀ opposed฀ to฀ relative,฀ ages฀ can฀ only฀ be฀ estimated฀ given฀ a฀ (rate-smoothed)฀ ultrametric฀tree฀calibrated฀using฀fossil฀or฀geological฀data฀to฀set฀or฀limit฀the฀age฀of฀ one฀or฀more฀nodes.฀This฀process฀is฀controversial,฀often฀afflicted฀by฀considerable฀ uncertainty฀(Bremer,฀2000;฀Smith฀&฀Peterson,฀2002;฀Hedges฀&฀Kumar,฀2004).฀ Further฀uncertainty฀can฀be฀associated฀with฀the฀topology฀of฀the฀phylogeny฀in฀ question฀(phylogenetic฀uncertainty),฀and฀in฀the฀estimation฀of฀branch฀lengths฀ �olecular dating in Annonaceae • 25 optimised฀ onto฀ it฀ given฀ the฀ character฀ sampling฀ used฀ (‘substitutional฀ noise’)฀ (Sanderson฀&฀Doyle,฀2001).฀Assessing฀the฀effects฀of฀the฀latter฀source฀of฀error฀ is฀possible฀through฀bootstrap฀re-sampling฀techniques,฀whereas฀the฀effects฀of฀ phylogenetic฀ uncertainty฀ can฀ be฀ avoided฀ by฀ limiting฀ dating฀ to฀ significantly฀ supported฀nodes. Error฀ stemming฀ from฀ taxon฀ sampling฀ in฀ the฀ estimation฀ of฀ ages฀ for฀ particular฀ nodes฀ is฀ less฀ straightforward฀ to฀ determine,฀ as฀ the฀ taxon฀ sampling฀ itself฀may฀define฀the฀node฀in฀question.฀For฀example,฀the฀first฀branching฀lineage฀ represented฀by฀Amborella฀effectively฀defines฀the฀age฀of฀the฀crown฀group฀of฀ Angiosperms.฀Given฀that฀one฀has฀identified฀and฀included฀such฀a฀lineage฀in฀the฀ analysis,฀ methods฀ of฀ molecular฀ dating฀ should฀ provide฀ consistent฀ estimations฀ of฀the฀age฀of฀the฀MRCA฀of฀the฀crown฀group,฀irrespective฀of฀levels฀of฀taxon฀ sampling.฀This฀ is฀ of฀ particular฀ importance฀ since฀ both฀ comprehensively฀ and฀ representatively฀sampled฀groups฀are฀subject฀to฀‘natural฀taxon฀sampling฀bias’฀in฀ the฀form฀of฀extinction,฀and฀techniques฀to฀estimate฀levels฀of฀extinction฀(Nee฀et฀ al.,฀1994;฀Barraclough฀&฀Reeves,฀2005)฀are฀based฀on฀the฀same฀molecular฀dating฀ methods.฀Given฀the฀large฀differences฀in฀estimated฀numbers฀of฀extant฀species฀ between฀the฀four฀genera฀under฀study฀here,฀(i.e.฀96฀species,฀and฀32฀accessions฀ available฀for฀Duguetia,฀250/49฀for฀Guatteria,฀35/23฀for฀Cremastosperma฀and฀ 15/7฀for฀Mosannona)฀we฀therefore฀consider฀it฀important฀to฀explore฀whether฀ taxon฀sampling฀itself฀might฀influence฀the฀outcome฀of฀the฀analysis.฀Thus,฀we฀ want฀to฀test฀whether฀dating฀results฀from฀analyses฀including฀all฀available฀taxa฀in฀ a฀study฀are฀also฀observed฀in฀analyses฀using฀equal฀sized฀subsets฀of฀those฀taxa.฀ The฀aims฀of฀this฀study฀were฀to฀test฀monophyly฀of฀the฀genera฀Cremastosperma,฀ Mosannona,฀Duguetia฀and฀Guatteria฀more฀thoroughly฀than฀Richardson฀et฀al.฀ (2004)฀ and฀ arrive฀ at฀ preliminary฀ estimates฀ for฀ the฀ ages฀ of฀ their฀ most฀ recent฀ common฀ancestors฀(MRCAs).฀We฀then฀used฀these฀estimations฀to฀investigate฀ whether฀diversification฀within฀these฀contrasting฀genera฀can฀be฀shown฀to฀have฀ occurred฀within฀the฀same฀or฀significantly฀different฀time฀slices.฀Finally,฀in฀order฀ to฀ assess฀ whether฀ and฀ to฀ what฀ extent฀ the฀ outcome฀ of฀ age฀ estimations฀ thus฀ produced฀ might฀ be฀ influenced฀ either฀ by฀ character฀ or฀ taxon฀ sampling,฀ resampling฀approaches฀were฀used฀and฀results฀compared. Materials and methods Ta x o n s a m p l i n g This฀study฀utilises฀both฀unpublished฀and฀published฀(Sauquet฀et฀al.,฀2003;฀ Mols฀et฀al.,฀2004;฀Mols฀et฀al.,฀submitted)฀rbcL,฀matK฀and฀trnL-F฀sequences฀from฀a฀ total฀of฀132฀taxa฀(see฀Appendix฀A).฀Twenty฀three฀accessions฀of฀Cremastosperma,฀ (representing฀around฀half฀of฀the฀(ca.฀35)฀described฀and฀un-described฀species฀ in฀the฀genus),฀7฀samples฀of฀ Mosannona,฀(representing฀just฀under฀half฀of฀the฀ 26 • Chapter 2 species),฀32฀samples฀of฀Duguetia,฀(representing฀one฀third฀of฀the฀species)฀and฀ 49฀samples฀of฀Guatteria,฀(representing฀around฀a฀fifth฀of฀the฀estimated฀number฀ of฀species)฀were฀included฀in฀this฀study฀(geographical฀localities฀of฀specimens฀ sampled฀are฀indicated฀in฀Appendix฀A฀and฀Fig.฀3).฀In฀addition,฀major฀clades฀in฀ Annonaceae,฀ including฀ sister฀ or฀ putative฀ sister฀ groups฀ of฀ the฀ four฀ genera฀ as฀ identified฀by฀Richardson฀et฀al.฀(2004),฀were฀represented฀here฀by฀23฀accessions.฀ These฀ represented฀ 11฀ species฀ each฀ from฀ the฀ long฀ branch฀ and฀ short฀ branch฀ clades฀(including฀taxa฀from฀across฀the฀tropics)฀plus฀Anaxagorea,฀sister฀group฀ to฀the฀rest฀of฀the฀Annonaceae.฀The฀numbers฀of฀accessions฀were฀limited฀in฀this฀ way฀in฀order฀to฀restrain฀computational฀times฀to฀reasonable฀lengths.฀Outgroups฀ were฀selected฀from฀other฀families฀within฀the฀Magnoliales฀sensu฀APG฀(2003),฀ i.e.฀Eupomatia฀(Eupomatiaceae฀is฀sister฀group฀of฀the฀Annonaceae฀(Qiu฀et฀al.,฀ 2000)),฀Magnolia฀(Magnoliaceae)฀and฀Coelocaryon฀(Myristicaceae). DNA extraction, PCR amplification and sequencing Total฀ genomic฀ DNA฀ was฀ extracted฀ using฀ a฀ modified฀ cetyl฀ trimethyl฀ ammonium฀bromide฀(CTAB)฀method฀(Doyle฀&฀Doyle,฀1987):฀50฀mg฀silica฀dried฀ or฀herbarium฀leaf฀material฀was฀homogenised฀in฀1300฀µl฀CTAB฀and฀incubated฀ for฀20฀minutes฀with฀ 12 µl฀2-mercaptoethanol฀at฀65°C,฀followed฀by฀90฀minutes฀ ambient฀mixing฀with฀1฀ml฀24:1฀chloroform:isoamylalcohol.฀Following฀10฀minutes฀ centrifugation฀at฀13,000฀rpm,฀300฀μl฀supernatant฀was฀purified฀using฀Wizard฀DNA฀ purification฀system฀(Promega฀corp.).฀We฀omitted฀isopropanol฀precipitation,฀in฀order฀ to฀avoid฀co-precipitation฀of฀oxidised฀material฀(Savolainen฀et฀al.,฀1995). PCR฀ amplification฀ conditions฀ were฀ modified฀ depending฀ on฀ the฀ qualities฀ of฀the฀DNA฀sample฀available.฀Samples฀extracted฀from฀herbarium฀material฀often฀ contain฀ lower฀ quantities฀ of฀ more฀ fragmented฀ DNA,฀ as฀ well฀ as฀ higher฀ levels฀ of฀ PCR฀inhibiting฀compounds฀(Savolainen฀et฀al.,฀1995).฀In฀most฀cases฀the฀rbcL฀gene฀ was฀amplified฀in฀two฀pieces฀and฀sequenced฀using฀primers฀1F/724R฀(Olmstead฀et฀ al.,฀1992)฀and฀636F/1460R฀(Fay฀et฀al.,฀1997;฀Fay฀et฀al.,฀1998).฀Where฀amplification฀ was฀ unsuccessful฀ (particularly฀ in฀ lower฀ quality฀ herbarium฀ extracted฀ DNA฀ samples)฀ further฀ internal฀ primers,฀ 217F:฀ 5’-GGACTTACCAGCCTTGATCG3 ’ ,฀9 2 2 F :฀5 ’ - G G T A T G C A C T T T C G T G T A C T A G C - 3 ’ ,฀5 3 6 R :฀ 3 ’ - G G T T A T C C G C C A A G A A C T A C G G - 5 ’ ฀a n d ฀1 1 0 4 R : ฀3 ’ GGCAGAACACCTGGCAAAGAGACC-5’฀ were฀ designed฀ and฀ applied฀ in฀ combination฀1F/536R,฀217F/724R,฀636F/1104R,฀and฀922F/1460R฀to฀amplify฀ the฀gene฀in฀four฀overlapping฀pieces฀of฀ca.฀500฀bp฀long.฀The฀angiosperm-universal฀ primers฀described฀by฀(Taberlet฀et฀al.,฀1991)฀were฀used฀to฀amplify฀and฀sequence฀the฀ trnL฀intron฀(primers฀C/D)฀and฀trnL-trnF฀spacer฀(primers฀E/F)฀separately.฀Partial฀ matK฀sequences฀were฀amplified฀and฀sequenced฀using฀primers฀390F฀and฀1326R฀ (Cuénoud฀ et฀ al.,฀ 2002).฀Where฀ amplification฀ was฀ unsuccessful฀ further฀ internal฀ primers,฀ MintF:฀ 5’-TCCTTTGGAACTGTTCTTGAGC-3’฀ and฀ MintR:฀ 5’GATCCTGTGCGGTTGAGACC-3’฀were฀designed฀and฀applied฀in฀combination฀ �olecular dating in Annonaceae • 27 390F/MintR฀and฀MintF/1326R฀in฀order฀to฀amplify฀the฀gene฀in฀two฀overlapping฀ pieces฀of฀ca.฀500฀bp฀long.฀A฀standard฀PCR฀protocol฀was฀used฀throughout,฀with฀the฀ addition฀of฀0.4฀%฀BSA฀(which฀was฀found฀to฀increase฀amplification฀in฀all฀samples).฀ The฀ PCR฀ profile฀ used฀ for฀ all฀ regions฀ comprised฀ 35฀ cycles฀ of฀ 30฀ sec.฀ at฀ 94°C,฀ 1฀min.฀at฀55°C,฀2฀min.฀at฀72°C,฀with฀an฀initial฀4฀min.฀at฀94°C฀and฀final฀7฀min.฀ at฀ 72°C.฀ PCR฀ products฀ were฀ purified฀ using฀ QIAquick฀ PCR฀ purification฀ kits฀ (Qiagen),฀sequenced฀with฀the฀PCR฀primers,฀and฀analysed฀by฀electrophoresis฀using฀ an฀automatic฀sequencer฀ABI฀3730XL. DNA sequence, phylogenetic & dating analyses DNA฀sequences฀were฀edited฀in฀SeqMan฀4.0฀(DNAStar฀Inc.,฀Madison,฀WI)฀and฀ aligned฀manually,฀resulting฀in฀alignments฀of฀1431฀positions฀(rbcL),฀1333฀positions฀ (trnL-F),฀and฀831฀positions฀(matK).฀Gaps฀in฀the฀trnL-F฀alignment฀were฀coded฀as฀ present/absent฀characters฀where฀they฀could฀be฀coded฀unambiguously,฀following฀ Simmons฀&฀Ochoterena฀(2000),฀and฀areas฀where฀the฀assessment฀of฀homology฀was฀ ambiguous฀were฀excluded฀from฀the฀analyses.฀Sequences฀were฀deposited฀at฀Genbank฀ (see฀Appendix฀A).฀ Data฀were฀analysed฀using฀the฀parsimony฀algorithm฀of฀the฀software฀package฀ PAUP*฀4.0b10฀(Swofford,฀2000),฀under฀the฀equal฀and฀unordered฀weights฀criterion฀ (Fitch฀ parsimony;฀ Fitch,฀ 1971).฀ Support฀ was฀ estimated฀ for฀ the฀ three฀ markers฀ independently฀ and,฀ after฀ checking฀ for฀ significantly฀ supported฀ incongruencies,฀ combined,฀using฀a฀500฀replicate฀bootstrap฀analysis฀with฀‘full’฀heuristic฀searches฀of฀50฀ random฀addition฀sequences,฀TBR,฀saving฀50฀trees฀each฀time.฀A฀maximum฀parsimony฀ search฀was฀performed฀on฀the฀combined฀matrix฀with฀500฀random฀taxon฀additions,฀ saving฀100฀trees฀per฀replicate,฀using฀the฀tree-bisection-reconnection฀(TBR)฀branch฀ swapping฀algorithm.฀ModelTest฀3.06฀(Posada฀&฀Crandall,฀1998)฀was฀used฀to฀select฀ the฀substitution฀model฀best฀fitting฀the฀(entire)฀dataset.฀This฀substitution฀model฀was฀ used฀to฀select฀the฀most฀likely฀of฀the฀maximum฀parsimony฀topologies฀as฀estimated฀ above฀and฀calculate฀branch฀lengths฀using฀the฀maximum฀likelihood฀(ML)฀criterion฀ as฀implemented฀in฀PAUP*.฀ In฀order฀to฀test฀whether฀our฀data฀set฀exhibited฀clocklike฀behaviour฀a฀likelihood฀ ratio฀test฀was฀performed฀on฀one฀of฀the฀parsimonious฀tree฀topologies:฀that฀with฀the฀ highest฀likelihood,฀derived฀from฀the฀complete฀matrix฀including฀all฀taxa.฀Likelihood฀ of฀the฀data฀with฀and฀without฀constraint฀of฀a฀molecular฀clock,฀were฀calculated฀and฀ the฀likelihood฀ratio฀statistic฀compared฀with฀ χ2฀critical฀value฀with฀135฀degrees฀ of฀ freedom฀ (i.e.,฀ number฀ of฀ taxa฀ minus฀ 2).฀Thereafter,฀ Sanderson’s฀ method฀ of฀nonparametric฀rate฀smoothing฀(NPRS)฀was฀applied฀(Sanderson,฀1997)฀as฀ implemented฀in฀the฀software฀package฀r8s฀(Sanderson,฀2002)฀in฀order฀to฀estimate฀ divergence฀times.฀ As฀ stated฀ above,฀ the฀ focus฀ of฀ this฀ chapter฀ is฀ on฀ the฀ relative฀ ages฀ of฀ the฀ four฀ genera.฀ However,฀ to฀ represent฀ these฀ ages฀ within฀ an฀ absolute฀ timescale฀ the฀molecular฀clock฀was฀calibrated฀by฀constraining฀the฀MRCA฀of฀the฀clade฀ 28 • Chapter 2 including฀ all฀Annonaceae฀ except฀ the฀ genus฀ Anaxagorea฀ and฀ the฀ basal฀ clade฀ including฀Cananga฀(see฀Fig.฀1)฀to฀between฀65฀and฀71.3฀mya.฀This฀age฀is฀based฀ on฀the฀presence฀of฀seeds฀with฀lamelliform฀ruminations฀in฀the฀Maastrichtian฀of฀ Nigeria฀(Chesters,฀1955).฀When฀used฀as฀single฀calibration฀point฀in฀preliminary฀ analyses฀this฀age฀produced฀a฀result฀congruent฀with฀the฀age฀of฀the฀MRCA฀of฀ the฀ genera฀ Annona฀ and฀ Asimina฀ (25฀ mya)฀ as฀ estimated฀ by฀Wikström฀ et฀ al.฀ (2001)฀(data฀not฀shown).฀The฀Annona/Asimina฀node฀was฀therefore฀also฀fixed฀ for฀further฀analyses. Divergence฀times฀were฀estimated฀for฀the฀complete฀matrix,฀as฀well฀as฀for฀ character฀re-sampled฀(analysis฀I)฀and฀taxon฀re-sampled฀matrices฀(analysis฀II).฀ In฀analysis฀I,฀confidence฀limits฀on฀branch฀lengths,฀reflecting฀stochasticity฀in฀ the฀sampling฀of฀character฀changes฀(‘substitutional฀noise’),฀were฀estimated฀by฀ 100฀replicates฀of฀bootstrap฀re-sampling฀(as฀also฀described฀in฀Wikström฀et฀ al.,฀ 2001),฀including฀all฀taxa,฀with฀subsequent฀ML฀branch฀length฀estimation฀on฀a฀ constrained฀tree฀topology฀(as฀above)฀for฀each฀bootstrap฀replicate.฀This฀results฀ in฀100฀trees฀comprising฀a฀range฀of฀estimated฀lengths฀for฀each฀branch฀of฀the฀ topology.฀Divergence฀times฀were฀then฀estimated฀for฀the฀4฀nodes฀representing฀ the฀ MRCAs฀ of฀ each฀ of฀ the฀ four฀ genera฀ in฀ each฀ of฀ these฀ 100฀ trees฀ and฀ the฀ results฀summarised฀giving฀mean฀values฀with฀standard฀deviation฀for฀specified฀ nodes฀(the฀MRCA฀for฀each฀genus)฀using฀the฀‘profile’฀command฀in฀r8s. In฀analysis฀II,฀100฀smaller฀matrices฀of฀38฀taxa฀were฀constructed฀by฀excluding฀ randomly฀all฀but฀three฀species฀each฀of฀Cremastosperma,฀Mosannona,฀Duguetia฀ and฀Guatteria,฀leaving฀the฀other฀Annonaceae฀and฀outgroup฀taxa฀in฀(see฀above).฀ This฀was฀done฀in฀order฀to฀assess฀whether฀and฀to฀what฀extent฀taxon฀sampling฀ might฀influence฀the฀outcome฀of฀age฀estimations.฀Each฀of฀these฀100฀taxon฀resampled฀matrices฀was฀then฀subjected฀to฀phylogenetic฀analysis฀followed฀by฀r8s฀ analysis฀as฀above.฀The฀oldest฀of฀the฀100฀estimations฀for฀the฀MRCA฀for฀each฀ genus฀ (this฀ time฀ represented฀ by฀ three฀ accessions)฀ is฀ then฀ interpreted฀ as฀ the฀ oldest฀ age฀ estimate฀ for฀ the฀ MRCA฀ of฀ that฀ genus.฀This฀ estimate฀ should฀ be฀ independent฀of฀numbers฀of฀taxa฀sampled฀across฀the฀four฀genera฀and฀thus฀suitable฀ to฀test฀whether฀there฀is฀a฀bias฀present฀in฀the฀form฀of฀uneven฀taxon฀sampling฀in฀ analysis฀I.฀Should฀previous฀analyses฀of฀all฀available฀taxa฀have฀resulted฀in฀a฀single฀ supported฀ topology,฀ this฀ could฀ have฀ been฀ used฀ to฀ constrain฀ relationships฀ in฀ each฀of฀the฀taxon฀re-sampled฀analyses.฀However,฀this฀was฀not฀the฀case.฀In฀order฀ to฀exclude฀the฀possibility฀of฀phylogenetic฀error฀causing฀erroneously฀old฀age฀ estimates,฀the฀trees฀from฀which฀the฀oldest฀age฀estimate฀overall฀for฀each฀genus฀ were฀obtained฀was฀inspected฀to฀confirm฀that฀the฀topology฀did฀not฀conflict฀with฀ the฀bootstrap฀consensus฀including฀all฀available฀taxa฀(Fig.฀4).฀ �olecular dating in Annonaceae • 29 Results Bootstrap฀ analysis฀ of฀ the฀ three฀ markers฀ independently฀ revealed฀ no฀ significantly฀ supported฀ conflict฀ (bootstrap฀ frequencies฀ ≥70%);฀ therefore฀ the฀ data฀were฀combined฀for฀further฀analyses.฀Parsimony฀analysis฀of฀the฀combined฀ data฀produced฀<50,000฀most฀parsimonious฀trees.฀Statistics฀of฀the฀three฀markers฀ separately฀are฀given฀in฀Table฀1,฀numbers฀of฀variable฀and฀parsimony฀informative฀ characters฀per฀marker฀within฀each฀of฀the฀four฀genera฀are฀given฀in฀Table฀2.฀Support฀ estimated฀using฀a฀500฀replicate฀bootstrap฀analysis฀is฀represented฀in฀Fig.฀4฀(see฀ below).฀Results฀were฀congruent฀with฀those฀produced฀by฀Richardson฀et฀al.฀(2004;฀ see฀Fig.฀3).฀Monophyly฀of฀clades฀including฀all฀accessions฀of฀Cremastosper ma,฀ Duguetia,฀Guatteria฀and฀Mosannona฀was฀supported฀by฀bootstrap฀percentages฀ of฀99,฀100,฀100฀and฀100฀respectively,฀but฀sequence฀variation฀within฀the฀clades฀ was฀generally฀low,฀as฀reflected฀by฀the฀small฀numbers฀of฀internal฀nodes฀supported฀ by฀<70฀%฀BS฀(1,฀2,฀4฀and฀3฀respectively).฀Numbers฀of฀variable฀characters฀were฀ higher฀in฀the฀long-branch฀clade฀genera฀Guatter ia฀and฀Duguetia฀than฀within฀ Cremastosperma฀and฀Mosannona,฀and฀those฀within฀Guatteria฀were฀highest฀of฀ the฀four฀(see฀Table฀2).฀However,฀the฀majority฀of฀maximum฀parsimony฀analyses฀ performed฀for฀analysis฀II฀resulted฀in฀a฀single฀most฀parsimonious฀estimation฀of฀ the฀relationships฀between฀the฀three฀accessions฀included฀for฀each฀of฀the฀four฀ genera. Ta b l e 1 . Details of maximum parsimony search �ength of variable parsimony % variable/ alignment characters informative informative on tree on tree characters characters in �igure 4 in �igure 4 r��L 1387 293 tr�L-F 1179 406 mat� 831 363 164 CI optimised RI optimised 21 / 12 0�521 0�885 190 34 / 16 0�733 0�927 198 44 / 24 0�706 0�920 Ta b l e 2 . Numbers of variable / parsimony informative characters, according to marker and in total within each of the four genera r��L tr�L-F mat� �otal �osa��o�a 22/11 17/4 9/1 48/16 Cremastosperma 35/9 22/6 18/6 75/21 ����et�a 74/46 41/8 52/9 167/63 ��atter�a 77/43 87/18 49/15 213/76 30 • Chapter 2 Coe�o�ar�o��pre�ss�� ��yristicaceae� �a��o��a��o��s ��agnoliaceae� ��pomat�a��e��et�� ��upomatiaceae� ��a�a�orea�s���at��a C�m�opeta��m�tor��os�m 100 �r����aea��a��e�peta�a �eoste�a�t�era�m�r�st����o��a 97 �oniothalamus griffithii 100 ���o�a�m�r��ata 100 �s�m��a�tr��o�a �o�o�ora�m�r�st��a 100 �as�mas��a�o��sootepe�se 100 F�ss�st��ma���a��es�e�s ��ar�a������a 99 89 99 100 100 100 ����et�a 100 100 �usaea peruviana 100 97 100 90 84 91 100 80 ��atter�a 96 99 97 �������a�p��osa �ree��a�o�e��ro��o���er� ��ptost��ma�morte�a�� �a�mea�s�r��ame�s�s 100 �����opeta��m�per�����o ��o�ops�s�r��es�e�s ��aro�e��a������ata 98 92 100 84 74 100 78 �osa��o�a �o�o�arp�a�e��e�ra �o��a�t��a��e�e���a �apra�t��s���r�����or�s �se��o�a��ra������a 99 Cremastosperma 98 0.01 F i g . 4 . Phylogram of Annonaceae (this study, analysis I) including all samples of Cremastosperma, Mosannona, Duguetia and Guatteria, with Magnoliales outgroups based on rbcL, trnL-F and matK. One of 50,000 most parsimonious trees, the one with highest maximum likelihood (ML) score, is shown, with branch lengths optimised based on the ML model described in the text. Bootstrap values at nodes are based on parsimony analysis. Scale bar indicates 0.01 subst. per site. �olecular dating in Annonaceae • 31 The฀substitution฀model฀selected฀as฀best฀fitting฀the฀(entire)฀dataset฀was฀the฀ general฀time฀reversible฀model฀(Tavare,฀1986)฀with฀a฀proportion฀of฀invariable฀ sites฀and฀gamma฀distributed฀rate฀(GTR฀+฀I฀+฀ Γ,฀α฀=฀0.995).฀One฀maximum฀ parsimony฀topology,฀the฀one฀with฀optimal฀likelihood฀of฀the฀data,฀is฀represented฀ in฀Fig.฀4฀with฀optimised฀ML฀branch฀lengths.฀Results฀of฀a฀likelihood฀ratio฀test฀ rejected฀ a฀ molecular฀ clock฀ (p฀ <฀ 0.0005).฀ Results฀ of฀ molecular฀ dating฀ using฀ NPRS฀are฀represented฀in฀Table฀3฀in฀the฀following฀way:฀firstly,฀the฀mean฀age฀of฀ the฀MRCAs฀of฀the฀four฀clades฀with฀estimation฀of฀error฀due฀to฀substitutional฀ noise;฀(analysis฀I).฀Second,฀the฀earliest฀age฀estimations฀derived฀from฀the฀taxonresampled฀matrices฀(analysis฀II).฀And฀finally฀the฀difference฀(where฀observed)฀ between฀the฀minimum฀age฀estimated฀for฀analysis฀I,฀and฀the฀value฀for฀analysis฀ II,฀which฀was฀not฀found฀to฀exceed฀the฀oldest฀age฀estimate฀under฀analysis฀I฀in฀ results฀for฀any฀of฀the฀four฀genera. Ta b l e 3 . Age estimations for the MRCAs of the four genera included in this study. Analysis I (see text) bootstrap re-sampled characters (mean with standard deviation), Analysis II; the oldest age estimates derived from taxon re-sampled matrices, and (the minimum value of) I minus II, where a difference was observed Analysis I �mya� Analysis II �mya� I - II Cremastosperma 35�57 ±7�38 28�82 - �osa��o�a 24�01 ± 4�74 22�55 - ����et�a 29�04 ± 4�52 20�34 4�18 ��atter�a 36�65 ± 2�50 19�42 14�73 The฀ latest฀ (youngest)฀ estimated฀ age฀ of฀ the฀ MRCA฀ of฀ Guatter ia,฀ with฀ confidence฀limits฀according฀to฀substitutional฀noise,฀was฀significantly฀older฀than฀ the฀earliest฀estimations฀for฀Mosannona฀and฀Duguetia.฀The฀mean฀age฀estimated฀ for฀the฀MRCA฀of฀Cremastosperma฀was฀similar฀to฀that฀of฀Guatteria,฀however฀ confidence฀limits฀for฀Cremastosper ma,฀Mosannona฀and฀Duguetia฀overlapped,฀ as฀did฀those฀for฀Cremastosperma฀and฀Guatteria฀(see฀Table฀3),฀due฀to฀the฀high฀ standard฀deviation฀in฀the฀result฀for฀Cremastosperma. The฀earliest฀age฀estimations฀derived฀from฀the฀taxon฀re-sampled฀matrices฀ fell฀ within฀ the฀ confidence฀ limits฀ according฀ to฀ substitutional฀ noise฀ in฀ Cremastosperma฀and฀Mosannona.฀The฀value฀for฀Duguetia฀was฀slightly฀younger฀ than฀the฀youngest฀confidence฀limit฀according฀to฀substitutional฀noise,฀and฀for฀ Guatteria,฀substantially฀younger.฀We฀did฀not฀further฀test฀for฀significance฀of฀the฀ differences฀observed฀as฀it฀is฀not฀obvious฀what฀test฀/chance฀distribution฀should฀ be฀applied. 32 • Chapter 2 Discussion Phylogeny reconstruction and monophyly of Cremastosperma, Duguetia, Guatteria, and Mosannona Phylogenetic฀analysis฀of฀an฀increased฀proportion฀of฀species฀of฀the฀genera฀ Cremastosper ma,฀ Mosannona,฀ Duguetia฀ and฀ Guatter ia฀ compared฀ with฀ previous฀studies฀has฀served฀to฀support฀the฀monophyly฀of฀all฀four฀genera฀with฀ a฀ high฀ degree฀ of฀ confidence.฀Taxa฀ were฀ selected฀ as฀ representatives฀ of฀ major฀ clades฀ in฀ the฀ family,฀ and฀ of฀ the฀ sister฀ groups฀ (or฀ most฀ likely฀ sister฀ groups)฀ of฀ Cremastosper ma,฀ Mosannona,฀ Duguetia฀ and฀ Guatter ia,฀ as฀ revealed฀ by฀ Richardson฀ et฀ al.฀(2004).฀Such฀a฀selection฀resulted฀in฀the฀reconstruction฀of฀ topologies฀congruent฀with฀those฀found฀in฀that฀study,฀limiting฀the฀scope฀of฀this฀ chapter฀to฀the฀questions฀at฀hand,฀thus฀making฀further฀discussion฀of฀the฀wider฀ topology฀here฀inappropriate. In฀ our฀ results,฀ resolution฀ of฀ relationships฀ within฀ Cremastosper ma,฀ Mosannona,฀ Duguetia฀ and฀ Guatter ia฀ was฀ low,฀ despite฀ the฀ extra฀ characters฀ sampled฀as฀compared฀with฀Richardson฀et฀al.฀(2004).฀The฀number฀of฀informative฀ characters฀contributed฀to฀the฀analyses฀by฀matK฀was฀relatively฀high฀(especially฀ given฀the฀length฀of฀the฀sequence)฀and฀levels฀of฀homoplasy฀comparable฀to฀those฀ in฀tr nL-F,฀both฀rather฀lower฀than฀in฀rbcL฀(see฀Table฀1),฀suggesting฀it฀to฀be฀a฀ useful฀marker฀for฀further฀application฀within฀Annonaceae.฀However,฀numbers฀of฀ characters฀informative฀at฀the฀intrageneric฀level,฀particularly฀in฀Cremastosperma฀ and฀ Mosannona,฀ were฀ low฀ (see฀Table฀ 2),฀ which฀ is฀ reflected฀ by฀ the฀ lack฀ of฀ resolution฀in฀relationships฀within฀them฀in฀Fig.฀4.฀This฀suggests฀that฀more,฀and฀ more฀variable,฀sequence฀data฀is฀necessary฀in฀order฀to฀resolve฀relationships฀at฀this฀ level,฀and฀that,฀ideally,฀AFLPTM฀markers฀should฀be฀applied฀as฀well.฀Alternatively,฀ low฀copy฀nuclear฀encoded฀genes฀may฀provide฀such฀variation,฀and฀is฀in฀any฀case฀ necessary฀in฀order฀to฀attempt฀to฀infer฀species฀rather฀than฀simply฀chloroplast฀ phylogenies. Molecular dating Calibrating฀a฀molecular฀clock฀by฀means฀of฀fossil฀data฀or฀geological฀events฀ under-฀and฀over-estimates฀ages฀respectively,฀to฀an฀unknowable฀degree,฀and฀must฀ be฀interpreted฀with฀caution฀(Bremer,฀2000;฀Smith฀&฀Peterson,฀2002;฀Hedges฀&฀ Kumar,฀2004).฀In฀attempting฀to฀make฀a฀comparison฀between฀the฀ages฀of฀the฀four฀ clades฀the฀analysed฀in฀this฀study,฀the฀impact฀of฀error฀arising฀from฀the฀calibration฀ of฀the฀molecular฀clock฀was฀avoided฀by฀including฀them฀in฀a฀single฀analysis฀using฀ the฀same฀calibration.฀This฀allows฀direct฀comparison฀of฀relative฀ages฀to฀be฀made,฀ even฀if฀the฀accuracy฀of฀the฀absolute฀dates฀remains฀un-assessed. Rate฀ heterogeneity฀ is฀ apparent฀ in฀ Annonaceae,฀ as฀ exemplified฀ by฀ the฀ differences฀in฀proportions฀of฀variable฀characters฀found฀between฀the฀short-branch฀ and฀ long-branch฀ clades.฀ It฀ can฀ be฀ addressed฀ using฀ a฀ number฀ of฀ techniques฀ �olecular dating in Annonaceae • 33 assuming฀autocorrelation฀of฀rates฀across฀the฀tree,฀including฀parametric฀Bayesian฀ approaches฀(e.g.฀Thorne฀et฀al.,฀1998)฀and฀semi-parametric฀Penalised฀Likelihood฀ (Sanderson,฀2002).฀The฀nonparametric฀rate฀smoothing฀technique฀(Sanderson,฀ 1997)฀has฀been฀used฀in฀a฀number฀of฀studies฀representing฀all฀angiosperms฀(e.g.฀ Sanderson฀&฀Doyle,฀2001;฀Wikström฀et฀al.,฀2001;฀Davies฀et฀al.,฀2004),฀wherein฀ variation฀in฀generation฀time,฀a฀major฀determining฀factor฀in฀rates฀of฀molecular฀ evolution,฀is฀far฀greater฀than฀within฀Annonaceae,฀where฀herbaceous฀habit฀is฀ not฀ observed.฀ Despite฀ possible฀ drawbacks฀ of฀ the฀ method฀ under฀ particular฀ conditions฀(Sanderson,฀2002),฀NPRS฀was฀used฀in฀this฀study,฀in฀order฀to฀keep฀ the฀numbers฀and฀durations฀of฀the฀analyses฀within฀reasonable฀bounds,฀and฀to฀ enable฀comparison฀with฀results฀obtained฀in฀other฀studies฀(e.g.฀Barraclough฀&฀ Reeves,฀2005,฀and฀Bakker฀et฀al.,฀2005).฀ Because฀combined฀analysis฀of฀rbcL,฀trnL-F฀and฀matK฀sequences฀failed฀to฀ resolve฀relationships฀within฀the฀four฀genera฀it฀was฀necessary฀to฀select฀effectively฀ arbitrarily฀ resolved฀ topologies฀ (the฀ most฀ parsimonious฀ trees฀ with฀ the฀ best฀ maximum฀likelihood฀scores)฀with฀which฀to฀make฀age฀estimations.฀Long-branch฀ attraction฀ in฀ one฀ of฀ the฀ examples฀ provided฀ by฀ Sanderson฀ &฀ Doyle฀ (2001)฀ resulted฀in฀overestimation฀of฀the฀age฀of฀the฀MRCA฀of฀Angiosperms฀due฀to฀ spurious฀placing฀of฀Oryza฀as฀sister฀group฀of฀the฀Angiosperms.฀The฀possibility฀of฀ such฀error฀affecting฀our฀age฀estimations฀when฀analysing฀all฀taxa฀cannot฀entirely฀ be฀excluded;฀however,฀the฀topology฀used฀showed฀no฀conflict฀with฀provisional฀ results฀ of฀ phylogeny฀ reconstruction฀ in฀ the฀ four฀ genera฀ based฀ on฀ analysis฀ of฀ more฀characters฀(data฀not฀shown).฀No฀further฀attempt฀was฀made฀to฀estimate฀ the฀effect฀of฀phylogenetic฀uncertainty฀on฀those฀date฀estimations,฀as฀the฀ages฀of฀ interest,฀those฀of฀the฀MRCAs฀of฀each฀of฀the฀four฀genera,฀were฀represented฀by฀ nodes฀which฀were฀present฀in฀the฀strict฀consensus฀of฀all฀most฀parsimonious฀trees฀ (not฀shown)฀and฀which฀received฀97-100%฀bootstrap฀support. In฀the฀taxon฀re-sampling฀technique฀(Analysis฀II)฀detailed฀here,฀the฀ages฀ of฀ the฀ MRCAs฀ of฀ the฀ four฀ genera฀ are฀ represented฀ by฀ the฀ oldest฀ of฀ the฀ age฀ estimates฀(see฀Table฀3).฀This฀is฀because฀many฀of฀the฀replications฀are฀likely฀to฀ involve฀ species฀ with฀ relatively฀ recent฀ MRCAs,฀ where฀ the฀ relatively฀ fewer฀ lineages฀ which฀ effectively฀ define฀ the฀ age฀ of฀ the฀ group฀ are฀ missed.฀ Ideally฀ a฀ topology฀supported฀by฀analyses฀including฀all฀taxa฀would฀be฀imposed฀for฀each฀ of฀the฀taxon฀re-sampled฀matrices,฀and฀in฀principle฀a฀range฀of฀estimations฀could฀ thus฀be฀produced฀for฀each฀node฀in฀the฀topology.฀At฀the฀least,฀given฀support฀for฀ the฀first-branching฀(‘basal’)฀lineage฀in฀a฀topology,฀this฀lineage฀could฀be฀included฀ in฀all฀re-sampled฀matrices฀in฀order฀to฀arrive฀at฀a฀thorough฀test฀of฀the฀age฀of฀the฀ MRCA.฀No฀fully฀supported฀topology฀was฀available฀for฀any฀of฀the฀genera฀based฀ on฀these฀results,฀and฀only฀in฀Guatteria฀was฀there฀support฀for฀the฀first฀branching฀ lineage฀within฀the฀genus.฀It฀was฀therefore฀decided฀to฀treat฀each฀genus฀equally฀ by฀ re-sampling฀ randomly฀ without฀ imposing฀ any฀ topological฀ constraints.฀We฀ have฀no฀reason฀to฀believe฀that฀the฀topologies฀produced฀for฀analysis฀II฀would฀all฀ 34 • Chapter 2 be฀congruent฀with฀a฀‘true’฀phylogeny฀of฀all฀the฀taxa฀available,฀however,฀those฀ points฀representing฀the฀oldest฀age฀estimates฀were฀inspected฀and฀found฀not฀to฀ conflict฀with฀the฀supported฀relationships฀based฀on฀analysis฀of฀all฀taxa฀(as฀in฀Fig.฀ 4). From฀the฀standard฀deviations฀according฀to฀substitutional฀noise฀it฀appeared฀ that฀ diversifications฀ in฀ Mosannona฀ and฀ Duguetia฀ occurred฀ within฀ roughly฀ the฀same฀period฀of฀time,฀beginning฀significantly฀later฀than฀that฀of฀Guatter ia฀ (see฀Table฀3).฀Lack฀of฀precision฀in฀the฀result฀for฀Cremastosper ma฀may฀be฀an฀ effect฀of฀the฀low฀number฀of฀variable฀characters฀in฀the฀DNA฀sequence฀data,฀ the฀difference฀compared฀with฀Mosannona฀explained฀by฀the฀higher฀number฀of฀ accessions฀included.฀A฀parametric฀or฀semi-parametric฀approach฀may฀be฀more฀ appropriate฀for฀Cremastosper ma,฀given฀the฀abundance฀of฀short฀branches฀in฀a฀ phylogeny฀of฀this฀genus฀based฀on฀this฀data,฀as฀nonparametric฀rate฀smoothing฀ tends฀to฀over-fit฀the฀data฀in฀such฀cases,฀leading฀to฀rapid฀fluctuations฀in฀rate฀ (Sanderson,฀2002). Ages฀ for฀ Cremastosper ma฀ and฀ Mosannona฀ estimated฀ from฀ taxon฀ resampling฀(analysis฀II)฀fell฀within฀the฀confidence฀intervals฀obtained฀by฀re-sampling฀ characters฀(analysis฀I).฀However,฀the฀taxon฀re-sampling฀estimations฀for฀the฀age฀ of฀ the฀ MRCAs฀ of฀ Duguetia฀ and,฀ more฀ markedly,฀ that฀ for฀ Guatter ia,฀ were฀ younger฀than฀the฀upper฀confidence฀limits฀estimated฀by฀character฀re-sampling.฀ The฀MRCA฀of฀Guatteria฀yielded฀the฀most฀precise฀result฀in฀the฀character฀resampling฀analysis฀judged฀by฀the฀smallest฀standard฀deviations,฀which฀is฀possibly฀ a฀ reflection฀ of฀ the฀ higher฀ number฀ of฀ informative฀ characters฀ present฀ in฀ the฀ Guatteria฀data.฀Guatteria฀was฀represented฀by฀the฀largest฀number฀of฀accessions฀ in฀the฀analysis,฀i.e.฀49฀compared฀to฀23,฀7฀and฀32฀in฀Cremastosperma,฀Mosannona฀ and฀Duguetia,฀respectively.฀It฀would฀thus฀be฀expected฀that฀if฀difference฀in฀the฀ number฀ of฀ samples฀ were฀ to฀ bias฀ the฀ result,฀ that฀ this฀ effect฀ would฀ be฀ most฀ noticeable฀in฀Guatteria. The฀ observation฀ of฀ this฀ conflict฀ raises฀ a฀ conundrum:฀ should฀ apparently฀ under-representing฀a฀species-rich฀genus฀such฀as฀Guatter ia฀by฀including฀only฀ three฀species฀in฀the฀taxon฀re-sampling฀lead฀to฀an฀underestimation฀of฀the฀age฀of฀ the฀MRCA?฀Or฀should฀including฀an฀absolutely฀higher฀number฀of฀accessions฀ of฀Guatteria,฀relative฀to฀those฀of฀the฀other฀genera฀in฀analysis฀I,฀be฀interpreted฀ as฀an฀overestimation฀in฀age?฀If฀the฀former฀conclusion฀were฀to฀be฀drawn,฀and฀ molecular฀dating฀only฀be฀performed฀on฀fully฀sampled฀clades,฀do฀we฀then฀need฀to฀ factor฀in฀a฀more฀realistic฀assessment฀of฀the฀levels฀of฀sampling,฀by฀first฀estimating฀ levels฀of฀extinction?฀If฀so,฀how฀would฀we฀ensure฀that฀this฀estimation฀itself,฀based฀ on฀the฀same฀analyses,฀is฀not฀subject฀to฀the฀same฀uncertainty? Alternatively,฀these฀inconsistencies฀may฀be฀a฀result฀of฀the฀particular฀method฀ used.฀One฀possible฀explanation฀might฀be฀multiple฀substitutions฀as฀revealed฀by฀ increased฀taxon฀sampling.฀ML฀branch฀lengths,฀as฀used฀here,฀should฀be฀less฀affected฀ by฀ this฀ phenomenon฀ than฀ those฀ reconstructed฀ under฀ maximum฀ parsimony.฀ �olecular dating in Annonaceae • 35 However,฀ML฀may฀not฀be฀entirely฀immune฀to฀this฀effect฀(T.฀Barraclough,฀pers.฀ com.),฀and฀where฀sequence฀variation฀in฀a฀clade฀is฀relatively฀higher,฀as฀is฀the฀ case฀in฀Guatteria฀(and฀Duguetia฀to฀a฀lesser฀extent),฀and฀homoplasy฀significant฀ (particularly฀in฀rbcL;฀see฀Table฀1),฀this฀might฀magnify฀the฀effect.฀A฀second฀possible฀ explanation฀might฀be฀provided฀by฀Barraclough฀&฀Reeves฀(2005):฀bias฀caused฀by฀ low฀sequence฀variation฀(characterised฀by฀a฀large฀number฀of฀unresolved฀inner฀ branches฀in฀a฀phylogeny)฀leading฀to฀an฀apparent฀slow฀down฀in฀LTT฀plots.฀That฀ this฀would฀be฀apparent฀in฀Guatteria฀and฀Duguetia,฀and฀not฀in฀Cremastosperma฀ and฀Mosannona฀where฀sequence฀variation฀was฀lower,฀might฀be฀explained฀in฀ analysis฀I฀by฀the฀imprecision฀in฀the฀result฀for฀Cremastosper ma฀and฀the฀lower฀ ratio฀of฀taxa฀to฀variable฀characters฀in฀Mosannona. One฀ of฀ the฀ few฀ supported฀ nodes฀ within฀ Guatter ia฀ is฀ that฀ between฀ G.฀ anomala฀and฀the฀rest฀of฀the฀genus.฀The฀earliest฀age฀estimations฀in฀analysis฀II฀for฀ the฀MRCA฀of฀Guatteria฀were฀derived฀from฀re-sampling฀which฀included฀this฀ taxon.฀Although฀sampling฀of฀Guatteria฀in฀analysis฀I฀was฀the฀highest฀of฀the฀four฀ genera,฀it฀actually฀represented฀the฀lowest฀proportion฀of฀the฀total฀numbers฀of฀ extant฀species฀for฀the฀four฀genera.฀Targeted฀sampling฀throughout฀the฀range฀of฀ geographic฀and฀morphological฀diversity฀within฀such฀a฀group฀might฀be฀hoped฀ to฀ reveal฀ such฀ anomalies,฀ but฀ even฀ when฀ a฀ larger฀ proportion฀ of฀ the฀ extant฀ species฀have฀been฀sampled฀the฀possibility฀remains฀that฀more฀early฀branching฀ lineages฀will฀be฀discovered.฀This฀would฀definitively฀increase฀the฀age฀estimation฀ for฀the฀MRCA฀of฀the฀genus.฀However,฀given฀the฀results฀shown฀here,฀it฀would฀ seem฀sensible฀to฀investigate฀whether฀increased฀sampling฀of฀the฀younger฀clade฀ representing฀the฀majority฀of฀sampled฀species฀of฀Guatteria฀might,฀less฀intuitively,฀ and฀perhaps฀misleadingly,฀lead฀to฀the฀same฀result. Conclusions As฀sister฀groups,฀the฀‘short฀branch’฀and฀‘long฀branch’฀clades฀of฀Annonaceae฀ are฀ of฀ equal฀ age.฀ However,฀ the฀ latter฀ consists฀ of฀ twice฀ as฀ many฀ species,฀ the฀ majority฀of฀these฀clustered฀within฀a฀small฀number฀of฀genera.฀How฀meaningful฀ is฀this฀difference?฀Is฀it฀simply฀a฀question฀of฀taxonomy,฀or฀does฀it฀reflect฀a฀real฀ difference฀in฀the฀evolutionary฀history฀of฀these฀groups?฀Could฀such฀differences฀ be฀ a฀ result฀ of฀ the฀ morphological฀ innovations,฀ such฀ as฀ syncarpous฀ fruits฀ and฀ liana฀habit,฀only฀found฀in฀the฀long฀branch฀clade,฀and฀could฀these฀have฀caused฀ differences฀in฀rates฀of฀speciation฀and฀extinction฀compared฀with฀the฀short฀branch฀ clade? Results฀of฀phylogeny฀reconstruction฀presented฀here฀support฀monophyly฀ in฀the฀genera฀Cremastosperma,฀Mosannona,฀Duguetia฀and฀Guatteria,฀although฀ more฀ stringent฀ tests฀ would฀ require฀ increased฀ sampling,฀ to฀ a฀ lesser฀ extent฀ in฀ Duguetia฀but฀especially฀in฀Guatter ia฀where฀the฀proportion฀of฀extant฀species฀ 36 • Chapter 2 sampled฀ was฀ lowest.฀ Estimation฀ of฀ the฀ ages฀ of฀ the฀ MRCAs฀ of฀ these฀ genera,฀ with฀confidence฀limits฀reflecting฀error฀due฀to฀finite฀sampling฀of฀stochastically฀ evolving฀characters,฀resulted฀in฀significantly฀greater฀age฀for฀Guatteria฀than฀for฀ Duguetia฀or฀Mosannona฀(the฀result฀for฀Cremastosperma฀was฀too฀imprecise฀to฀ be฀able฀to฀draw฀such฀conclusions). In฀addressing฀questions฀of฀taxon฀sampling,฀with฀respect฀to฀molecular฀dating,฀ we฀encountered฀an฀apparent฀conflict.฀A฀more฀species฀rich฀group฀might฀either฀ be฀hypothesised฀to฀have฀originated฀earlier฀than฀one฀comprising฀fewer฀species฀ (such฀as฀was฀first฀apparent฀in฀Guatteria),฀or฀to฀have฀been฀the฀result฀of฀a฀recent฀ more฀rapid฀radiation.฀In฀a฀large฀group,฀achieving฀complete฀sampling,฀or฀even฀ knowing฀the฀proportion฀of฀complete฀sampling฀that฀has฀been฀achieved,฀may฀be฀ a฀practical฀impossibility.฀It฀would฀seem฀intuitively฀reasonable฀to฀assume฀that,฀ with฀knowledge฀of฀the฀broad฀topology฀of฀such฀a฀group,฀a฀subset฀of฀taxa฀can฀be฀ chosen฀to฀represent฀clades฀for฀which฀ages฀can฀then฀be฀estimated฀(for฀example,฀ in฀angiosperms:฀Sanderson฀&฀Doyle,฀2001;฀Wikström฀et฀al.,฀2001;฀Davies฀et฀al.,฀ 2004).฀However,฀results฀produced฀here฀suggest฀that฀increased฀taxon฀sampling฀ from฀within฀a฀clade฀(and฀not฀simply฀of฀lineages฀‘basal’฀to฀it)฀may฀result฀in฀older฀ age฀estimations฀for฀that฀clade. It฀ is฀ necessary฀ to฀ determine฀ whether฀ this฀ result฀ is฀ robust฀ to฀ different฀ methods฀of฀age฀estimation,฀to฀use฀with฀more฀resolved฀phylogenies฀and฀different฀ taxonomic฀ groups.฀The฀ implication,฀ should฀ it฀ indeed฀ prove฀ to฀ be฀ robust,฀ is฀ either฀that฀species฀rich฀groups฀may฀have฀diversified฀more฀recently฀than฀a฀densely฀ sampled฀analysis฀might฀suggest,฀or฀alternatively,฀that฀a฀reduced฀analysis฀of฀carefully฀ selected฀‘placeholder’฀taxa฀might฀underestimate฀that฀age.฀The฀predominance฀of฀ large฀genera฀in฀the฀long-branch฀clade฀might฀reflect฀evolutionary฀history,฀or฀ the฀figment฀of฀a฀taxonomist’s฀system.฀Further฀investigation฀will฀be฀required฀to฀ discern฀between฀these฀two฀competing฀explanations. Acknowledgements Permission฀ to฀ extract฀ DNA฀ from฀ herbarium฀ specimens฀ was฀ kindly฀ granted฀by฀herbaria฀NY฀and฀MO.฀Study฀of฀specimens฀at฀the฀Swedish฀Museum฀ of฀Natural฀History฀was฀supported฀by฀a฀grant฀from฀the฀High฀Lat฀programme,฀ which฀was฀made฀available฀by฀the฀European฀Community฀-฀Access฀to฀Research฀ Infrastructure฀action฀of฀the฀Improving฀Human฀Potential฀Programme฀(MDP).฀ The฀ authors฀ further฀ thank฀ Freek฀ Bakker฀ and฀Tim฀ Barraclough฀ for฀ valuable฀ comments฀and฀suggestions.฀Fig.฀4฀was฀produced฀using฀the฀program฀TreeGraph฀ (Müller฀&฀Müller฀2004). �olecular dating in Annonaceae • 37 38 • Chapter 2 Chapter 3 ‘Andean-centred’ genera in the short branch clade of Annonaceae: testing biogeographic hypotheses using phylogeny reconstruction and molecular dating M i c h a e l D . P i r i e 1 , L a r s W. C h a t r o u 1 , J o h a n B . M o l s 2 , Roy H. J. Erkens1 and Jessica Oosterhof1 Submitted to Journal of Biogeography 1 ฀ Nat i onaal Her bar i um Neder l and, Uni ver s i t ei t Ut r ec ht br anc h, Hei del ber gl aan 2, 3584 CS Ut r ec ht , The Net her l ands 2 ฀ Nat i onaal Her bar i um Neder l and, Uni ver s i t ei t Lei den br anc h, P. � . Box 9514, 2300 RA Lei den, The Net her l ands Biogeography of Andean-centred Annonaceae • 39 Abstract A฀number฀of฀Neotropical฀genera฀of฀Annonaceae฀display฀‘Andean-centred’฀ distribution฀patterns,฀with฀high฀species฀richness฀on฀both฀sides฀of฀the฀Andes฀ mountain฀ range.฀We฀ test฀ biogeographic฀ hypotheses฀ regarding฀ the฀ origin฀ of฀ Andean฀centred฀plant฀groups฀by฀reconstructing฀phylogeny฀and฀estimating฀the฀ timing฀of฀diversifications฀in฀four฀genera:฀Cremastosperma,฀Klarobelia,฀Malmea,฀ and฀Mosannona,฀using฀chloroplast฀DNA฀sequences฀of฀related฀Annonaceae฀taxa฀ plus฀outgroups.฀Almost฀all฀accessions฀of฀the฀four฀genera฀formed฀monophyletic฀ groups฀nested฀within฀a฀clade฀including฀all฀of฀the฀South฀American-centred฀‘short฀ branch฀clade’฀genera,฀to฀the฀exclusion฀of฀two฀largely฀Asian฀and฀two฀African฀ clades.฀We฀infer฀a฀common฀ancestor฀of฀the฀four฀genera฀to฀have฀originated฀in฀ South฀America,฀but฀not฀by฀vicariance฀of฀an฀ancestral฀population฀on฀Gondwana.฀ Radiations฀of฀these฀clades฀could฀have฀been฀influenced฀by฀the฀Andean฀orogeny,฀ but฀further฀tests฀require฀greater฀precision฀in฀the฀molecular฀dating฀results.฀ Keywords:฀ Andes,฀ Annonaceae,฀ biogeography,฀ Cremastosperma,฀ Klarobelia,฀ Malmea,฀ Mosannona,฀Neotropics,฀phylogeny฀reconstruction,฀molecular฀dating. Abbreviations:฀ BS฀ (bootstrap฀ support);฀ LBC฀ (long฀ branch฀ clade);฀ MP฀ (maximum฀ parsimony);฀ MRCA฀ (most฀ recent฀ common฀ ancestor);฀ NPRS฀ (nonparametric฀ rate฀ smoothing);฀ PL฀ (penalized฀ likelihood);฀ PP฀ (posterior฀ probability);฀ SAC฀ clade฀ (South฀ American฀centred฀clade);฀SBC฀(short฀branch฀clade). Introduction Around฀ a฀ third฀ of฀ all฀ flowering฀ plants฀ are฀ found฀ in฀ the฀ Neotropics฀ (tropical฀America)฀(Smith฀et฀ al.,฀2004).฀Two฀areas฀within฀the฀Neotropics;฀the฀ tropical฀Andes฀(including฀forests฀on฀the฀eastern฀side฀of฀the฀Andes฀extending฀ from฀ Colombia฀ through฀ eastern฀ Ecuador฀ and฀ Peru฀ as฀ far฀ south฀ as฀ Bolivia),฀ and฀the฀Chocó/Darién/Western฀Ecuador฀region฀(the฀narrow฀tropical฀zone฀on฀ the฀Pacific฀Ocean฀side฀of฀the฀Andean฀mountain฀chain)฀together฀house฀22,500฀ endemic฀ plant฀ species฀ representing฀ 7.5%฀ of฀ all฀ species฀ of฀ plants฀ worldwide฀ (Myers฀et฀al.,฀2000).฀Understanding฀the฀origins฀of฀these฀biodiverse฀areas฀in฀the฀ Neotropics฀will฀help฀us฀to฀determine฀why฀this฀region฀is฀so฀species฀rich.฀ In฀his฀1982฀paper,฀Al฀Gentry฀proposed฀historical฀biogeographical฀scenarios฀ that฀might฀explain฀the฀high฀floristic฀diversity฀of฀the฀Neotropics฀in฀general,฀and฀ the฀areas฀surrounding฀the฀Andes฀in฀particular.฀Gentry฀considered฀the฀majority฀ of฀South฀American฀taxa฀to฀be฀Gondwanan-derived฀and฀he฀distinguished฀two฀ groups฀within฀them฀on฀the฀basis฀of฀distribution฀patterns฀and฀growth฀forms.฀ Amazon-centred฀ taxa,฀ sensu฀ Gentry,฀ are฀ largely฀ canopy฀ trees฀ and฀ lianas.฀ Andean-centred฀taxa,฀in฀contrast,฀are฀diverse฀in฀the฀lowlands฀near฀the฀base฀of฀ 40 • Chapter 3 the฀mountains฀and฀in฀middle฀elevation฀cloud฀forests,฀areas฀corresponding฀to฀the฀ tropical฀Andes฀and฀Chocó/Darién/Western฀Ecuador฀region,฀with฀very฀poor฀ representation฀in฀Amazonia.฀They฀are฀chiefly฀shrubs,฀epiphytes฀and฀palmettos.฀ Gentry’s฀explanation฀for฀the฀distribution฀of฀Andean-centred฀taxa฀was฀that฀they฀ were฀subject฀to฀speciation฀as฀a฀result฀of฀the฀orogeny฀of฀the฀Andean฀mountain฀ chain.฀ However,฀ South฀American฀ geological฀ history฀ of฀ the฀ last฀ few฀ tens฀ of฀ millions฀of฀years฀is฀complex,฀and฀a฀number฀of฀overlapping฀events฀may฀be฀critical฀ in฀determining฀current฀distribution฀patterns฀-฀in฀order฀of฀decreasing฀age:฀the฀ Andean฀ orogeny,฀ closure฀ of฀ Panama฀ isthmus฀ and฀ the฀ numerous฀ episodes฀ of฀ climatic฀changes฀occurring฀throughout฀the฀Pleistocene฀(Burnham฀&฀Graham,฀ 1999).฀Even฀the฀relative฀importance฀of฀historical฀versus฀present฀day฀(ecological)฀ factors฀in฀determining฀current฀plant฀species฀distributions฀remains฀a฀point฀of฀ contention฀(e.g.฀see฀Tuomisto฀&฀Ruokolainen,฀1997). The฀pantropically฀distributed฀family฀Annonaceae฀comprises฀around฀2500฀ species฀of฀trees฀and฀lianas,฀found฀predominantly฀in฀tropical฀rain฀forests.฀Over฀ 900฀species฀are฀recognised฀in฀the฀Neotropics฀(Chatrou฀et฀al.,฀2004),฀where฀they฀ represent฀a฀significant฀part฀of฀plant฀diversity,฀both฀in฀terms฀of฀number฀of฀species฀ and฀ number฀ of฀ individuals฀ (Valencia฀ et฀ al.,฀ 1994;฀Ter฀ Steege฀ et฀ al.,฀ 2000).฀ The฀majority฀of฀Annonaceae฀genera฀in฀South฀America฀would฀be฀considered฀ Amazon-centred฀according฀to฀Gentry,฀i.e.฀most฀species฀are฀medium฀to฀large฀ trees฀distributed฀across฀Amazonia.฀However,฀some฀groups,฀notably฀the฀genera฀ Cremastosper ma,฀ Klarobelia,฀ Malmea,฀ and฀ Mosannona฀ display฀ markedly฀ Andean฀centred฀distributions฀(lowland฀to฀pre-montane฀forest฀usually฀only฀up฀ to฀1,500฀m)฀and฀are฀usually฀small฀under-story฀trees. Cremastosperma฀can฀be฀distinguished฀from฀other฀Neotropical฀Annonaceae฀ by฀ its฀ raised฀ midrib฀ with฀ a฀ unique฀ longitudinal฀ groove฀ and฀ comprises฀ approximately฀35฀species.฀Most฀species฀are฀found฀along฀the฀Andes฀in฀Peru฀and฀ Ecuador,฀ with฀ significant฀ diversity฀ also฀ extending฀ north฀ through฀ Colombia฀ into฀Panama฀and฀Costa฀Rica฀(see฀Fig.฀1a).฀Only฀four฀species฀have฀been฀found฀ further฀east:฀C.฀ macrocar pum฀and฀C.฀ venezuelanum฀in฀coastal฀Venezuela,฀C.฀ brevipes฀ on฀ the฀ Guiana฀ shield฀ and฀ C.฀ monosper mum฀ widespread฀ from฀ Peru฀ across฀northern฀Bolivia฀and฀central฀Brazil.฀Of฀the฀19฀species฀originally฀described฀ under฀Malmea,฀12฀were฀subsequently฀moved฀to฀three฀new฀genera฀described฀by฀ Chatrou฀(1998)฀on฀the฀basis฀of฀leaf,฀inflorescence฀and฀seed฀characters:฀Klarobelia฀ (now฀ including฀ 12฀ recognised฀ species),฀ Mosannona฀ (14)฀ and฀ Pseudomalmea฀ (3).฀ Malmea฀ now฀ includes฀ 6฀ recognised฀ species.฀The฀ distribution฀ of฀ species฀ of฀Klarobelia,฀Malmea,฀and Mosannona฀(Fig.฀1b,฀c฀&฀d)฀are฀markedly฀similar฀ to฀ those฀ of฀ Cremastosper ma.฀ In฀ all฀ four฀ genera฀ no฀ single฀ species฀ is฀ found฀ either฀on฀both฀sides฀of฀the฀Andean฀mountain฀chain,฀or฀distributed฀across฀more฀ than฀one฀of฀the฀further฀disjunct฀areas฀in฀the฀Guianas,฀Venezuela฀and฀tropical฀ Andes.฀However,฀there฀are฀differences:฀Mosannona฀is฀distributed฀further฀into฀ Central฀America,฀ with฀ one฀ species฀ found฀ as฀ far฀ north-west฀ as฀ the฀ Mexican฀ Biogeography of Andean-centred Annonaceae • 41 a b 42 • Chapter 3 c d F i g . 1 . Distribution maps: Mollweide projection. Collections denoted by grey dots, location of those sampled for DNA by black asterisks. 1a: Cremastosperma 1b: Klarobelia (black star: Pseudephedranthus fragrans) 1c: Malmea 1d: Mosannona Biogeography of Andean-centred Annonaceae • 43 states฀of฀Veracruz฀and฀Nayarit,฀and,฀as฀is฀also฀the฀case฀for฀Malmea,฀no฀species฀of฀ Mosannona฀have฀been฀collected฀in฀coastal฀Venezuela.฀Further,฀the฀type฀species฀ of฀Malmea,฀M.฀ obovata฀is฀only฀known฀from฀one฀collection฀from฀the฀Atlantic฀ coast฀of฀Brazil,฀and฀M.฀manausensis฀is฀found฀in฀the฀heart฀of฀the฀Amazon฀basin฀ (Fig.฀1c):฀in฀neither฀of฀these฀two฀areas฀have฀species฀of฀the฀other฀three฀genera฀ been฀collected.฀ Phylogeny reconstruction in Annonaceae: are Cremastosperma, Klarobelia, Malmea and Mosannona monophyletic? The฀ informal฀ names฀ long฀ branch฀ clade฀ (LBC)฀ and฀ short฀ branch฀ clade฀ (SBC),฀have฀been฀applied฀to฀the฀two฀major,฀pan-tropically฀distributed฀clades฀ comprising฀the฀majority฀of฀species฀in฀Annonaceae฀(see฀Fig.฀2),฀reflecting฀the฀ large฀ difference฀ in฀ the฀ rate฀ of฀ molecular฀ change฀ apparent฀ as฀ branch฀ lengths฀ representing฀inferred฀changes฀on฀phylogenetic฀trees฀(Richardson฀et฀ al.,฀2004;฀ Pirie฀et฀al.,฀this฀thesis,฀Chapter฀2).฀Phylogenetic฀analyses฀of฀Annonaceae฀using฀ plastid฀rbcL฀and฀trnL-F฀(Mols฀et฀al.,฀2004;฀Richardson฀et฀al.,฀2004)฀and฀rbcL,฀ tr nL-F฀ and฀ matK฀ (Pirie฀ et฀ al.,฀ this฀ thesis,฀ Chapter฀ 2)฀ DNA฀ sequence฀ data฀ placed฀Cremastosper ma,฀Klarobelia,฀Malmea฀and฀Mosannona,฀and฀a฀number฀ of฀other฀Neotropical฀genera฀with฀imbricate฀sepals,฀one฀basal,฀lateral,฀or฀apical฀ ovule,฀and฀apocarpous฀fruits,฀in฀the฀SBC. Phylogenetic฀ analyses฀ of฀ Annonaceae฀ presented฀ by฀ Richardson฀ et฀ al.฀ (2004)฀and฀Pirie฀et฀ al.฀(this฀thesis,฀Chapter฀2)฀indicate฀that฀Cremastosper ma,฀ Klarobelia,฀Malmea฀and฀Mosannona฀are฀likely฀to฀represent฀monophyletic฀groups.฀ However,฀the฀former฀study฀included฀limited฀taxon฀sampling฀within฀the฀four฀ genera,฀and฀revealed฀only฀low฀support฀for฀monophyly฀of฀Mosannona.฀The฀latter฀ study฀included฀extra฀sampling฀of฀both฀characters฀and฀taxa.฀The฀low฀support฀ for฀Mosannona฀(Richardson฀et฀ al.,฀2004)฀is฀probably฀a฀result฀of฀insufficient฀ character฀sampling.฀Increased฀sampling฀of฀species฀within฀these฀genera฀and฀their฀ closest฀relatives,฀including฀taxa฀distributed฀outside฀the฀‘Andean-centred’฀regions฀ is฀necessary. The SBC phylogeny: Pantropical disjunct distribution patterns and their implications for biogeographic hypotheses in the Neotropics Recent฀studies฀of฀angiosperm฀groups,฀including฀Annonaceae,฀previously฀ regarded฀as฀Gondwanan฀distributed,฀have฀demonstrated฀using฀molecular฀dating฀ techniques฀that฀these฀distributions฀may฀have฀originated฀later฀than฀the฀estimated฀ timing฀of฀separation฀of฀its฀constituent฀continents฀(Renner฀et฀ al.,฀2001;฀Davis฀ et฀ al.,฀2002;฀Richardson฀et฀ al.,฀2004).฀A฀combination฀of฀the฀presence฀of฀land฀ connections฀and฀of฀suitable฀climatic฀conditions฀made฀a฀number฀of฀dispersal฀ paths฀between฀currently฀isolated฀tropical฀zones฀available฀during฀periods฀since฀ the฀ break฀ up฀ of฀ Gondwana฀ (Morley,฀ 2003;฀ Pennington฀ &฀ Dick,฀ 2004).฀The฀ 44 • Chapter 3 �iliusoid clade �o�o�arp�a ��aro�e��a �se��ep�e�ra�t��s �osa��o�a Cremastosperma ��o�ops�s clade SBC �se��o�a��ra �a�mea �o��a�t��a ��po�e��a complex ��ptost��ma � �ree��a�o�e��ro� �������a LBC Annonaceae Ca�a��a � Ambavioids ��a�a�orea �upomatiaceae other �agnoliales outgroups �aurales F i g . 2 . Phylogeny of the Annonaceae. Summary of maximum parsimony 50% bootstrap consensus topology adapted from Richardson et al. (2004). view฀that฀since฀its฀split฀from฀Africa฀around฀100฀mya฀the฀geological฀history฀of฀ South฀America฀is฀that฀of฀an฀island฀continent,฀only฀recently฀connected฀to฀North฀ America฀via฀the฀Panamanian฀isthmus฀in฀the฀Pliocene฀(Burnham฀&฀Graham,฀ 1999)฀may฀thus฀underestimate฀the฀role฀of฀dispersal฀in฀the฀origin฀of฀pan-tropical฀ disjunctions฀(Morley,฀2003). One฀proposed฀explanation฀for฀the฀pantropical฀distribution฀of฀clades฀within฀ Annonaceae฀is฀through฀dispersal฀across฀the฀‘boreotropics’.฀This฀dispersal฀route,฀ unlike฀ the฀ presumption฀ of฀ a฀ common฀ ancestor฀ on฀ Gondwana,฀ would฀ not฀ necessarily฀involve฀an฀ancestral฀area฀in฀South฀America.฀The฀phylogeny฀of฀the฀ Biogeography of Andean-centred Annonaceae • 45 SBC฀and฀positions฀of฀Cremastosper ma,฀Klarobelia,฀Malmea,฀and฀Mosannona฀ within฀ it฀ has฀ important฀ implications฀ for฀ possible฀ reconstructions฀ of฀ their฀ biogeographical฀ history.฀ Levels฀ of฀ resolution฀ on฀ the฀ basis฀ of฀ rbcL฀ and฀ tr nLF฀ do฀ not฀ exclude฀ the฀ possibility฀ that฀ most฀ of฀ the฀ Neotropical฀ SBC฀ genera฀ together฀ could฀ represent฀ a฀ monophyletic฀ group฀ (Richardson฀ et฀ al.,฀ 2004).฀ Should฀in฀particular฀the฀Cremastosperma฀and฀Malmea฀clades฀be฀demonstrated฀ to฀ be฀ sister฀ to฀ other฀ South฀American฀ centred฀ clades฀ this฀ would฀ indicate฀ a฀ common฀ geographical฀ origin฀ (in฀ the฀ Neotropics)฀ dating฀ back฀ at฀ least฀ as฀ far฀ as฀ the฀ MRCA฀ of฀ those฀ groups.฀ However,฀ should฀ such฀ a฀ Neotropical฀ clade฀ share฀a฀most฀recent฀last฀common฀ancestor฀with฀a฀(largely)฀Asian฀clade,฀then฀its฀ Neotropical฀distribution฀could฀theoretically฀be฀as฀recent฀as฀the฀age฀of฀its฀crown฀ group.฀The฀origin฀of฀its฀constituent฀species฀in฀South฀America฀would฀then฀need฀ to฀be฀reassessed. The฀ importance฀ of฀ this฀ question฀ for฀Andean฀ biogeography฀ is฀ that฀ it฀ is฀ currently฀ impossible฀ to฀ reject฀ the฀ possibility฀ that฀ Cremastosper ma฀ and/or฀ Malmea฀originated฀in฀different฀geographic฀areas฀to฀Klarobelia฀or฀Mosannona:฀ within฀ or฀ without฀ the฀ South฀American฀ continent,฀ with฀ recent฀ dispersal฀ to฀ South,฀or฀Central฀America฀followed฀by฀radiation.฀The฀similarity฀in฀their฀current฀ distribution฀patterns฀might฀not฀reflect฀common฀biogeographic฀history,฀and฀the฀ explanation฀of฀Gentry฀for฀their฀diversification,฀i.e.฀in฀response฀to฀the฀Andean฀ orogeny,฀ might฀ even฀ fall฀ outside฀ the฀ time฀ frame฀ of฀ their฀ presence฀ in฀ South฀ America. Aims: Gentry’s฀explanation฀for฀the฀distribution฀of฀Andean-centred฀taxa฀was฀that฀ they฀1)฀originated฀in฀South฀America฀(i.e.฀Gondwanan)฀and฀2)฀were฀subject฀to฀ a฀recent฀burst฀of฀speciation฀as฀a฀result฀of฀the฀orogeny฀of฀the฀Andean฀mountain฀ chain.฀This฀ represents฀ two฀ hypotheses฀ for฀ the฀ origins฀ of฀ Cremastosper ma,฀ Klarobelia,฀Malmea฀and฀Mosannona฀which฀can฀be฀tested฀using฀phylogenetic฀ reconstruction฀ and฀ molecular฀ dating฀ techniques.฀The฀ aims฀ of฀ this฀ study฀ are฀ therefore฀firstly฀to฀determine฀the฀sister฀groups฀of฀the฀four฀genera฀and฀the฀age฀of฀ these฀sister฀groups.฀This฀may฀help฀us฀determine฀the฀geographic฀origin฀of฀these฀ Andean฀centred฀genera.฀Secondly,฀we฀aim฀to฀test฀the฀monophyly฀of฀each฀of฀ these฀genera฀and฀also฀to฀determine฀the฀age฀of฀the฀crown฀group฀of฀each฀genus.฀ This฀ may฀ help฀ us฀ to฀ determine฀ whether฀ they฀ have฀ congruent฀ evolutionary฀ histories฀that฀may฀all฀have฀been฀affected฀by฀the฀Andean฀orogeny.฀Should฀the฀ similar฀distribution฀patterns฀of฀these฀Andean-centred฀genera฀in฀Annonaceae฀ be฀due฀to฀a฀common฀biogeographical฀history,฀then฀reconstructing฀that฀history฀ might฀ also฀ offer฀ insight฀ into฀ the฀ origins฀ of฀ the฀ high฀ diversity฀ of฀ other฀ taxa฀ in฀ north฀ western฀ South฀ America.฀ Alternatively,฀ the฀ species฀ represented฀ by฀ these฀Annonaceae฀genera฀might฀have฀originated฀in฀different฀ways฀and/or฀over฀ different฀periods฀of฀time. 46 • Chapter 3 Materials and Methods Ta x o n s a m p l i n g This฀study฀largely฀utilised฀previously฀unpublished฀sequence฀data,฀as฀well฀ as฀published฀sequences฀(Sauquet฀et฀ al.,฀2003;฀Mols฀et฀ al.,฀2004;฀Pirie฀et฀ al.,฀ this฀thesis,฀Chapter฀2;฀Chatrou฀et฀al.,฀in฀prep.).฀Around฀half฀the฀total฀numbers฀ of฀described฀and฀un-described฀species฀for฀each฀of฀the฀four฀genera฀under฀study฀ were฀represented฀by฀fourteen฀samples฀of฀Cremastosperma,฀six฀of฀Klarobelia,฀four฀ of฀Malmea฀and฀seven฀of฀Mosannona.฀Geographical฀distribution฀of฀specimens฀ sampled฀ is฀ indicated฀ on฀ the฀ distribution฀ maps฀ (see฀ Fig.฀ 1a-d).฀A฀ total฀ of฀ 77฀ SBC฀taxa฀were฀sampled,฀including฀increased฀sampling฀of฀species฀representing฀ all฀related฀South฀American฀genera฀as฀represented฀by฀Richardson฀et฀al.฀(2004),฀ five฀Asian฀ and฀ Central฀American฀ taxa฀ representative฀ of฀ the฀ miliusoid฀ clade,฀ including฀first-branching฀lineage฀Monocarpia (Mols et฀al.,฀2004),฀three฀samples฀ representing฀the฀Asian฀‘Polyalthia฀ hypoleuca฀complex’฀clade฀(Rogstad,฀1989;฀ Mols฀et฀al.,฀2004)฀and฀eight฀representing฀the฀African฀genera฀Greenwayodendron,฀ Piptostigma,฀and฀Annic kia,฀first฀branching฀lineages฀of฀the฀SBC฀(Mols฀et฀ al.,฀ 2004;฀Richardson฀et฀ al.,฀2004).฀Major฀clades฀within฀the฀LBC฀and฀the฀firstbranching฀ lineages฀ of฀ Annonaceae,฀ the฀‘Ambavioid’฀ clade฀ and฀ Anaxagorea฀ (Doyle฀ &฀ Le฀Thomas,฀ 1996;฀ Richardson฀ et฀ al.,฀ 2004)฀ were฀ represented฀ by฀ sixteen฀accessions.฀Outgroups฀were฀selected฀from฀other฀families฀of฀the฀order฀ Magnoliales;฀ Magnolia฀ and฀ Lir iodendron฀ (Magnoliaceae)฀ and฀ Coelocaryon฀ (Myristicaceae).฀Genbank฀accession฀numbers฀and฀voucher฀details฀are฀presented฀ in฀Appendix฀A.฀ Character sampling For฀all฀96฀accessions฀the฀chloroplast฀DNA฀markers฀rbcL,฀tr nL-tr nF฀and฀ psbA-tr nH฀were฀sampled฀(Matrix฀1:฀see฀Appendix฀A),฀and฀for฀23฀of฀these฀96฀ accessions฀the฀additional฀markers,฀matK,฀ndhF,฀trnT-trnL,฀trnS-trnG฀and฀atpBrbcL฀(Matrix฀2:฀see฀Appendix฀A).฀LBC฀accessions฀were฀excluded฀in฀Matrix฀2,฀ as฀ high฀ sequence฀ divergence฀ within฀ this฀ clade฀ made฀ homology฀ assessment฀ in฀ the฀alignment฀of฀non-coding฀markers฀ambiguous.฀Having฀confirmed฀the฀best฀ outgroups฀for฀the฀SBC฀in฀Matrix฀1,฀both฀alignment฀problems฀and฀sequencing฀ effort฀were฀minimised฀by฀the฀exclusion฀of฀more฀distant฀outgroups฀in฀Matrix฀2. DNA extraction, PCR amplification and sequencing Total฀ genomic฀ DNA฀ was฀ extracted฀ using฀ a฀ modified฀ cetyl฀ trimethyl฀ ammonium฀bromide฀(CTAB)฀method฀(Doyle฀&฀Doyle,฀1987):฀50฀mg฀silica฀dried฀ or฀herbarium฀leaf฀material฀was฀homogenised฀in฀1300฀ μl฀CTAB฀and฀incubated฀ for฀20฀minutes฀with฀12฀μl฀2-mercaptoethanol฀at฀65°C,฀followed฀by฀90฀minutes฀ ambient฀mixing฀with฀1฀ml฀24:1฀chloroform:฀isoamylalcohol.฀After฀10฀minutes฀ centrifuged฀at฀13,000฀rpm,฀300฀μl฀supernatant฀was฀purified฀using฀Wizard฀DNA฀ Biogeography of Andean-centred Annonaceae • 47 purification฀system฀(Promega฀corp.)฀(i.e.฀without฀isopropanol฀precipitation,฀ avoiding฀the฀co-precipitation฀of฀oxidised฀material;฀Savolainen฀et฀al.,฀1995).฀ PCR฀amplification฀conditions฀were฀modified฀depending฀on฀the฀qualities฀ of฀the฀DNA฀sample฀available.฀Samples฀extracted฀from฀herbarium฀material฀often฀ contain฀lower฀quantities฀of฀more฀fragmented฀DNA,฀and฀higher฀levels฀of฀PCR฀ inhibiting฀compounds฀(Savolainen฀et฀ al.,฀1995).฀In฀most฀cases฀the฀rbcL฀gene฀ was฀amplified฀in฀two฀pieces฀and฀sequenced฀using฀primers฀1F/724R฀(Olmstead฀ et฀ al.,฀ 1992)฀ and฀ 636F/1460R฀ (Fay฀ et฀ al.,฀ 1997;฀ Fay฀ et฀ al.,฀ 1998).฀Where฀ amplification฀was฀unsuccessful฀further฀internal฀primers,฀217F,฀922F,฀536R฀and฀ 1104R฀(Pirie฀et฀ al.,฀this฀thesis,฀Chapter฀2),฀and฀376R:฀5’-GGGTTCAAAGC TCTACGAGCTCTACG-3’฀and฀444F:฀5’-GGTCCGCCCCATGGCATCC-3’฀ were฀applied฀in฀combination฀1F/536R,฀217F/724R฀(or฀1F/376R,฀217F/536R฀ and฀444F/724R),฀636F/1104R฀and฀922F/1460R฀to฀amplify฀the฀gene฀in฀up฀ to฀ five฀ overlapping฀ pieces฀ of฀ between฀ 300฀ and฀ 500฀ bp฀ long.฀ Plant฀ universal฀ primers฀of฀Taberlet฀et฀al.฀(1991)฀were฀used฀to฀amplify฀separately฀and฀sequence฀ the฀trnL฀intron฀(primers฀C/D)฀and฀trnL-trnF฀spacer฀(primers฀E/F).฀The฀psbAtrnH฀intergenic฀spacer฀was฀amplified฀and฀sequenced฀using฀primers฀psb฀A฀and฀ trn฀H฀(GUG)฀(Hamilton,฀1999).฀Partial฀matK฀sequences฀were฀amplified฀using฀ primers฀390F฀and฀1326R฀(Cuénoud฀et฀al.,฀2002),฀and฀MintF฀and฀MintR฀(Pirie฀ et฀ al.,฀this฀thesis,฀Chapter฀2),฀in฀combination฀390F/1326R฀or฀390F/MintR฀ and฀MintF/1326R฀and฀sequenced฀using฀primers฀390F฀and฀1326R.฀The฀ndhF฀ gene฀ was฀ amplified฀ and฀ sequenced฀ in฀ two฀ overlapping฀ pieces฀ using฀ primers฀ 1,฀972฀and฀2110R฀(Olmstead฀&฀Sweere,฀1994)฀and฀1165R฀(Kim฀et฀ al.,฀2001)฀ in฀combination฀1/1165R฀and฀972/2110R.฀The฀ tr nT-tr nL฀intergenic฀spacer฀ was฀amplified฀using฀primers฀A฀and฀B฀(Taberlet฀et฀ al.,฀1991),฀AintFCrem:฀5’CCGTTCCGGTATTCCAAATCGAGC-3’฀and฀ABintR:5’-CGTTGATGTAT CCGCAATTCAATATG-3’฀in฀combination฀A/B฀or฀A/BintR฀and฀AintFCrem/ B฀ and฀ sequenced฀ using฀ primers฀A฀ and฀ B.฀The฀ tr nS-tr nG฀ intergenic฀ spacer฀ was฀amplified฀and฀sequenced฀using฀primers฀trn฀S฀(GCU)฀and฀trn฀G฀(UCC)฀ (Hamilton,฀1999)฀and฀the฀atpB-rbcL฀intergenic฀spacer฀using฀primers฀atprbc3฀ (complimentary฀to฀S20฀of฀Hoot฀et฀ al.,฀1995)฀and฀atprbc2฀(Scharaschkin฀and฀ Doyle,฀pers.฀com.). A฀standard฀PCR฀protocol฀was฀used฀throughout,฀with฀the฀addition฀of฀1฀µl฀ 0.4฀%฀BSA฀per฀25฀ μl฀reaction฀(which฀was฀found฀to฀increase฀amplification฀in฀ all฀samples),฀35฀cycles฀of฀30฀sec.:฀94°C;฀1฀min.:฀55°C;฀2฀min.:฀72°C,฀with฀an฀ initial฀4฀min.:฀94°C฀and฀final฀7฀min.:฀72°C.฀PCR฀products฀were฀purified฀using฀ QIAquick฀PCR฀purification฀kits฀(Qiagen),฀sequenced฀with฀the฀PCR฀primers,฀ and฀analysed฀by฀electrophoresis฀using฀an฀automatic฀sequencer฀ABI฀3730XL. Phylogenetic Analysis DNA฀ sequences฀ were฀ edited฀ in฀ SeqMan฀ 4.0฀ (DNAStar฀ Inc.,฀ Madison,฀ WI)฀ and฀ aligned฀ manually,฀ resulting฀ in฀ alignments฀ of฀ 1496฀ positions฀ (rbcL,฀ 48 • Chapter 3 including฀a฀34-41฀aligned฀positions฀long฀non-coding฀region฀on฀the฀3’฀end),฀ 1402฀ positions฀ (tr nL-F),฀ 798฀ positions฀ (psbA-tr nH),฀ 843฀ positions฀ (matK),฀ 2043฀positions฀(ndhF),฀1157฀positions฀(trnT-trnL),฀953฀positions฀(trnS-trnG)฀ and฀828฀positions฀(atpB-rbcL).฀Areas฀of฀the฀alignments฀where฀the฀assessment฀of฀ homology฀was฀ambiguous฀were฀excluded฀from฀the฀analyses.฀ Gaps฀ in฀ the฀ alignments฀ were฀ coded฀ as฀ present/absent฀ characters฀ where฀ they฀ could฀ be฀ coded฀ unambiguously,฀ following฀ Simmons฀ and฀ Ochoterena฀ (2000).฀Two฀excluded฀regions,฀one฀in฀psbA-tr nH,฀the฀other฀in฀tr nT-tr nL,฀of฀ 15฀and฀12฀positions฀respectively,฀appeared฀to฀represent฀inversions,฀with฀around฀ half฀the฀accessions฀possessing฀almost฀exact฀reverse-compliment฀sequences฀of฀ the฀others.฀Under฀the฀assumption฀that฀these฀inversions฀had฀occurred฀with฀high฀ frequency,฀ the฀ bases฀ in฀ one฀ version฀ were฀ aligned฀ with฀ those฀ of฀ the฀ reverse฀ compliment฀ of฀ the฀ other.฀These฀ characters฀ displayed฀ little฀ or฀ no฀ homoplasy฀ when฀optimised฀onto฀the฀bootstrap฀topologies฀and฀were฀therefore฀presumed฀to฀ contain฀phylogenetic฀signal฀and฀included฀in฀further฀analyses. Maximum฀ parsimony฀ (MP)฀ analysis:฀ Data฀ were฀ analysed฀ using฀ the฀ parsimony฀algorithm฀of฀the฀software฀package฀PAUP*฀4.0b10฀(Swofford,฀2000),฀ under฀ the฀ equal฀ and฀ unordered฀ weights฀ criterion฀ (Fitch฀ parsimony;฀ Fitch,฀ 1971).฀The฀length฀of฀the฀shortest฀trees฀were฀estimated฀for฀Matrix฀1฀using฀the฀ parsimony฀ratchet฀(Nixon,฀1999)฀as฀implemented฀using฀PAUP*฀and฀PAUPRat฀ (Sikes฀&฀Lewis,฀2001).฀All฀shortest฀trees฀were฀calculated฀using฀the฀‘branch฀and฀ bound’฀method฀for฀Matrix฀2,฀and฀Bremer฀support฀(Bremer,฀1994)฀estimated฀ using฀the฀program฀TreeRot฀(Sorenson,฀1999).฀Support฀was฀also฀estimated฀using฀ bootstrap฀analyses฀of฀500฀replicates฀with฀‘full’฀heuristic฀searches฀of฀50฀random฀ addition฀ sequences,฀TBR,฀ saving฀ 50฀ trees฀ each฀ time.฀ Bootstrap฀ percentages฀ were฀interpreted฀following฀Richardson฀et฀al.฀(2004):฀50-74%฀represents฀weak฀ support,฀ 75-84%฀ moderate฀ support฀ and฀ 85-100%฀ strong฀ support.฀ Bootstrap฀ analyses฀ were฀ performed฀ on฀ Matrix฀ 2฀ with฀ and฀ without฀ the฀ rbcL฀ sequence฀ data฀(see฀Results).฀ Selecting฀the฀best฀fitting฀DNA฀substitution฀model:฀ModelTest฀3.06฀(Posada฀ &฀Crandall,฀1998)฀was฀used฀to฀select฀the฀substitution฀model฀best฀fitting฀each฀ sequence฀data฀partition,฀and฀the฀combined฀sequence฀data฀of฀Matrix฀1,฀using฀an฀ arbitrary฀most฀parsimonious฀tree฀topology฀as฀estimated฀above฀for฀Matrix฀1฀and฀ the฀most฀parsimonious฀topology฀for฀Matrix฀2. Bayesian฀ analysis:฀ The฀ combined฀ datasets฀ were฀ also฀ analysed฀ using฀ Bayesian฀ inference,฀ as฀ implemented฀ in฀ MrBayes฀ version฀ 3.0฀ (Huelsenbeck,฀ 2000).฀The฀data฀was฀partitioned฀according฀to฀the฀separate฀markers฀used฀and฀ both฀rates฀and฀substitution฀models฀were฀allowed฀to฀vary฀across฀the฀partitions.฀ Prior฀values฀for฀the฀DNA฀substitution฀models฀were฀applied฀to฀each฀partition฀ (as฀determined฀using฀ModelTest฀above).฀Prior฀probabilities฀for฀all฀topologies฀ were฀ equal.฀ Coelocaryon฀ preussii฀ (Myristicaceae,฀ sister฀ group฀ to฀ rest฀ of฀ Magnoliales;฀Sauquet฀et฀ al.,฀2003)฀was฀chosen฀as฀the฀single฀outgroup฀taxon฀ Biogeography of Andean-centred Annonaceae • 49 permitted฀by฀MrBayes฀in฀Matrix฀1,฀Cleistopholis฀glauca฀as฀outgroup฀for฀Matrix฀ 2.฀MCMC฀analyses฀were฀run฀for฀5,000,000฀generations฀with฀four฀simultaneous฀ MCMC฀chains฀to฀calculate฀posterior฀probabilities฀(PP)฀and฀one฀tree฀per฀100฀ generations฀was฀saved.฀The฀burn-in฀values฀were฀determined฀empirically฀from฀ the฀likelihood฀values฀and฀50%฀majority฀rule฀consensus฀trees฀calculated฀together฀ with฀approximations฀of฀the฀PP฀for฀the฀observed฀bipartitions.฀ Molecular dating Topology:฀Nodes฀present฀in฀the฀Bayesian฀consensus฀of฀Matrix฀1฀and฀not฀ contradicted฀by฀results฀from฀Matrix฀2฀were฀used฀to฀constrain฀two฀further฀MP฀ searches฀of฀the฀sequence฀data฀of฀Matrix฀1฀(‘full’฀heuristic,฀100฀random฀taxon฀ addition฀ sequences,฀TBR,฀ saving฀ maximum฀ of฀ 50฀ shortest฀ trees฀ each฀ time),฀ from฀which฀single฀arbitrary฀most฀parsimonious฀topologies฀were฀selected.฀The฀ two฀searches฀included฀(A)฀all฀taxa฀(96฀in฀total),฀and฀(B)฀all฀taxa฀minus฀nine฀ accessions฀of฀Cremastosper ma,฀three฀of฀Mosannona,฀and฀three฀of฀Klarobelia,฀ thus฀ leaving฀ each฀ of฀ these฀ clades,฀ and฀ that฀ of฀ Malmea,฀ represented฀ by฀ four฀ accessions฀(including฀those฀representing฀first฀branching฀lineages,฀thus฀ensuring฀ that฀the฀crown฀nodes฀remained฀comparable),฀in฀order฀to฀explore฀possible฀bias฀ in฀age฀estimations฀according฀to฀numbers฀of฀taxa฀sampled. Fossil฀ calibration:฀ The฀ oldest฀ unambiguously฀ identifiable฀ fossil฀ Annonaceae฀ remains฀ have฀ been฀ found฀ in฀ the฀ Maastrichtian฀ of฀ Nigeria฀ (seeds฀ with฀ perichalazal฀ ring฀ and฀ ruminate฀ endosperm;฀ Chesters,฀ 1955)฀ and฀ Colombia฀(reticulate฀monosulcate฀pollen;฀Sole฀de฀Porta,฀1971).฀However,฀their฀ precise฀ placement฀ on฀ the฀Annonaceae฀ phylogeny฀ is฀ ambiguous,฀ due฀ to฀ the฀ reticulate฀nature฀of฀the฀characters฀available฀for฀their฀identification.฀The฀fossil฀ taxon฀ Arc haeanthus฀ (Dilcher฀ &฀ Crane,฀ 1984)฀ was฀ used฀ following฀ Doyle฀ et฀ al.฀(2004)฀and฀Richardson฀et฀ al.฀(2004)฀to฀assign฀a฀minimum฀age฀of฀98฀mya฀ to฀ the฀ stem฀ node฀ of฀ Magnoliaceae฀ (due฀ to฀ the฀ distinctive฀ stipules,฀ elongate฀ receptacle฀ and฀ fruits).฀ This฀ interpretation฀ is฀ not฀ entirely฀ uncontroversial:฀ Arc haeanthus฀has฀recently฀been฀explicitly฀excluded฀from฀age฀estimations฀in฀ angiosperms฀ by฀ Crepet฀ et฀ al.฀ (2004).฀ Crepet฀ et฀ al.฀ instead฀ used฀ two฀ fossil฀ flowers,฀Cronquistiflora฀and฀Detrusandra฀(Crepet฀&฀Nixon,฀1998),฀to฀impose฀ the฀more฀conservative฀minimum฀age฀of฀90฀mya฀on฀the฀Magnoliales.฀However,฀ results฀of฀Doyle฀et฀al.฀(2004)฀and฀Richardson฀et฀al.฀(2004)฀broadly฀agreed฀with฀ ages฀estimated฀in฀angiosperm฀wide฀studies฀(e.g.฀Wikström฀et฀al.,฀2001;฀Davies฀ et฀ al.,฀2004),฀which฀suggest฀the฀calibration฀of฀Crepet฀et฀ al.฀may฀represent฀a฀ (greater)฀underestimation฀of฀the฀true฀age. In฀order฀to฀test฀whether฀the฀sequence฀data฀of฀Matrix฀1฀exhibited฀clocklike฀ behaviour,฀ a฀ likelihood฀ ratio฀ test฀ was฀ performed฀ on฀ the฀ first฀ of฀ the฀ above฀ (constrained)฀most฀parsimonious฀tree฀topologies.฀Likelihood฀of฀the฀data฀with฀ and฀without฀constraint฀of฀a฀molecular฀clock,฀were฀calculated฀and฀the฀likelihood฀ 50 • Chapter 3 ratio฀statistic฀compared฀with฀ χ 2฀critical฀value฀with฀94฀degrees฀of฀freedom฀(i.e.,฀ number฀of฀taxa฀minus฀2).฀ Molecular฀ dating฀ using฀ NPRS฀ and฀ PL:฀This฀substitution฀model฀selected฀ using฀ModelTest฀was฀used฀to฀calculate฀branch฀lengths฀for฀the฀above฀topologies฀ based฀ on฀ the฀ original฀ data฀ using฀ the฀ maximum฀ likelihood฀ (ML)฀ criterion฀ as฀ implemented฀ in฀ PAUP*.฀ Confidence฀ limits฀ on฀ branch฀ lengths,฀ reflecting฀ stochasticity฀in฀the฀sampling฀of฀character฀changes฀(‘substitutional฀noise’),฀were฀ estimated฀ by฀ 100฀ replicates฀ of฀ bootstrap฀ re-sampling฀ (as฀ also฀ described฀ in฀ Wikström฀et฀al.,฀2001),฀with฀subsequent฀ML฀branch฀length฀estimation฀on฀the฀ constrained฀tree฀topology฀for฀each฀bootstrap฀replicate.฀This฀resulted฀in฀100฀trees฀ comprising฀a฀range฀of฀estimated฀lengths฀for฀each฀branch฀of฀the฀topology.฀ Thereafter,฀Sanderson’s฀methods฀of฀nonparametric฀rate฀smoothing฀(NPRS)฀ (Sanderson,฀ 1997)฀ and฀ penalized฀ likelihood฀ (PL)฀ (Sanderson,฀ 2002a)฀ were฀ applied฀ as฀ implemented฀ in฀ the฀ software฀ package฀ r8s฀ (Sanderson,฀ 2002b)฀ in฀ order฀to฀estimate฀divergence฀times.฀Divergence฀times฀were฀estimated฀for฀nodes฀ representing฀the฀MRCAs฀(i.e.฀the฀crown฀groups)฀of฀each฀of฀the฀four฀genera,฀ and฀those฀representing฀the฀MRCAs฀of฀a฀number฀of฀further฀SBC฀sub-clades฀ (see฀Table฀3),฀with฀the฀Magnoliaceae฀stem฀lineage฀(i.e.฀the฀root฀node,฀once฀ the฀initial฀outgroup฀has฀been฀pruned฀out฀in฀the฀analyses)฀fixed฀at฀98฀million฀ years฀old.฀Analyses฀were฀performed฀on฀the฀trees฀with฀branch฀lengths฀derived฀ from฀the฀original฀data,฀and฀for฀each฀of฀the฀100฀trees฀each฀with฀branch฀lengths฀ derived฀from฀bootstrap฀re-sampled฀data.฀The฀results฀of฀the฀latter฀analyses฀were฀ summarised฀giving฀mean฀values฀with฀standard฀deviation฀for฀specified฀nodes฀ using฀the฀‘profile’฀command฀in฀r8s. Molecular฀ dating฀ using฀ a฀ Bayesian฀ technique:฀Bayesian฀molecular฀dating฀ was฀performed฀following฀Rutschmann฀(2004),฀Renner฀(2004),฀and฀the฀PAML฀ (Yang,฀1997)฀and฀multidivtime฀(Thorne฀&฀Kishino,฀2002)฀manuals.฀Nucleotide฀ substitutions฀ in฀ the฀ combined฀ sequence฀ data฀ were฀ estimated฀ using฀ PAML’s฀ ‘baseml’฀program฀and฀the฀F84฀+฀ Γ฀model฀(with฀five฀rate฀categories),฀with฀the฀ single฀topology฀as฀above.฀Using฀the฀multidivtime฀package,฀each฀baseml฀output฀ was฀converted฀using฀‘pamlmodelinf ’฀for฀input฀in฀‘estbranches’,฀to฀estimate฀branch฀ lengths฀and฀calculate฀the฀variance-covariance฀structure฀of฀those฀estimates.฀These฀ were฀then฀used฀as฀input฀for฀‘multidivtime’฀to฀calculate฀node฀divergence฀times฀ with฀the฀following฀settings:฀100,000฀generations฀of฀the฀Markov฀chain,฀sampled฀ every฀10฀after฀a฀burn฀in฀of฀10,000.฀Prior฀number฀of฀time฀units฀between฀tip฀and฀ root:฀98฀(i.e.฀millions฀of฀years),฀SD฀of฀prior:฀98,฀prior฀rate฀at฀root฀node:฀0.0003฀ (derived฀from฀r8s฀results฀using฀PL),฀nu:฀1,฀SD฀nu:฀1,฀and฀the฀single฀constraint฀on฀ node฀times฀was฀the฀same฀as฀the฀calibration฀in฀the฀r8s฀analyses฀as฀above.฀ Biogeography of Andean-centred Annonaceae • 51 Results Matrix 1 Both฀100฀and฀1000฀iterations฀of฀the฀parsimony฀ratchet฀recovered฀trees฀of฀ 2137฀steps,฀CI฀=0.648,฀RI฀=0.777.฀Sequence฀and฀alignment฀lengths,฀numbers฀ of฀variable฀and฀parsimony฀informative฀characters฀and฀tree฀statistics฀for฀Matrix฀ 1฀are฀presented฀in฀Table฀1.฀The฀marker฀rbcL฀provided฀the฀lowest฀total฀number฀ of฀ parsimony฀ informative฀ characters,฀ despite฀ representing฀ higher฀ numbers฀ of฀ sequenced฀ bases,฀ and฀ exhibited฀ significantly฀ higher฀ levels฀ of฀ homoplasy฀ compared฀with฀the฀other฀two฀markers. Ta b l e 1 : D e t a i l s o f M a t r i x 1 , m a x i m u m p a r s i m o n y s e a r c h a n d b e s t f i t t i n g substitution models Marker Sequence Alignment length length Variable Pars. inf. characters characters Pars. d CI/RI Model inf. indels b 1470-1480 r��L 912-996 � tr�L- 1496 a 310 187 0 0�519/0�712 ��R+I+Γ 1402 454 270 30 0�729/0�810 K81uf+Γ 798 340 215 10 0�689/0�812 �V�+Γ tr�F �c 274� � ps��- 412-511 tr�H� a Alignment length of Chatrou et al.; b Including 3’ non coding region; c Annona muricata 0525: large deletion; d Optimised onto the combined topology The฀ best฀ fitting฀ substitution฀ model฀ for฀ all฀ sequence฀ data฀ of฀ matrix฀ 1฀ estimated฀ using฀ ModelTest฀ was฀ GTR+I+ Γ.฀ Best฀ fitting฀ models฀ for฀ each฀ marker฀ individually฀ are฀ presented฀ in฀ Table฀ 1.฀ MP฀ bootstrap฀ analysis฀ and฀ Bayesian฀inference฀of฀Matrix฀1฀resulted฀in฀congruent฀consensus฀topologies,฀ that฀of฀Bayesian฀inference฀being฀significantly฀more฀resolved.฀Both฀results฀are฀ summarised฀in฀Fig.฀3,฀with฀bootstrap฀support฀(BS)฀values฀below฀the฀nodes฀and฀ posterior฀ probabilities฀ above.฀ Monophyly฀ of฀ Cremastosper ma,฀ Malmea,฀ and฀ Mosannona฀were฀confirmed฀by฀100%฀BS.฀A฀clade฀including฀all฀accessions฀of฀ Klarobelia฀plus฀Pseudephedranthus฀received฀90%฀BS.฀Further฀clades฀supported฀ by฀>50%฀BS฀correspond฀largely฀to฀those฀revealed฀by฀Richardson฀et฀al.฀(2004),฀ except฀for฀a฀clade฀including฀all฀the฀‘South฀American฀centred’฀SBC฀taxa฀(except฀ a฀ single฀ accession,฀ unidentified฀ to฀ genus,฀ falling฀ with฀ 100%฀ BS฀ within฀ the฀ miliusoid/Monocarpia฀clade)฀with฀63%฀BS.฀This฀will฀be฀further฀referred฀to฀as฀ the฀‘SAC฀clade’.฀Sampling฀of฀taxa฀within฀the฀miliusoid฀clade฀was฀not฀sufficient฀ 52 • Chapter 3 1�00 100 0�58 �50 0�99 86 0�99 100 0�99 74 0�99 100 1�00 100 0�99 100 0�99 100 0�99 100 0�99 94 0�60 74 0�92 58 0�99 74 0�96 64 0�98 63 0�99 90 0�69 �50 0�99 71 0�99 61 0�99 81 0�99 91 0�99 82 0�99 77 0�99 96 1�00 92 1�00 100 0�97 86 0�99 89 0�99 100 0�99 99 0�99 94 0�99 94 0.50 �50 0�94 �50 1�00 95 0�98 84 0�99 100 0�56 �50 0�99 98 0�99 94 0�99 66 0�99 97 0�99 63 0�99 90 0�99 95 0�99 100 0�99 98 1�00 100 0�99 99 0�99 100 0�99 100 0�99 99 0�99 95 0�97 �50 0�93 69 1�00 94 0�91 88 0�99 100 0�97 �50 0�83 �50 0�99 78 0�72 �50 0�99 62 SBC 0�99 100 �BC 0�99 96 Africa 0�99 98 0�99 100 miliusoid Asia 0�99 88 0�99 100 1�00 100 0�99 99 0�97 50 SAC clade outgroups 0�99 100 0�99 100 Coe�o�ar�o��pre�ss�� L�r�o�e��ro������e�se �a��o��a��o��s ��a�a�orea�p�aeo�arpa ��a�a�orea�s���at��a Ca�a��a�o�orata C�e�stop�o��s���a��a �etramera�t��s�����e� C�m�opeta��m�sp��0098 C�m�opeta��m�sp��0203 ��atter�a�a�oma�a ��atter�a�a�stra��s ����et�a���r�sea F�saea�per���a�a �rta�otr�s��e�apeta��s ���op�a��r�tes�e�s ��ar�a������a ���o�a�m�r��ata �s�m��a�tr��o�a �������a���mmer�ae �������a����ora�t�a �������a�p��osa �ree��a�o�e��ro��o���er� �ree��a�o�e��ro��s�a�eo�e�s ��ptost��ma��as�����ata ��ptost��ma�morte�a�� ��ptost��ma�p��os�m �o��a�t��a�s�matra�a �o��a�t��a���s�o�or �o��a�t��a���a��a �o�o�arp�a�e��e�ra��Asia �nknown genus 0750 S America �o��a�t��a�s��erosa �apra�t��s���r�����or�s �esmops�s�m��ro�arpa �te�a�o�a��ostar��e�s�s �����opeta��m�per�����o �o�a�eops�s�m��t���ora �o�a�eops�s�p�e�osperma ��o�ops�s�st�p�tata ��o�ops�s�p�tt�er� ��o�ops�s�r��es�e�s �osa��o�a���s�o�or �osa��o�a�pap���osa �osa��o�a��as��e��� �osa��o�a�sp��0226 �osa��o�a��ostar��e�s�s �osa��o�a��ar�oo��� �osa��o�a�pa�����a ��a��ra�as�e���� ��a��ra�e��e�ra ��a��ra�po��a�t�a ��a��ra��o���peta�a ��a��ra�esp��ta�a ��a��ra���t��a ��a��ra��e�e��e�a�a �se��oma�mea�������a �se��oma�mea�������a ��aro�e��a�me�a�o�arpa ��aro�e��a��a�����ora ��aro�e��a�st�p�tata �se��ep�e�ra�t��s��ra�ra�s ��aro�e��a�sp��1027 ��aro�e��a��a����a ��aro�e��a������ata ����o�e��ro��o�a�e �p�e�ra�t��s�par����or�s �p�e�ra�t��s�sp��0105 �p�e�ra�t��s�sp��0284 ��a��ra��a�r��o��a ��a��ra�ma�rop����a ��a��ra�sp�aero�arpa ��a��ra����op�o��es �se��o�a��ra�sp�r�t�s-sa��t� �se��o�a��ra��a��e�s�s �se��o�a��ra��o���pes �se��o�a��ra�pa�����a �se��o�a��ra������a �se��o�a��ra�po��p��e�a �a�mea�sp��0197 �a�mea�s�r��ame�s�s �a�mea���e�s�a�a �a�mea���mera Cremastosperma��re��pes Cremastosperma��e�e��e�a��m Cremastosperma�ma��a�e�ae Cremastosperma�ma�ro�arp�m Cremastosperma�spe���� Cremastosperma��ama�a�ate�se Cremastosperma��a�����or�m Cremastosperma�me�a�op�����m Cremastosperma�m��ro�arp�m Cremastosperma�����at�m Cremastosperma��e�op�����m Cremastosperma�mo�osperm�m Cremastosperma�pe�����m basal grade 1�00 100 1�00 97 F i g . 3 . Summary of results of phylogeny reconstruction using maximum parsimony and Bayesian analysis: Matrix 1. MP bootstrap support values below the nodes, Bayesian analysis posterior probabilities above. Biogeography of Andean-centred Annonaceae • 53 to฀further฀infer฀the฀position฀of฀the฀unidentified฀accession,฀and฀it฀was฀omitted฀ from฀further฀character฀sampling.฀One฀clade,฀that฀of฀Malmea฀as฀sister฀group฀to฀ Cremastosperma,฀received฀97%฀posterior฀probability,฀but฀less฀than฀50%฀BS฀(see฀ results฀of฀Matrix฀2฀below). Matrix 2 MP฀branch฀and฀bound฀search฀of฀Matrix฀2฀excluding฀rbcL฀sequence฀data฀ resulted฀ in฀ a฀ single฀ tree฀ of฀ 1555฀ steps,฀ CI฀ =0.871,฀ RI฀ =0.791,฀ presented฀ in฀ Fig.฀4฀(with฀posterior฀probabilities฀above฀the฀nodes฀and฀BS฀values฀and฀Bremer฀ support฀below).฀Inclusion฀of฀rbcL฀resulted฀in฀3฀trees฀of฀1754฀length,฀CI฀=0.851,฀ RI฀=0.761.฀Sequence฀and฀alignment฀lengths,฀numbers฀of฀variable฀and฀parsimony฀ informative฀characters฀and฀tree฀statistics฀for฀Matrix฀2฀are฀presented฀in฀Table฀2.฀ Levels฀of฀homoplasy฀in฀rbcL฀were฀higher฀than฀those฀of฀all฀the฀other฀markers.฀ The฀CI฀values฀for฀rbcL,฀trnL-trnF฀and฀psbA-trnH฀were฀higher฀in฀Matrix฀2฀than฀ in฀Matrix฀1,฀which฀is฀to฀be฀expected฀given฀the฀lower฀taxon฀sampling฀density:฀ less฀ homoplasy฀ is฀ revealed.฀Values฀ for฀ RI฀ were฀ lower฀ in฀ all฀ three฀ markers฀ in฀ Matrix฀2,฀but฀that฀for฀rbcL฀decreased฀markedly.฀Analyses฀were฀thus฀performed฀ on฀Matrix฀2฀with฀and฀without฀the฀rbcL฀sequence฀data,฀and฀results฀compared. Ta b l e 2 : Details of Matrix 2, maximum parsimony search and best fitting substitution models Marker c f Sequence Alignment Variable Pars. inf. Pars. length length characters characters inf.indels 1496 127 59 0 0�688/0�554 �rN+I+ Γ a CI/RI Model r��L 1470-1480 mat� 843 843 142 69 0 0�881/0�833 K81uf+Γ ���F 2043 2043 370 170 0 0�840/0�763 K81uf+Γ 641-947 1157 178 68 8 0�877/0�745 �V�+Γ 1402 174 64 7 0�907/0�816 K81uf+ Γ b 798 120 50 1 0�838/0�782 K81uf+ Γ 953 134 50 0 0�890/0�769 �I�+ Γ 828 135 61 4 0�918/0�874 K81uf+I �t r��-tr�L b �t r�L-tr�F �p s��-tr�H 912-996 �e 274� a 412-511 �d 265-� �t r��-tr�� 643-851 745-770 �a tp�-r��L a Alignment length of Chatrou et al.; b Alignment/length derived from Matrix 1; c Including 3’ non coding region; d Malmea dielsiana 0260: large deletion; e Annona muricata 0525: large deletion; e Optimised onto the combined topology 54 • Chapter 3 outgroups Best฀fitting฀substitution฀models฀for฀each฀marker฀as฀estimated฀by฀ModelTest฀ are฀ presented฀ in฀Table฀ 2.฀ Bootstrap฀ analysis฀ of฀ Matrix฀ 2฀ resulted฀ in฀ a฀ fully฀ resolved฀ consensus฀ tree,฀ though฀ one฀ node฀ received฀ only฀ weak฀ support฀ (see฀ Fig.฀ 4).฀The฀ SAC฀ clade฀ received฀ strong฀ support฀ (87%฀ BS).฀ A฀ sister฀ group฀ relationship฀ between฀ Pseudoxandra฀ and฀ Cremastosper ma฀ was฀ also฀ strongly฀ supported฀(100%฀BS),฀contradicting฀the฀results฀of฀Bayesian฀(but฀not฀of฀MP)฀ analysis฀of฀Matrix฀1.฀The฀bootstrap฀consensus฀when฀including฀rbcL฀included฀ two฀polytomies.฀The฀two฀nodes฀with฀Bremer฀support฀of฀only฀1฀step฀without฀ rbcL฀(and฀subject฀to฀moderate฀to฀weak฀BS)฀were฀no฀longer฀recovered฀when฀ Ca�a��a�o�orata C�e�stop�o��s���a��a �������a����ora�t�a 100 100 / 46 100 97 / 9 ��ptost��ma�morte�a�� �o��a�t��a���s�o�or �o��a�t��a�s�matra�a �o�o�arp�a�e��e�ra 100 100 / 25 �����opeta��m�per�����o 100 100 / 28 84 75 / 1 ��o�ops�s�st�p�tata 100 100 / 52 100 87 / 3 100 76/2 94 58 / 1 100 100 / 7 �o��a�t��a�s��erosa �apra�t��s���r�����or�s��C�Am�� 100 100 / 23 99 82 / 2 �a�mea���e�s�a�a �a�mea�sp��0197 Cremastosperma��re��pes Cremastosperma��a�����or�m �se��o�a��ra������a �se��o�a��ra�sp�r�t�s-sa��t� �osa��o�a��ostar��e�s�s 100 100 / 17 100 100 / 12 �p�e�ra�t��s�sp��0105 99 85 / 3 SAC �South/Central America� 100 100 / 14 100 100 / 34 miliusoid 100 100 / 21 75 / 2 Asia 100 100 / 123 Africa �������a�p��osa �ree��a�o�e��ro��o���er� ��aro�e��a������ata �se��oma�mea�������a F i g . 4 . Summary of results of phylogeny reconstruction using maximum parsimony and Bayesian analysis: Matrix 2 (excluding rbcL sequence data). Single most parsimonious topology presented with MP bootstrap and Bremer support values below the nodes, Bayesian analysis posterior probabilities above. Biogeography of Andean-centred Annonaceae • 55 rbcL฀was฀included.฀These฀were฀the฀node฀of฀the฀Mosannona/Klarobelia฀clade฀ as฀sister฀to฀the฀Cremastosper ma/Malmea/Pseudoxandra฀clade฀(58%฀BS),฀and฀ that฀of฀the฀miliusoid/Monocarpia฀clade฀as฀sister฀to฀the฀SAC฀clade฀(75%฀BS).฀BS฀ values฀across฀the฀topology฀were฀slightly฀lower฀when฀rbcL฀was฀included฀(data฀ not฀ shown).฀ Bayesian฀ analysis฀ recovered฀ a฀ congruent฀ topology,฀ though฀ with฀ one฀polytomy:฀relationships฀between฀the฀two฀African฀clades฀and฀the฀rest฀of฀the฀ ingroup฀were฀not฀resolved.฀ Molecular dating The฀ likelihood฀ of฀ the฀ sequence฀ data฀ of฀ Matrix฀ 1฀ given฀ the฀ constraint฀ topology฀ were฀ significantly฀ different฀ according฀ to฀ clock฀ constrained฀ and฀ unconstrained฀substitution฀models฀(P<0.01),฀indicating฀rate฀heterogeneity฀i.e.฀ the฀rejection฀of฀a฀molecular฀clock.฀NPRS,฀PL฀and฀Bayesian฀(using฀multidivtime)฀ methods฀were฀thus฀applied฀to฀produce฀full฀(A)฀and฀reduced฀(B)฀taxon฀sampled฀ ultrametric฀ trees.฀The฀ cross฀ validation฀ test฀ for฀ PL฀ resulted฀ in฀ a฀ smoothing฀ parameter฀of฀31.62.฀Maximum฀ages฀for฀the฀MRCAs฀of฀the฀four฀genera฀were฀ estimated฀ at฀ ca.฀ 20-27฀ mya฀ using฀ NPRS,฀ 13-16฀ mya฀ using฀ PL,฀ and฀ 23-28฀ mya฀(95%฀probability)฀using฀multidivtime.฀The฀maximum฀age฀estimated฀for฀the฀ Asian/Neotropical฀crown฀group฀was฀ca.฀42฀mya฀(NPRS),฀37฀mya฀(PL)฀and฀51฀ mya฀(multidivtime฀-฀95%฀probability).฀Age฀estimations฀with฀standard฀deviations฀ using฀NPRS฀and฀PL,฀and฀mean฀age฀estimations฀with฀standard฀deviations฀and฀ 95%฀ confidence฀ limits฀ derived฀ using฀ multidivtime฀ are฀ presented฀ in฀Table฀ 3฀ for฀ key฀ nodes฀ referred฀ to฀ in฀ the฀ discussion฀ and฀ a฀ chronogram฀ for฀ the฀ SBC฀ (based฀on฀results฀using฀PL)฀with฀timing฀of฀geological฀events฀referred฀to฀in฀the฀ discussion฀is฀presented฀in฀Fig.฀5฀(see฀below). The฀bounds฀for฀age฀estimations,฀according฀to฀standard฀deviations฀or฀95%฀ confidence฀ limits,฀ overlapped฀ for฀ all฀ for฀ nodes฀ compared฀ across฀ the฀ three฀ different฀ methods,฀ and฀ both฀ taxon฀ selections.฀ For฀ results฀ using฀ NPRS฀ and฀ multidivtime฀all฀age฀estimations฀for฀nodes฀including฀all฀96฀taxa฀(A)฀were฀older฀ than฀those฀produced฀when฀only฀81฀were฀included฀(B).฀The฀largest฀apparent฀ difference฀was฀observed฀in฀the฀Cremastosperma฀crown฀node,฀from฀which฀clade฀ the฀largest฀number฀of฀accessions฀had฀been฀excluded฀in฀(B).฀In฀contrast,฀ages฀ produced฀for฀the฀two฀taxon฀selections฀using฀PL฀were฀almost฀identical,฀with฀no฀ consistent฀difference฀apparent.฀More฀rigorous฀tests฀are฀required฀to฀determine฀ whether฀ the฀ results฀ of฀ any฀ of฀ these฀ methods฀ are฀ sensitive฀ to฀ levels฀ of฀ taxon฀ sampling.฀ In฀ this฀ study฀ the฀ widest฀ bounds฀ of฀ the฀ age฀ estimations฀ produced฀ using฀the฀different฀methods฀are฀interpreted฀as฀the฀most฀stringent฀test฀for฀the฀ biogeographic฀hypotheses. 56 • Chapter 3 Ta b l e 3 : Ages estimations derived using DNA sequence data of Matrix 1 and the nonparametric rate smoothing (NPRS), penalized likelihood (PL) and Bayesian (Multidivtime) methods: (A) including all 97 taxa and (B) including only 4 samples each of Cremastosperma, Mosannona and Klarobelia (82 taxa in total) Cremastosperma NPRS (A) NPRS (B) PL (A) PL (B) Multidivtime (A) Multidivtime (B) 22.33 13.65 7.17 7.54 16.49 9.07 �S�D� = 4�99� Malmea Klarobelia Mosannona Mosannona west of Andes (1) �S�D� = 4�99� �S�D� = 8�53� �S�D� = 5�83� �S�D� = 5�14 �S�D� = 3�84 95% = 8�01 - 27�73� 95% = 3�45 - 18�20� 15.90 13.09 9.54 9.88 13.42 10.47 �S�D� = 4�23� �S�D� = 3�29� �S�D� = 5�35� �S�D� = 5�02� �S�D� = 4�97 �S�D� = 4�04 95% = 5�49 - 24�76� 95% = 4�28 - 19�94� 15.96 10.71 7.69 7.46 12.51 9.25 �S�D� = 3�31� �S�D� = 3�01� �S�D� = 5�20� �S�D� = 4�16� �S�D� = 4�59 �S�D� = 3�58325 95% = 5�25 - 23�12� 95% = 3�72 - 17�64� 17.38 10.22 8.62 7.88 13.58 10.18 �S�D� = 3�28� �S�D� = 2�72� �S�D� = 4�45� �S�D� = 3�77� �S�D� = 4�87 �S�D� = 3�82 95% = 6�08 - 24�85� 95% = 4�62 - 19�22� 12.67 - 6.04 - 8.76 - a a � � � � �S�D� = 3�76 95% = 3�26 - 17�78� Mosannona east of Andes (2) 11.98 - �a � 5.27 - �a � 5.97 - �S�D� = 3�19 95% = 1�35 - 13�69� Mosannona / Klarobelia clade 31.16 25.57 19.42 19.60 �S�D� = 2�95� �S�D� = 2�60� �S�D� = 8�21� �S�D� = 7�91� 26.40 22.15 �S�D� = 6�08 �S�D� = 5�40 95% = 16�13 - 39�81� 95% = 12�89 - 33�82� SAC clade 37.65 33.09 24.76 25.43 31.81 26.66 (node ‘A’) �a � �a � �a � �a � �S�D� = 6�23 �S�D� = 5�82 95% = 21�23 - 45�44� 95% = 16�21 - 39�20� 39.23 34.87 26.29 27.07 36.72 31.70 Neotropics / Asia (node ‘B’) Piptostigma / �S�D� = 3�15� 48.44 Greenwayodendron / �a � Neotropics / Asia �S�D� = 2�77� �S�D� = 9�64� �S�D� = 9�11� �S�D� = 6�60 �S�D� = 6�35 95% = 25�29 - 50�93� 95% = 20�25 - 44�94� 44.78 34.89 35.63 42.76 38.13 �a � �a � �a � �S�D� = 6�83 �S�D� = 6�64 95% = 30�81 - 56�62� 95% = 25�83 - 51�833� (node ‘C’) a No S�D� calculated for nodes subtended by �ero-length branches in any of the bootstrap replicate trees Biogeography of Andean-centred Annonaceae • 57 Discussion ‘Backbone’ phylogeny of the SBC The฀resolution฀obtained฀in฀this฀study฀suggests฀that฀sampling฀a฀larger฀number฀ of฀characters฀for฀a฀careful฀selection฀of฀placeholder฀taxa฀would฀be฀an฀efficient฀ approach฀for฀resolving฀relationships฀between฀further฀clades฀identified฀on฀the฀basis฀ of฀wider฀taxon฀sampling฀in฀Annonaceae.฀A฀number฀of฀previously฀unidentified฀ clades฀ were฀ discovered:฀ sister฀ group฀ relationships฀ between฀ Cremastosper ma฀ and฀ Pseudoxandra฀ and฀ between฀ the฀ Cremastosper ma/Pseudoxandra฀ clade฀ and฀Malmea,฀and฀the฀SAC฀clade.฀The฀latter฀result฀in฀particular฀has฀important฀ implications:฀ in฀ the฀ first฀ instance,฀ optimisation฀ of฀ ancestral฀ areas฀ results฀ in฀ the฀ stem฀ nodes฀ of฀ each฀ of฀ the฀ clades฀ forming฀ the฀ SAC฀ clade฀ resolving฀ as฀ Neotropical.฀Consequently,฀the฀minimum฀age฀which฀can฀be฀inferred฀for฀the฀ presence฀of฀their฀ancestral฀lineages฀in฀the฀Neotropics฀is฀greatly฀constrained.฀ However,฀despite฀the฀large฀amount฀of฀DNA฀sequence฀data฀analysed,฀the฀ numbers฀of฀characters฀supporting฀the฀‘backbone’฀nodes฀were฀low.฀Two฀nodes฀ in฀particular฀are฀subject฀to฀weak฀or฀moderate฀support฀and฀no฀longer฀recovered฀ when฀rbcL฀was฀included฀in฀the฀analyses:฀the฀node฀of฀the฀Mosannona/Klarobelia฀ clade฀as฀sister฀to฀the฀Cremastosperma/Malmea/Pseudoxandra฀clade฀and฀that฀of฀ the฀miliusoid/Monocarpia฀clade฀as฀sister฀to฀the฀South฀American฀centred฀clade.฀ The฀latter฀uncertainty฀effectively฀renders฀the฀optimisation฀of฀ancestral฀areas฀ for฀the฀nodes฀subtending฀the฀Asian฀and฀SAC฀clades฀ambiguous,฀increasing฀the฀ maximum฀possible฀age฀for฀the฀presence฀of฀the฀SAC฀clade฀in฀the฀Neotropics฀ (see฀below฀and฀Fig.฀5). Biogeographic history in the SBC of Annonaceae: tracking the origins of Andean-centred genera on the South American continent The฀oldest฀estimations฀of฀the฀age฀of฀the฀SBC฀crown฀group฀produced฀here฀ (58.76฀mya),฀in฀agreement฀with฀estimations฀of฀Richardson฀et฀al.฀(2004)฀(53.162.5฀mya฀s.d.฀3.6)฀were฀significantly฀younger฀than฀the฀timing฀of฀the฀AfricaSouth฀America฀break-up฀(ca.฀100฀mya;฀Burnham฀&฀Graham,฀1999).฀The฀age฀of฀ the฀MRCA฀of฀the฀SAC฀clade฀(Fig.฀5,฀indicated฀by฀an฀arrow฀and฀‘A’),฀estimated฀ at฀between฀around฀45฀and฀16฀mya฀here,฀represents฀the฀time฀after฀which฀this฀ lineage฀ can฀ be฀ said฀ with฀ certainty฀ to฀ have฀ been฀ present฀ in฀ the฀ Neotropics.฀ According฀ to฀ the฀ SBC฀ topology฀ presented฀ here,฀ the฀ actual฀ age฀ of฀ the฀ first฀ Neotropical฀ancestor฀of฀the฀SAC฀clade฀could฀date฀back฀as฀far฀as฀the฀SAC฀clade฀ stem฀node฀(Fig.฀5฀‘B’:฀51-16฀mya).฀However,฀should฀the฀miliusoid/Monocarpia฀ clade฀prove฀instead฀to฀be฀sister฀to฀the฀Polyalthia฀ hypoleuca฀clade,฀this฀would฀ push฀ this฀ age฀ back฀ as฀ far฀ as฀ that฀ of฀ the฀ Piptostigma/Greenwayodendron/฀ Neotropics/Asia฀node฀(Fig.฀5฀‘C’:฀57-26฀mya). 58 • Chapter 3 Andean orogeny �AARlandia C �iliusoid / �o�o�arp�a �o��a�t��a Asia Africa Panama isthmus �uiana �est �ast 2 ��aro�e��a 1 SAC clade A �osa��o�a Cremastosperma �a�mea B ��A -30 -20 -10 0 F i g . 5 . Chronogram: branch lengths proportional to time, as estimated using PL. Nodes A, B and C represent minimum ages of a common Neotropical ancestor of the SAC clade according to different topologies as referred to in the text. Nodes 1 and 2 represent the clades in Mosannona found west and east of the Andes respectively. Ages are subject to standard deviations (where estimated) of 4.45-9.64 my. proefschrift Pirie fig 05�ai mrt 2005 Biogeography of Andean-centred Annonaceae • 59 Neither฀of฀these฀maximum฀ages฀are฀ancient฀enough฀to฀allow฀explanation฀ of฀ the฀ distribution฀ of฀ the฀ major฀ clades฀ within฀ the฀ SBC฀ through฀ either฀ the฀ splitting฀ of฀ west฀ Gondwana฀ or฀ transatlantic฀ dispersal฀ from฀Africa฀ to฀ South฀ America฀(which฀according฀to฀Morley฀(2003)฀may฀have฀been฀possible฀across฀ island฀chains฀up฀to฀around฀76฀mya).฀Furthermore,฀the฀corresponding฀minimum฀ ages฀are฀older฀than฀the฀first฀point฀at฀which฀North฀and฀South฀America฀were฀ directly฀connected฀by฀the฀present฀day฀Central฀American฀land฀bridge฀(Fig.฀5)฀ (3.5-3.1฀mya;฀Burnham฀&฀Graham,฀1999).฀ If฀these฀conclusions฀are฀interpreted฀as฀support฀for฀a฀Boreotropical฀dispersal฀ of฀SBC฀subclades,฀then฀South฀American฀origin฀of฀the฀SAC฀clade฀would฀have฀ to฀imply฀dispersal฀between฀North฀and฀South฀America฀prior฀to฀the฀closure฀of฀ the฀Panama฀isthmus.฀Both฀animal฀and฀plant฀fossil฀evidence฀suggest฀that฀this฀ may฀have฀been฀possible฀(Morley,฀2003).฀The฀plate-kinematic฀model฀of฀Pindall฀et฀ al.฀(1988),฀is฀interpreted฀to฀suggest฀two฀windows฀of฀opportunity฀for฀dispersal฀ associated฀with฀the฀formation฀of฀island฀chains฀at฀the฀leading฀edge฀the฀eastdrifting฀Caribbean฀plate฀(Morley,฀2003;฀Pennington฀&฀Dick,฀2004).฀Firstly,฀the฀ proto-Greater฀Antilles฀formed฀a฀bridge฀between฀Yucatán฀and฀Colombia,฀ca.฀50฀ mya,฀which฀was฀subsequently฀fractured฀as฀the฀plate฀drifted฀further฀east.฀Secondly,฀ a฀ land฀ mass฀ including฀ the฀ Greater฀ Antilles฀ and฀ Aves฀ Ridge฀ (GAARlandia;฀ Iturralde-Vinent฀&฀McPhee,฀1999)฀formed฀around฀35-33฀mya฀which฀may฀have฀ provided฀a฀dispersal฀route฀for฀around฀3฀million฀years,฀before฀fragmenting฀to฀ form฀the฀present฀day฀Caribbean฀islands฀(Morley,฀2003).฀The฀Panama฀isthmus฀ itself฀formed฀in฀the฀Pliocene฀from฀the฀island฀arc฀associated฀with฀the฀trailing฀ edge฀of฀the฀Caribbean฀plate฀(Pindall฀et฀al.,฀1988). These฀two฀pre-Panama฀dispersal฀opportunities฀appear฀not฀to฀have฀been฀ utilised฀ by฀ ancestors฀ of฀ the฀ genera฀ Desmopsis,฀ Sapranthus,฀ Stenanona,฀ and฀ Tr idimer is,฀which฀are฀nested฀with฀high฀support฀within฀the฀otherwise฀almost฀ exclusively฀Asian฀ miliusoid฀ clade฀ (Mols฀ et฀ al.,฀ 2004).฀ Both฀ Sapranthus฀ and฀ Tridimeris฀are฀endemic฀to฀Central฀America,฀as฀are฀most฀species฀of฀Desmopsis฀ and฀ Stenanona:฀ the฀ few฀ South฀American฀ species฀ of฀ these฀ genera฀ are฀ found฀ only฀ on฀ the฀ Pacific฀ coast฀ of฀ Colombia฀ (P.฀ Maas,฀ pers.฀ comm.),฀ apparently฀ limited฀in฀their฀distribution฀into฀South฀America฀by฀the฀barrier฀presented฀by฀ the฀Andean฀mountain฀chain฀(see฀below).฀Clades฀within฀the฀SAC฀clade,฀such฀ as฀those฀of฀Cremastosper ma,฀Klarobelia,฀Malmea,฀or฀Mosannona,฀are฀diverse฀ in฀ South฀America.฀ If฀ they฀ were฀ to฀ have฀ originated฀ in฀ Central฀America฀ and฀ dispersed฀ south฀ across฀ the฀ Panama฀ isthmus฀ (i.e.฀ within฀ the฀ last฀ 3.5฀ million฀ years),฀ it฀ would฀ be฀ difficult฀ to฀ explain฀ why฀ those฀ miliusoid฀ clades฀ did฀ not.฀ More฀data฀is฀needed฀to฀get฀more฀accurate฀dates฀for฀the฀divergences฀of฀Central฀ American฀taxa,฀and฀thus฀better฀to฀understand฀how฀the฀distributions฀of฀SBC฀taxa฀ were฀affected฀by฀pre-Panama฀isthmus฀land-bridges.฀However,฀in฀the฀absence฀ of฀a฀fully฀sampled฀and฀completely฀resolved฀phylogeny฀of฀the฀SAC฀clade,฀we฀ consider฀these฀results฀to฀provide฀convincing฀support฀for฀a฀common฀ancestor฀of฀ 60 • Chapter 3 Cremastosperma,฀Klarobelia,฀Malmea,฀and฀Mosannona฀having฀been฀present฀on฀ the฀South฀American฀continent. Monophyly of Cremastosperma, Klarobelia, Malmea, and Mosannona In฀general฀monophyly฀among฀genera฀sampled฀was฀confirmed.฀A฀possible฀ exception฀ was฀ the฀ genus Oxandra,฀ the฀ accessions฀ of฀ which฀ fell฀ within฀ two฀ clades฀plus฀one฀isolated฀lineage.฀Monophyly฀was฀confirmed฀in฀Cremastosperma,฀ Malmea฀and฀Mosannona.฀Klarobelia฀proved฀broadly฀monophyletic,฀with฀the฀ exception฀ of฀ a฀ single฀ accession.฀ Pseudephedranthus฀ fragrans,฀ representing฀ a฀ monotypic฀ genus฀ from฀ the฀ upper฀ Rio฀ Negro฀ on฀ the฀ Brazil/Venezuela/ Colombia฀ border,฀ was฀ nested฀ within฀ Klarobelia.฀ It฀ was฀ sister฀ to฀ a฀ clade฀ of฀ species฀ not฀ represented฀ by฀ Richardson฀ et฀ al.฀ (2004)฀ which฀ together฀ forms฀ the฀ sister฀ group฀ to฀ the฀ rest฀ of฀ Klarobelia,฀ thus฀ explaining฀ why฀ its฀ position฀ had฀not฀been฀previously฀discovered.฀This฀result฀is฀however฀curious,฀given฀the฀ morphological฀characters฀(such฀as฀impressed฀as฀opposed฀to฀raised฀midrib฀and฀ open฀rather฀than฀closed฀bud฀development)฀which฀otherwise฀appear฀to฀represent฀ synapomorphies฀for฀Klarobelia฀to฀the฀exclusion฀of฀Pseudephedranthus,฀and฀also฀ warrants฀further฀investigation.฀The฀‘Andean฀centred’฀distribution฀patterns฀were฀ thus฀not฀contradicted,฀though฀the฀unique฀distribution฀of฀Pseudephedranthus฀ falls฀between฀those฀of฀the฀disjunct฀localities฀of฀other฀species฀in฀the฀tropical฀ Andes฀and฀coastal฀Venezuela.฀ Progress towards reconstructing the biogeographic history of Cremastosperma, Klarobelia, Malmea, and Mosannona One฀biogeographical฀hypothesis฀concerning฀Andean฀centred฀groups฀such฀ as฀Cremastosperma,฀Klarobelia,฀Malmea,฀and฀Mosannona฀was฀that฀speciation฀in฀ these฀groups฀occurred฀as฀a฀result฀of฀the฀Andean฀orogeny.฀The฀inferred฀presence฀ of฀ancestors฀of฀the฀four฀clades฀on฀the฀South฀American฀continent฀prior฀to฀the฀ closure฀of฀the฀Panama฀isthmus,฀and฀confirmation฀of฀their฀distribution฀patterns฀ through฀ testing฀ monophyly฀ of฀ the฀ accessions฀ of฀ these฀ taxa฀ sampled,฀ failed฀ to฀ reject฀ this฀ hypothesis.฀A฀ further฀ test฀ was฀ applied฀ using฀ molecular฀ dating฀ techniques฀ to฀ examine฀ the฀ possibility฀ that฀ diversifications฀ in฀ these฀ groups฀ occurred฀ within฀ the฀ time฀ frame฀ of฀ the฀ elevation฀ of฀ the฀Andes.฀ Burnham฀ &฀ Graham฀(1999)฀considered฀the฀Andean฀orogeny฀to฀have฀been฀an฀influence฀in฀ the฀history฀of฀Neotropical฀vegetation฀since฀the฀Miocene.฀This฀would฀place฀the฀ effective฀timeframe฀for฀diversifications฀that฀might฀have฀been฀associated฀with฀ the฀Andean฀orogeny฀at฀anywhere฀between฀around฀23.3฀mya฀and฀the฀present.฀ However,฀estimations฀of฀paleoelevation฀of฀the฀Central฀and฀Colombian฀Andes฀ suggest฀ that฀ much฀ of฀ the฀Andean฀ uplift฀ occurred฀ in฀ the฀ late฀ Miocene฀ and฀ Pliocene฀(Gregory-Wodzicki,฀2000),฀when฀elevations฀increased฀by฀more฀than฀ 3,500฀ m฀ (Lundberg฀ &฀ Chernoff,฀ 1992).฀The฀ central฀Andes฀ had฀ reached฀ no฀ Biogeography of Andean-centred Annonaceae • 61 more฀than฀half฀its฀modern฀elevation฀by฀10฀mya฀and฀the฀Eastern฀Cordillera฀of฀ the฀Colombian฀Andes฀was฀at฀no฀more฀than฀40%฀of฀its฀modern฀elevation฀by฀4฀ mya฀(Gregory-Wodzicki,฀2000).฀ Molecular฀dating฀results฀produced฀here฀showed฀the฀ages฀of฀MRCAs฀of฀ each฀of฀the฀four฀genera฀to฀fall฀within฀roughly฀the฀same฀wide฀time฀window฀(the฀ oldest฀dates฀we฀estimated฀are฀those฀using฀NPRS฀that฀give฀ages฀of฀from฀22.33฀ +/-฀ 4.99฀ mya฀ to฀ 15.96฀ +/-฀ 3.31฀ mya).฀The฀ Multidivtime฀ and฀ PL฀ methods฀ suggest฀even฀younger฀ages฀for฀diversification฀in฀these฀genera฀that฀even฀more฀ clearly฀fall฀within฀the฀timeframe฀of฀Andean฀orogeny.฀These฀dates฀are฀consistent฀ with฀suggestions฀by฀Burnham฀and฀Graham฀(1999)฀that฀the฀Andean฀orogeny฀ has฀had฀an฀effect฀on฀the฀history฀of฀Neotropical฀vegetation฀since฀the฀Miocene.฀ However,฀imprecision฀in฀the฀results฀means฀that฀we฀cannot฀exclude฀the฀possibility฀ that฀speciation฀within฀each฀of฀the฀four฀clades฀actually฀occurred฀during฀rather฀ different฀time฀slices. That฀no฀one฀species฀of฀any฀of฀these฀four฀genera฀is฀to฀be฀found฀on฀both฀ sides฀ of฀ the฀Andes฀ provides฀ compelling฀ evidence฀ that฀ the฀ current฀ elevation฀ of฀the฀northern฀Andes฀forms฀an฀effective฀barrier฀to฀dispersal฀in฀these฀groups.฀ Furthermore,฀preliminary฀results฀in฀Mosannona฀presented฀here฀show฀most฀of฀ the฀species฀sampled฀to฀fall฀into฀two฀clades,฀one฀from฀west฀and฀one฀east฀of฀the฀ Andes฀(see฀Figs฀3฀and฀5).฀If฀the฀north-Andean฀uplift฀represents฀a฀vicariance฀ event฀common฀to฀this,฀and฀other฀SAC฀clade฀genera,฀then฀this฀should฀be฀reflected฀ in฀their฀species฀phylogenies,฀with฀congruent฀timing฀of฀subsequent฀divergences.฀ The฀ age฀ of฀ the฀ divergence฀ between฀ Mosannona฀ clades฀ on฀ the฀ west฀ and฀ on฀ the฀east฀of฀the฀Andes฀(represented฀by฀nodes฀1฀and฀2฀in฀Fig.฀5,฀and฀in฀Table฀3)฀ could฀be฀as฀early฀as฀15฀mya,฀or฀rather฀more฀recent฀according฀to฀our฀estimates,฀ which฀is฀consistent฀with฀the฀limitation฀in฀elevation฀below฀which฀species฀of฀ the฀SAC฀clade฀are฀found฀(mostly฀1,500฀m,฀very฀rarely฀up฀to฀around฀2,000฀m),฀ which฀would฀probably฀place฀a฀vicariance฀explanation฀for฀speciation฀in฀these฀ groups฀no฀earlier฀than฀the฀Pliocene.฀The฀absence฀of฀miliusoid฀clade฀taxa฀east฀of฀ the฀Andes฀would฀suggest฀the฀timing฀to฀have฀been฀prior฀to฀the฀closure฀of฀the฀ Panama฀isthmus.฀However,฀alternative฀(or฀additional)฀causes฀of฀speciation฀(or฀ indeed฀extinction)฀in฀these฀groups฀could฀have฀been฀linked฀to฀distribution฀shifts฀ along฀the฀Andean฀elevational฀range฀during฀climatic฀changes฀in฀the฀Pleistocene฀ (Hooghiemstra฀&฀van฀der฀Hammen,฀1998). Conclusions We฀consider฀the฀fully฀resolved฀(though฀not฀in฀all฀nodes฀highly฀supported)฀ topology฀ presented฀ here฀ to฀ represent฀ a฀ credible฀ hypothesis฀ of฀ phylogenetic฀ relationships฀between฀the฀major฀clades฀of฀the฀SBC฀-฀but฀one฀which฀should฀be฀ further฀tested฀with฀independent฀data. 62 • Chapter 3 The฀identification฀of฀a฀clade฀comprising฀all฀the฀SBC฀genera฀distributed฀ in฀South฀America฀and฀mostly฀only฀to฀a฀limited฀extent฀into฀Central฀America฀ (the฀SAC฀clade),฀to฀the฀exclusion฀particularly฀of฀Asian฀and฀Central฀American฀ endemic฀clades,฀suggests฀a฀common฀origin฀of฀the฀SAC฀clade฀in฀South฀America.฀ Origin฀of฀the฀SAC฀clade฀in฀South฀America฀as฀a฀result฀of฀dispersal฀across฀the฀ Boreotropics฀is฀supported฀by฀the฀age฀estimations฀presented฀here฀rather฀than฀ Gentry’s฀hypothesis฀of฀origin฀by฀Gondwanan฀vicariance.฀Broad฀monophyly฀of฀ the฀four฀genera฀leads฀us฀to฀conclude฀that฀the฀distribution฀patterns฀as฀observed฀ are฀not฀the฀arbitrary฀result฀of฀the฀definition฀of฀poly-฀or฀paraphyletic฀groups.฀ The฀ages฀estimated฀for฀the฀MRCA฀of฀each฀clade฀were฀not฀significantly฀different฀ from฀each฀other฀and฀appear฀to฀fall฀within฀the฀time฀frame฀of฀the฀orogeny฀of฀the฀ Northern฀Andes,฀though฀the฀strength฀of฀this฀test฀was฀limited฀by฀imprecision฀in฀ the฀molecular฀dating฀results. Further฀testing฀these฀biogeographic฀hypotheses฀requires฀the฀reconstruction฀ of฀species฀level฀phylogenies฀of฀Cremastosper ma,฀Klarobelia,฀Mosannona฀and฀ Malmea.฀ Further฀ work฀ should฀ be฀ concentrated฀ both฀ on฀ finding฀ further฀ age฀ calibration฀points฀for฀the฀Annonaceae฀phylogeny฀and฀on฀assessing฀sources฀of฀ error฀in฀the฀techniques฀used฀to฀derive฀ultrametric฀trees.฀This฀approach฀could฀ shed฀further฀light฀on฀the฀dynamic฀processes฀of฀invasion฀of฀Central฀America฀ (Chatrou,฀1997),฀and฀the฀origin฀of฀high฀species฀diversity฀in฀tropical฀America. Acknowledgements The฀authors฀would฀like฀to฀express฀gratitude฀to฀Jan฀Maas฀and฀Maria฀Paula฀ Balcázar฀Vargas฀for฀help฀in฀the฀lab,฀to฀Tanya฀Scharaschkin฀and฀Jim฀Doyle฀for฀ advice฀and฀access฀to฀unpublished฀Annonaceae฀primers,฀and฀to฀the฀herbaria฀NY฀ and฀WAG฀for฀granting฀permission฀for฀the฀extraction฀of฀DNA฀from฀herbarium฀ specimens.฀James฀Richardson฀and฀Jifke฀Koek-Noorman฀gave฀helpful฀criticism฀ of฀ the฀ manuscript.฀ Distribution฀ maps฀ in฀ Fig.฀1฀ were฀ produced฀ using฀ ESRI฀ data฀made฀available฀by฀the฀New฀York฀Botanical฀Garden’s฀‘Digital฀Basemap฀of฀ the฀Americas’.฀Phylogenetic฀trees฀in฀Figs฀3-5฀were฀produced฀using฀TreeGraph฀ (Müller฀&฀Müller,฀2004). Biogeography of Andean-centred Annonaceae • 63 64 • Chapter 3 Chapter 4 An ancient paralogue of the cpDNA tr nL (UAA)-tr nF (GAA) region in Annonaceae and its application in phylogeny reconstruction M i c h a e l D . P i r i e 1, M a r i a P a u l a B a l c á z a r Va r g a s 1, M a r l e e n B o t e r m a n s 2 , F r e e k T. B a k k e r 2 , a n d L a r s W. C h a t r o u 1 Submitted to Molecular Pylogenetics and Evolution 1 Nat i onaal Her bar i um Neder l and, Uni ver s i t ei t Ut r ec ht br anc h, Hei del ber gl aan 2, 3584 CS Ut r ec ht , The Net her l ands ; 2 Nat i onaal Her bar i um Neder l and, Wageni ngen Uni ver si t ei t br anc h, Gener aal Foul kes w eg 37, 6703 BL Wageni ngen, The Net her l ands A paralogue of tr�L-F in Annonaceae • 65 Abstract Phylogenetic฀signal฀of฀trnL-F฀in฀Annonaceae฀conflicts฀with฀that฀of฀other฀ chloroplast฀ markers฀ with฀ respect฀ to฀ the฀ position฀ of฀ the฀ Neotropical฀ genus฀ Unonopsis.฀This฀conflict฀is฀shown฀to฀be฀the฀result฀of฀preferential฀amplification฀ of฀ a฀ paralogous฀ copy฀ of฀ the฀ tr nL-F฀ region,฀ pseudtr nL-F,฀ which฀ diverged฀ from฀trnL-F฀in฀a฀direct฀ancestor฀of฀the฀Annonaceae.฀The฀pseudtrnL-F฀region฀ evolves฀at฀a฀faster฀rate฀than฀tr nL-F.฀Although฀the฀exon฀appears฀intact,฀some฀ pseudtr nL-F฀ intron฀ sequences฀ show฀ signs฀ of฀ disruption฀ of฀ the฀ secondary฀ structure฀otherwise฀conserved฀across฀land฀plants.฀The฀pseudtrnL-F฀region฀may฀ have฀phylogenetic฀utility฀for฀some฀groups฀of฀Annonaceae,฀but฀until฀its฀position฀ in฀the฀Annonaceae฀genome฀is฀determined฀the฀phylogenetic฀signal฀it฀contains฀ will฀have฀to฀be฀interpreted฀with฀caution. ฀ Keywords:฀ Annonaceae,฀ Magnoliales,฀ Unonopsis,฀ conflict,฀ paralogy,฀ lineage฀ sorting,฀฀ trnL-F. Abbreviations:฀ BS฀ (bootstrap฀ support);฀ LBC฀ (long฀ branch฀ clade);฀ SBC฀ (short฀ branch฀ clade);฀ SAC฀ clade฀ (South฀American฀ centred฀ clade);฀ MP฀ (maximum฀ parsimony);฀ ML฀ (maximum฀ likelihood);฀ MRCA฀ (most฀ recent฀ common฀ ancestor);฀ PL฀ (penalized฀ likelihood);฀PP฀(posterior฀probability). Introduction For฀ the฀ last฀ decade฀ DNA฀ sequences฀ from฀ the฀ chloroplast฀ genome฀ have฀ been฀widely฀exploited฀as฀an฀invaluable฀source฀of฀characters฀for฀phylogenetic฀ inference฀in฀plants.฀This฀is฀largely฀due฀to฀their฀uniparental฀mode฀of฀inheritance฀ and฀ effective฀ lack฀ of฀ recombination.฀ Phylogenies฀ inferred฀ using฀ chloroplast฀ DNA฀ (cpDNA)฀ sequence฀ data฀ may฀ or฀ may฀ not฀ reflect฀ species฀ phylogenies฀ (Nichols,฀2001),฀but฀they฀can฀in฀any฀case฀be฀regarded฀as฀gene฀trees฀for฀which฀ the฀congruence฀with฀organismal฀history฀becomes฀greater฀with฀increasing฀time฀ scale฀(Clegg฀&฀Zurawski,฀1992).฀ For฀practical฀purposes,฀polymerase฀chain฀reaction฀(PCR)฀amplification฀of฀ cpDNA฀markers฀is฀made฀easier฀by฀the฀mostly฀large฀numbers฀of฀chloroplasts฀ present฀in฀each฀plant฀cell.฀In฀a฀solution฀of฀total฀extracted฀DNA฀target฀sequences฀ have฀higher฀effective฀copy฀number฀than฀those฀of฀most฀markers฀located฀in฀the฀ cell฀nucleus.฀The฀practical฀application฀and฀patterns฀of฀evolution฀of฀different฀ potential฀cpDNA฀markers฀are฀increasingly฀well฀understood.฀Not฀only฀have฀a฀ number฀of฀complete฀chloroplast฀genomes฀been฀sequenced฀(e.g.฀Notsu฀et฀ al.,฀ 2002),฀ but฀ regions฀ within฀ the฀ chloroplast฀ have฀ been฀ sequenced฀ for฀ a฀ broad฀ selection฀of฀plants฀(e.g.฀Small฀et฀al.,฀1998;฀Shaw฀et฀al.,฀2005).฀ 66 • Chapter 4 T h e c h l o r o p l a s t e n c o d e d t r n T- F r e g i o n The฀cpDNA฀tr nL-F฀region฀in฀land฀plants฀consists฀of฀the฀transfer฀RNA฀ genes฀ tr nT UGU,฀ tr nL UAA฀ and฀ tr nF GAA฀ arranged฀ in฀ tandem,฀ separated฀ by฀ non-coding฀ spacer฀ regions.฀ It฀ is฀ positioned฀ in฀ the฀ large฀ single฀ copy฀ region,฀ approximately฀8kb฀downstream฀of฀rbcL.฀The฀tr nL฀ gene฀of฀cyanobacteria฀and฀ a฀number฀of฀chloroplast฀genomes,฀including฀that฀of฀all฀land฀plants,฀contains฀a฀ group-I฀intron฀positioned฀between฀the฀U฀and฀the฀A฀of฀the฀UAA฀anticodon฀ loop.฀This฀intron฀is฀inferred฀by฀phylogenetic฀analysis฀to฀have฀been฀present฀in฀ the฀cyanobacterial฀ancestor฀of฀the฀plastid฀lineages฀of฀Rhodophyta,฀Chlorophyta฀ and฀ Glaucocystophyta฀ and฀ subsequently฀ vertically฀ transmitted฀ (Besendahl฀฀ et฀al.,฀2000). The฀succession฀of฀conserved฀trn฀genes฀and฀the฀apparent฀absence฀of฀gene฀ rearrangements฀in฀the฀tr nT-F฀region฀facilitated฀the฀design฀of฀plant฀universal฀ primers฀by฀Taberlet฀et฀ al.฀(1991),฀since฀which฀in฀particular฀the฀ tr nL฀intron฀ and฀trnL-F฀spacer฀(together,฀the฀trnL-F฀region)฀has฀become฀one฀of฀the฀most฀ widely฀used฀chloroplast฀markers฀for฀phylogenetic฀analyses฀in฀plants฀(Quandt฀et฀ al.,฀2004).฀The฀accumulation฀of฀an฀increasingly฀large฀body฀of฀sequences฀of฀the฀ trn(T-)L-F฀region฀from฀a฀wide฀range฀of฀plants฀has฀allowed฀further฀study฀of฀its฀ structures,฀functions฀and฀evolution฀(e.g.฀in฀different฀orders฀of฀flowering฀plants:฀ Bakker฀et฀ al.,฀2000;฀in฀basal฀angiosperms:฀Borsch฀et฀ al.,฀2003;฀in฀land฀plants:฀ Quandt฀et฀al.,฀2004;฀in฀bryophytes:฀Quandt฀&฀Stech,฀2004). Use of trnL-F in phylogeny reconstruction in Annonaceae Sequences฀from฀the฀tr nL-F฀region฀(not฀including฀the฀tr nT-L฀region฀or฀ trnL฀5’฀exon)฀have฀recently฀been฀used฀in฀combination฀with฀those฀from฀further฀ chloroplast฀ markers฀ rbcL฀ and฀ matK฀ as฀ a฀ source฀ of฀ characters฀ for฀ phylogeny฀ reconstruction฀in฀the฀tropical฀flowering฀plant฀family฀Annonaceae฀(Sauquet฀et฀ al.,฀2003;฀Mols฀et฀al.,฀2004;฀Richardson฀et฀al.,฀2004;฀this฀thesis,฀Chapter฀2).฀The฀ majority฀of฀species฀of฀Annonaceae฀fall฀within฀two฀large฀clades฀(Richardson฀et฀ al.,฀2004).฀The฀informally฀named฀‘long฀branch฀clade’฀(LBC)฀represents฀around฀ 1,500฀of฀the฀total฀2,500฀species฀and฀is฀sister฀group฀to฀the฀‘short฀branch฀clade’฀ (SBC),฀ which฀ itself฀ represents฀ a฀ further฀ 700฀ species฀ (Fig.฀ 1).฀The฀ remaining฀ species฀fall฀within฀two฀clades฀forming฀a฀basal฀grade฀(this฀thesis,฀Chapter฀2).฀As฀ is฀to฀be฀expected฀from฀different฀sequences฀sampled฀from฀the฀plastid฀genome฀ (Chase฀&฀Cox,฀1998),฀these฀markers฀were฀shown฀to฀contain฀complementary฀ phylogenetic฀ signal,฀ and฀ were฀ thus฀ applied฀ in฀ combined฀ analyses฀ which฀ delivered฀more฀resolved฀phylogenies,฀subject฀to฀higher฀levels฀of฀support,฀than฀ those฀derived฀from฀each฀marker฀individually. In฀ contrast฀ to฀ this฀ result,฀ subsequent฀ maximum฀ parsimony฀ analysis฀ of฀ sequences฀of฀the฀trnL-F฀region฀for฀additional฀accessions฀resulted฀in฀a฀position฀ of฀the฀Neotropical฀genus฀Unonopsis฀conflicting฀with฀that฀derived฀from฀other฀ cpDNA฀markers.฀ A paralogue of tr�L-F in Annonaceae • 67 �a�����m�ma��e��ra�ea�a �e�e�er�a�rose���ora 100 L�r�o�e��ro������e�se 91 68 99 �a��o��a��o��s �mer�a�����a�e�s�s ��pomat�a��e��et�� Unonopsis elegantissima 100 �C� Unonopsis stipitata ��a�a�orea�p�aeo�arpa 100 �A� �e��ett�a�par����ora 99 Ca�a��a�o�orata 94 �B� 100 basal grade 89 SAC clade taxa 1 Coe�o�ar�o��pre�ss�� �aurales �ersea�amer��a�a �agnoliales outgroups C���amom�m��ass�a �r����aea��a��e�peta�a 67 99 Letest��o�a��e��a ����et�a���r�sea 100 100 �BC C�m�opeta��m�sp� 98 ��ar�a������a �o�a�t�ota��s����te� �o�o�arp�a�e��e�ra 99 93 ��p�o�sea��o��a�a 59 �o��a�t��a�s��erosa �apra�t��s���r�����or�s �������a����ora�t�a �o��a�t��a���a��a 94 �ree��a�o�e��ro��o���er� 85 95 SBC ��ptost��ma�morte�a�� �nychopetalum periquino 95 Cremastosperma��re��pes Cremastosperma��e�op�����m 59 100 �a�mea���e�s�a�a �a�mea�s�r��ame�s�s SAC clade taxa 2 Bocageopsis multiflora 56 �osa��o�a��ostar��e�s�s 70 76 ��aro�e��a��a�����ora ��a��ra�as�e���� F i g . 1 : Maximum parsimony 50% majority rule bootstrap consensus tree resulting from analysis of the entire trnL-F region for a selection of Annonaceae, other Magnoliales, and Laurales outgroups. Bootstrap support is indicated above the nodes. Nodes labelled A, B and C are those defining Unonopsis as a monophyletic sister group to Annonaceae. 68 • Chapter 4 Conflicting results in the placement of Unonopsis in Annonaceae trnL-F phylogeny Unonopsis฀has฀been฀grouped฀with฀two฀smaller฀South฀American฀genera,฀ Bocageopsis฀ and฀ Onyc hopetalum฀ (comprising฀ 4฀ species฀ each฀ to฀ the฀ 38฀ of฀ Unonopsis),฀on฀the฀basis฀of฀morphological฀similarity฀(they฀share฀small,฀usually฀ whitish฀ flowers฀ with฀ valvate฀ sepals฀ and฀ petals,฀ and฀ a฀ low฀ number฀ of฀ pitted,฀ flattened฀ seeds฀ in฀ lateral฀ position฀ with฀ a฀ canaliculate฀ raphe฀ and฀ spiniform฀ ruminations;฀Van฀Heusden,฀1992;฀Van฀Setten฀&฀Koek-Noorman,฀1992).฀However,฀ monophyly฀of฀Unonopsis฀with฀respect฀to฀Bocageopsis฀and฀Onychopetalum฀is฀ far฀from฀certain,฀and฀that฀of฀each฀of฀the฀three฀has฀yet฀to฀be฀stringently฀tested.฀ In฀ order฀ to฀ attempt฀ this฀ test,฀ and฀ as฀ part฀ of฀ wider฀ phylogenetic฀ research฀ in฀ Annonaceae,฀we฀compared฀sequences฀from฀the฀trnL-F,฀rbcL,฀matK฀and฀psbAtrnH฀regions฀for฀a฀number฀of฀samples฀of฀each฀of฀the฀three฀genera. Inclusion฀of฀the฀resulting฀trnL-F฀sequences฀in฀phylogenetic฀analysis฀with฀a฀ selection฀of฀Annonaceae,฀other฀Magnoliales฀and฀Laurales฀trnL-F฀sequences฀(as฀ used฀by฀Mols฀et฀al.,฀2004;฀Richardson฀et฀al.,฀2004)฀resulted฀in฀a฀monophyletic฀ Unonopsis฀as฀sister฀group฀to฀the฀rest฀of฀the฀Annonaceae฀(Fig.฀1).฀This฀result฀ was฀in฀direct฀conflict฀with฀the฀other฀plastid฀DNA฀sequence฀data.฀Phylogenetic฀ analysis฀of฀multiple฀chloroplast฀markers฀supports฀monophyly฀of฀the฀Unonopsis/ Bocageopsis/Onychopetalum฀clade,฀placed฀with฀high฀support฀within฀the฀South฀ American฀centred฀clade฀(SAC฀clade),฀itself฀nested฀within฀the฀SBC฀(this฀thesis,฀ Chapter฀3).฀ A฀number฀of฀potential฀problems฀in฀phylogeny฀reconstruction฀can฀lead฀to฀ apparently฀incongruent฀results.฀In฀particular,฀weak฀signal฀in฀the฀data,฀failings฀ of฀the฀inference฀method฀under฀particular฀conditions,฀and฀incorrect฀assessment฀ of฀homology฀-฀either฀in฀sequence฀alignment,฀or฀due฀to฀paralogy฀(Sanderson฀&฀ Shaffer,฀2002).฀No฀such฀immediately฀obvious฀errors฀were฀apparent.฀The฀position฀ of฀Unonopsis฀in฀the฀tr nL-F฀phylogeny฀received฀high฀bootstrap฀support.฀The฀ alignment฀ of฀ sequences฀ was฀ unambiguous.฀ No฀ unusually฀ large฀ insertion/ deletions฀were฀observed,฀nor฀was฀the฀subtending฀branch฀representing฀changes฀ in฀the฀two฀Unonopsis฀sequences฀unusually฀long.฀ The฀chloroplast฀genome฀is฀rarely฀subject฀to฀recombination.฀In฀the฀absence฀ of฀recombination,฀incongruence฀between฀the฀phylogenetic฀signal฀of฀different฀ chloroplast฀markers฀could฀be฀caused฀by฀paralogy,฀leading฀to฀lineage฀sorting.฀The฀ node฀of฀the฀most฀recent฀common฀ancestor฀(MRCA)฀of฀putatively฀orthologous฀ and฀paralogous฀sequences฀in฀the฀trnL-F฀(gene)฀phylogeny฀would฀represent฀the฀ divergence฀of฀the฀two฀paralogues.฀This฀itself฀could฀represent฀the฀duplication฀of฀ a฀region฀including฀tr nL-F.฀It฀could฀alternatively฀represent฀a฀speciation฀event,฀ followed฀by฀hybridisation฀with฀an฀unknown฀extra-Annonaceae฀Magnoliales฀ lineage,฀and฀heteroplasmy฀(though฀this฀might฀seem฀an฀outside฀possibility).฀In฀ either฀case,฀the฀phylogeny฀presented฀in฀Fig.฀1฀could฀be฀observed฀if฀in฀Unonopsis฀ A paralogue of tr�L-F in Annonaceae • 69 one฀ paralogue฀ was฀ amplified฀ by฀ PCR,฀ and฀ in฀ other฀Annonaceae,฀ the฀ other฀ paralogue.฀ The฀inference฀of฀lineage฀sorting฀leads฀to฀two฀predictions:฀firstly,฀tr nL-F฀ sequences฀with฀the฀same฀signal฀as฀the฀other฀chloroplast฀markers฀should฀be฀(or฀ should฀ have฀ been)฀ present฀ in฀ Unonopsis.฀ Secondly,฀ if฀ lineage฀ sorting฀ is฀ the฀ result฀ of฀ a฀ duplication,฀ a฀ second฀ copy฀ of฀ tr nL-F฀ should฀ be฀ (or฀ should฀ have฀ been)฀ present฀ in฀ all฀ other฀ taxa฀ descending฀ from฀ the฀ node฀ representing฀ the฀ inferred฀ duplication฀ event.฀According฀ to฀ the฀ preliminary฀ results,฀ thus,฀ in฀ all฀ Annonaceae. Aims and objectives In฀this฀paper฀we฀test฀the฀hypothesis฀that฀conflicting฀phylogenetic฀signal฀in฀ the฀trnL-F฀region฀in฀Annonaceae฀is฀the฀result฀of฀lineage฀sorting,฀using฀PCRbased฀and฀phylogenetic฀analysis฀techniques.฀Copy-specific฀PCR฀primers฀were฀ designed฀based฀on฀Annonaceae฀trnL-F฀sequences฀obtained฀previously฀(Sauquet฀ et฀ al.,฀2003;฀Mols฀et฀ al.,฀2004;฀Richardson฀et฀ al.,฀2004;฀this฀thesis,฀Chapter฀ 2).฀These฀might฀amplify฀the฀different฀paralogues,฀if฀present,฀both฀in฀Unonopsis฀ and฀in฀other฀Annonaceae.฀We฀then฀draw฀further฀conclusions฀with฀respect฀to฀the฀ timing฀of฀divergence฀and฀to฀the฀phylogenetic฀signal฀contained฀in฀both฀copies.฀ Support฀for฀the฀hypothesis฀raises฀a฀number฀of฀further฀issues฀concerning฀the฀ definition฀of฀homology฀in฀Magnoliales฀trnL-F฀regions.฀To฀address฀the฀question฀ of฀functional฀homology,฀comparisons฀are฀drawn฀between฀sequences฀obtained฀ in฀this฀study฀and฀proposed฀secondary฀structures฀and฀corresponding฀functional฀ constraints฀ in฀ the฀ tr nL฀ gene฀ and฀ Group฀ I฀ intron฀ in฀ land฀ plants฀ (Borsch฀ et฀ al.,฀2003;฀Quandt฀et฀ al.,฀2004).฀Positional฀homology฀(i.e.฀within฀or฀possibly฀ without฀the฀chloroplast฀genome)฀and฀the฀precise฀origin฀of฀paralogues฀is฀less฀ straightforward฀ to฀ determine฀ from฀ sequences฀ alone.฀A฀ number฀ of฀ potential฀ directions฀for฀future฀research,฀such฀as฀the฀application฀of฀additional฀PCR฀based฀ or฀southern฀blotting฀techniques,฀are฀therefore฀also฀discussed. Materials and Methods Ta x o n s a m p l i n g Recent฀improvements฀in฀both฀phylogenetic฀resolution฀and฀representation฀ of฀ taxa฀ (Sauquet฀ et฀ al.,฀ 2003;฀ Mols฀ et฀ al.,฀ 2004;฀ Richardson฀ et฀ al.,฀ 2004)฀ provide฀a฀robust฀framework฀for฀the฀choice฀of฀taxa฀in฀phylogeny฀reconstruction฀ in฀Magnoliales.฀This฀study฀utilised฀previously฀unpublished฀sequence฀data,฀as฀ well฀as฀published฀tr nL-F,฀rbcL,฀matK฀and฀psbA-tr nH฀sequences฀(Kojoma฀et฀ 70 • Chapter 4 al.,฀2002;฀Sauquet฀et฀ al.,฀2003;฀Mols฀et฀ al.,฀2004;฀this฀thesis,฀Chapters฀2฀&฀3;฀ Chatrou฀et฀al.,฀in฀prep.;฀see฀Appendix฀A). DNA extraction, PCR amplification and sequencing Total฀ genomic฀ DNA฀ was฀ extracted฀ using฀ a฀ modified฀ cetyl฀ trimethyl฀ ammonium฀ bromide฀ (CTAB)฀ method฀ (Doyle฀ &฀ Doyle,฀ 1987):฀ 50฀ mg฀ silica฀ dried฀ or฀ herbarium฀ leaf฀ material฀ was฀ homogenised฀ in฀ 1300฀ μl฀ CTAB฀ and฀ incubated฀for฀20฀minutes฀with฀12฀ μl฀2-mercaptoethanol฀at฀65°C,฀followed฀by฀ 90฀minutes฀ambient฀mixing฀with฀1฀ml฀24:1฀chloroform:฀isoamylalcohol.฀After฀ 10฀minutes฀centrifuged฀at฀13,000฀rpm,฀300฀ μl฀supernatant฀was฀purified฀using฀ Wizard฀DNA฀purification฀system฀(Promega฀corp.)฀(i.e.฀without฀isopropanol฀ precipitation,฀avoiding฀the฀co-precipitation฀of฀oxidised฀material;฀Savolainen฀et฀ al.,฀1995). A฀standard฀PCR฀protocol฀was฀used฀throughout,฀with฀the฀addition฀of฀1μl฀ 0.4฀%฀BSA฀per฀25฀ μl฀reaction฀(which฀was฀found฀to฀increase฀amplification฀in฀ all฀samples),฀35฀cycles฀of฀30฀sec.:฀94°C;฀1฀min.:฀55-58°C;฀2฀min.:฀72°C;฀with฀ an฀initial฀4฀min.:฀94°C;฀and฀final฀7฀min.:฀72°C.฀PCR฀products฀were฀purified฀ using฀ QIAquick฀ PCR฀ purification฀ kits฀ (Qiagen),฀ sequenced฀ with฀ selected฀ PCR฀and฀specially฀designed฀sequencing฀primers฀(see฀below),฀and฀analysed฀by฀ electrophoresis฀using฀an฀automatic฀sequencer฀ABI฀3730XL. BLAST฀ search฀ (Altschul฀ et฀ al.,฀ 1997)฀ was฀ employed฀ using฀ the฀ NCBI฀ website฀(http://www.ncbi.nlm.nih.gov/BLAST/)฀to฀compare฀the฀Unonopsis฀ tr nL-Fs฀ with฀ published฀ sequences.฀To฀ avoid฀ confusion฀ from฀ this฀ point฀ the฀ trnL-F฀copy฀homologous฀with฀those฀previously฀sequenced฀in฀Annonaceae฀will฀ be฀referred฀to฀as฀‘trnL-F’,฀and฀that฀homologous฀with฀those฀first฀sequenced฀only฀ in฀Unonopsis฀will฀be฀referred฀to฀as฀‘pseudtrnL-F’. tr�L intron tr�L 5� exon C tr�L-F spacer tr�L 3� exon trnLF�intFOR �no39� pseudtrnLF�FOR Primer sequences (5‘-3‘): Annonaceae partial trnl-F / pseudtrnL-F internal sequencing primers: trnLF�intFOR: GAGAGAAACATTTCTGGTCGG trnLF�intREV: GGGCAATCCTGAGCCAAATCC E D tr�F exon trnLF�intREV F �no807R pseudtrnLF�REV pseudtrnLF�specific primers: pseudtrnLF�FOR: GGAAACCTACTAGTCACTTCC pseudtrnLF�REV: GGAGATTCCTTGCCCATTATCTG Annonaceae trnLF�specific primers: �no39�: GGAAACCTACTAAGTGAGAACTTCC �no807R: GCCCATTCATTATCTGTTC F i g . 2 : Scheme of primers used to amplify and sequence trnL-F and pseudtrnL-F. A paralogue of tr�L-F in Annonaceae • 71 For฀ taxa฀ other฀ than฀ Unonopsis฀ the฀ tr nL-F฀ region฀ was฀ amplified฀ and฀ sequenced฀using฀plant฀universal฀primers฀of฀Taberlet฀et฀al.฀(1991)฀in฀combination฀ C/F฀ or฀ C/D฀ and฀ E/F฀ (the฀ positions฀ and฀ sequences฀ of฀ all฀ primers฀ used฀ to฀ amplify฀and฀sequence฀both฀copies฀of฀the฀trnL-F฀region฀are฀presented฀in฀Fig.฀2).฀ Further฀primers฀were฀designed฀to฀amplify฀specifically฀and฀sequence฀the฀different฀ trnL-F฀copies.฀In฀Unonopsis,฀trnL-F฀was฀amplified฀using฀primers฀Uno39F฀and฀ Uno807R.฀In฀other฀Annonaceae฀(and฀in฀Unonopsis฀samples฀where฀sequencing฀ using฀ primers฀ C/D,฀ E/F฀ failed)฀ pseudtr nL-F฀ was฀ amplified฀ using฀ primers฀ pseudtrnLF_FOR฀and฀pseudtrnLF_REV.฀Uno39F฀and฀pseudtrnLF_FOR฀were฀ designed฀to฀anneal฀to฀the฀same฀region฀near฀the฀beginning฀of฀the฀tr nL฀intron฀ where฀length฀differences฀were฀specific฀to฀the฀different฀copies,฀the฀same฀being฀ true฀of฀Uno807฀and฀pseudtrnLF_REV,฀located฀within฀the฀tr nL-F฀intergenic฀ spacer฀(Fig.฀3).฀The฀higher฀annealing฀temperature฀of฀58°C฀was฀employed฀to฀ reduce฀ the฀ chances฀ on฀ non-copy-specific฀ annealing฀ in฀ these฀ reactions.฀ Few฀ areas฀of฀the฀alignment฀displayed฀(small)฀copy-specific฀length฀differences฀suitable฀ trnL 3’ exon<=**************..............................trnL intron=> Persea americana ????????????GGTATGGAAACCTACTAAGTGATAACTTCCAAATT Coelocaryon preussii ????????????GGTATGGAAACCTACTAAGTGGTAACTTCCAAATT Magnolia kobus ????????????GGTAGGGAAACCTAcTAAGTGGTACCTTCCAAATT Eupomatia bennetii ????????????GGTATGGAAACCTACCAAGTGGTAACTTCCAAATT Malmea dielsiana ????????????GGTATGGAAACCTACTAAGTGAGAACTTCCAAATT Cremastosperma brevipes GGATTGAGCCTTGGTATGGAAACCTACTAAGTGAGAACTTCCAAATT Unonopsis elegantissima GGATTGAGCCTTGGTATGGAAACCTACTA-GTCA---CTTCCAAATT Unonopsis stipitata GGATTGAGCCTTGGTATGGAAACCTACTA-GTCA---CTTCCAATCT pseudtrnLF_FOR GGAAACCTACTA-GTCA---CTTCC=> Uno39F GGAAACCTACTAAGTGAGAACTTCC=> <=.........trnL-F intergenic spacer..........=> Persea americana TACAAATGAACATAT-AT----AGGCAAGGAATTTCCATTATTAAAT Coelocaryon preussii CACAAATGAACAGATTAT----GGGCAAGGAATCCCCATTATTGAAT Magnolia kobus TACAAATGCCCATATTAT----GGGCAAGGAATCTCCATTATTGAAT Eupomatia bennetii TACAAATGAACATATTAT----GGGCAAGGAATCTCCATTATTGAAT Malmea dielsiana TACAAATGAACAGATAATGAATGGGC---GAATCTCCACTATTGAAT Cremastosperma brevipes TACAAATGAACAGATAATGAATGGG----GAATCTCCACTATTGAAT Unonopsis elegantissima TACAAATGAACAGATAAT----GGGCAAGGAATCTCCATTATTGAAT Unonopsis stipitata TACAAATGAACAGATAAT----GGGCAAGGAATCTCCATTATTGAAT pseudtrnLF_REV Uno807R <=CAGATAAT----GGGCAAGGAATCTCC <=GAACAGATAATGAATGGGC F i g . 3 . Copy-specific primer sequences, illustrating length difference in pseudtrnL-F intron and spacer. 72 • Chapter 4 for฀primer฀design.฀Possible฀primer฀sequences฀were฀far฀from฀ideal:฀firstly,฀patterns฀ of฀ insertions/deletions฀ inferred฀ in฀Annonaceae฀ tr nL-F฀ sequences฀ suggested฀ that฀annealing฀sites฀may฀not฀be฀present฀in฀taxa฀not฀closely฀related฀to฀Unonopsis฀ (length฀differences฀were฀not฀consistent฀across฀all฀sequences฀available).฀Secondly,฀ primers฀ did฀ not฀ conform฀ to฀ the฀ basic฀ principles฀ of฀ primer฀ design฀ (such฀ as฀ >50%฀GC฀content,฀and฀including฀two฀G/Cs฀on฀the฀3’฀end).฀To฀alleviate฀the฀ latter฀problem,฀PCR฀products฀amplified฀using฀both฀sets฀of฀copy-specific฀primers฀ were฀sequenced฀using฀primers฀trnLF_intFOR฀and฀trnLF_intREV,฀which฀were฀ designed฀to฀anneal฀within฀the฀amplified฀fragments,฀following฀such฀principles฀in฀ order฀to฀improve฀efficiency฀of฀the฀sequencing฀reaction. Analysis Two฀matrices฀were฀constructed:฀1)฀a฀trnL-F฀matrix฀including฀trnL-F฀and฀ pseudtrnL-F฀(thus฀with฀some฀taxa฀present฀twice),฀2)฀a฀multi-gene฀matrix฀also฀ including฀rbcL,฀matK฀and฀psbA-tr nH฀where฀tr nL-F฀and฀pseudtr nL-F฀were฀ treated฀as฀separate฀partitions฀(and฀where฀taxa฀were฀thus฀present฀only฀once).฀In฀ the฀latter฀case฀trnL-F฀and฀pseudtrnL-F฀sequences฀(as฀identified฀from฀analysis฀ of฀the฀first฀matrix)฀were฀both฀aligned฀with฀non-Annonaceae฀outgroup฀trnL-F฀ sequences฀(which฀were฀thus฀present฀in฀both฀partitions),฀and฀where฀Annonaceae฀ pseudtr nL-F฀sequences฀were฀unavailable฀(i.e.฀in฀non-SAC฀clade฀taxa)฀these฀ taxa฀were฀excluded. DNA฀ sequences฀ were฀ edited฀ in฀ SeqMan฀ 4.0฀ (DNAStar฀ Inc.,฀ Madison,฀ WI)฀and฀aligned฀by฀eye.฀All฀areas฀of฀the฀alignment฀where฀the฀assessment฀of฀ homology฀was฀ambiguous฀were฀excluded฀from฀the฀analyses.฀In฀the฀analyses฀where฀ sequences฀of฀tr nL-F฀(in฀Unonopsis)฀or฀pseudtr nL-F฀(in฀other฀Annonaceae)฀ were฀ incomplete฀ (due฀ to฀ the฀ position฀ of฀ the฀ copy-specific฀ primers),฀ the฀ corresponding฀stretches฀of฀the฀alignment฀at฀both฀ends฀were฀excluded.฀ Maximum฀ parsimony฀ (MP)฀ analysis:฀ Data฀ were฀ analysed฀ using฀ the฀ parsimony฀algorithm฀of฀the฀software฀package฀PAUP*฀4.0b10฀(Swofford,฀2000),฀ under฀the฀equal฀and฀unordered฀weights฀criterion฀(Fitch฀parsimony;฀Fitch,฀1971).฀ The฀length฀of฀the฀shortest฀trees฀were฀estimated฀with฀‘full’฀heuristic฀searches฀of฀ 1000฀random฀addition฀sequences,฀TBR,฀saving฀100฀trees฀each฀time.฀Support฀ was฀ estimated฀ using฀ bootstrap฀ analyses฀ of฀ 500฀ replicates฀ with฀‘full’฀ heuristic฀ searches฀of฀100฀random฀addition฀sequences,฀TBR,฀saving฀50฀trees฀each฀time.฀ Bootstrap฀percentages฀were฀interpreted฀following฀Richardson฀et฀al.฀(2004):฀5074%฀represents฀weak฀support,฀75-84%฀moderate฀support฀and฀85-100%฀strong฀ support.฀ For฀ the฀ multi-gene฀ matrix,฀ support฀ was฀ estimated฀ for฀ the฀ markers฀ independently฀and฀in฀combined฀analysis. Selecting฀ the฀ best฀ fitting฀ DNA฀ substitution฀ model:฀ ModelTest฀ 3.06฀ (Posada฀ &฀ Crandall,฀ 1998)฀ was฀ used฀ to฀ select฀ the฀ substitution฀ model฀ best฀ fitting฀each฀sequence฀data฀partition฀for฀each฀matrix฀using฀a฀most฀parsimonious฀ tree฀topology.฀For฀matrix฀1฀ModelTest฀was฀run฀both฀with฀and฀without฀nonA paralogue of tr�L-F in Annonaceae • 73 Annonaceae฀ sequences฀ in฀ order฀ to฀ check฀ whether฀ omitting฀ the฀ outgroups฀ (and฀their฀relatively฀long฀branches)฀could฀have฀resulted฀in฀different฀models฀and฀ parameters. Bayesian฀ analysis:฀ Bayesian฀ inference฀ was฀ applied฀ as฀ implemented฀ in฀ MrBayes฀version฀3.0฀(Huelsenbeck,฀2000).฀The฀use฀of฀Markov฀chain฀Monte฀ Carlo฀analyses฀(MCMC;฀Geyer,฀1991)฀in฀Bayesian฀inference฀facilitates฀heuristic฀ searching฀of฀parameter฀value฀space฀for฀maximum฀likelihood฀models฀of฀DNA฀ substitution฀ in฀ phylogeny฀ reconstruction฀ (Huelsenbeck฀ et฀ al.,฀ 2001).฀ Prior฀ values฀ for฀ the฀ DNA฀ substitution฀ models฀ were฀ applied฀ to฀ each฀ partition฀ (as฀ determined฀ using฀ ModelTest฀ above).฀ Prior฀ probabilities฀ for฀ all฀ topologies฀ were฀equal.฀Persea฀ amer icana฀(Lauraceae)฀was฀chosen฀as฀the฀single฀outgroup฀ taxon฀permitted฀by฀MrBayes฀for฀the฀tr nL-F฀matrix฀(1),฀Coelocaryon฀ preussii฀ (Myristicaceae,฀sister฀group฀to฀rest฀of฀Magnoliales;฀Sauquet฀et฀ al.,฀2003)฀for฀ the฀multi-gene฀matrix฀(2).฀In฀the฀multi-gene฀matrix฀the฀data฀was฀partitioned฀ according฀ to฀ the฀ separate฀ markers฀ and฀ both฀ rates฀ and฀ substitution฀ models฀ allowed฀to฀vary฀across฀the฀partitions.฀MCMC฀analyses฀were฀run฀for฀5,000,000฀ generations฀ with฀ four฀ simultaneous฀ MCMC฀ chains฀ to฀ calculate฀ posterior฀ probabilities฀ (PP),฀ saving฀ one฀ tree฀ per฀ 100฀ generations.฀The฀ burn-in฀ values฀ were฀determined฀empirically฀from฀the฀log-likelihood฀values฀and฀50%฀majority฀ rule฀consensus฀trees฀were฀calculated฀together฀with฀approximations฀of฀the฀PP฀ for฀the฀observed฀bipartitions.฀ r8s฀analyses:฀A฀single฀MP฀topology฀was฀selected฀from฀the฀three฀found฀on฀ heuristic฀search฀(as฀above)฀of฀the฀trnL-F฀matrix฀(1),฀having฀excluded฀all฀missing฀ data฀(leaving฀635฀characters),฀and฀all฀Annonaceae฀taxa฀for฀which฀only฀one฀copy฀ of฀trnL-F฀was฀available฀(leaving฀33฀sequences),฀and฀constraining฀the฀relationships฀ between฀Magnoliales฀outgroups฀to฀conform฀to฀those฀demonstrated฀by฀Sauquet฀ et฀al.฀(2003).฀A฀likelihood฀ratio฀test฀was฀performed฀on฀this฀topology:฀likelihood฀ of฀the฀data฀with฀and฀without฀constraint฀of฀a฀molecular฀clock฀were฀calculated,฀ and฀the฀likelihood฀ratio฀statistic฀compared฀with χ2฀critical฀value฀with฀31฀degrees฀ of฀freedom฀(i.e.,฀number฀of฀sequences฀minus฀2).฀ML฀branch-lengths฀were฀then฀ calculated฀using฀the฀substitution฀model฀calculated฀as฀above฀with฀a)฀the฀original฀ matrix,฀ and฀ b)฀ 100฀ bootstrap฀ re-sampled฀ matrices.฀Thereafter,฀ the฀ penalized฀ likelihood฀(PL)฀method฀of฀Sanderson฀(2002a)฀was฀applied฀using฀the฀program฀ r8s฀(Sanderson,฀2002b)฀to฀estimate฀rates฀and฀divergence฀times฀(with฀standard฀ deviations฀according฀to฀variation฀in฀branch฀lengths฀derived฀from฀the฀bootstrap฀ re-sampled฀data).฀To฀calibrate฀the฀rate฀smoothed฀tree฀in฀absolute฀time฀two฀r8s฀ analyses฀were฀performed฀using฀the฀ages฀of฀different฀fossils฀to฀impose฀minimum฀ ages฀ on฀ basal฀ nodes:฀ 1)฀ Arc haeanthus฀ (Dilcher฀ &฀ Crane,฀ 1984),฀ imposing฀ a฀ minimum฀age฀of฀98฀mya฀for฀the฀Magnoliaceae฀stem฀node.฀2)฀Cronquistiflora฀ and฀Detrusandra฀(Crepet฀&฀Nixon,฀1998),฀following฀Crepet฀et฀ al.฀(2004),฀to฀ impose฀the฀more฀conservative฀minimum฀age฀of฀90฀mya฀on฀the฀Magnoliales฀ crown฀group. 74 • Chapter 4 Secondary structure of the trnL gene and intron The฀secondary฀structures฀of฀the฀trnL฀gene฀and฀intron฀were฀calculated฀for฀ representatives฀of฀both฀tr nL-F฀copies.฀Stem-loop฀regions฀were฀identified฀by฀ comparison฀to฀the฀structure฀proposed฀by฀Borsch฀et฀ al.฀(2003)฀for฀Nymphaea฀ odorata,฀with฀further฀reference฀to฀the฀conserved฀sequence฀motifs฀reported฀across฀ land฀plants฀by฀Quandt฀et฀al.฀(2004).฀Secondary฀structures฀of฀these฀regions฀were฀ then฀estimated฀individually฀using฀Mfold฀(Zuker,฀2003),฀except฀in฀the฀case฀of฀ the฀more฀variable฀P8฀region,฀which฀was฀not฀further฀considered฀as฀the฀precise฀ structure฀ in฀ angiosperms฀ is฀ not฀ yet฀ fully฀ understood฀ (Quandt,฀ pers.฀ com.),฀ rendering฀comparison฀between฀putatively฀conserved฀regions฀impossible.฀ Results and discussion Robustness of the position of Unonopsis in the trnL-F phylogeny BLAST฀search฀(Altschul฀et฀al.,฀1997)฀identified฀chloroplast฀trnL-F฀regions฀ derived฀from฀species฀of฀Magnoliaceae฀-฀a฀family฀of฀the฀same฀order฀as฀Annonaceae;฀ the฀Magnoliales฀(APG฀II,฀2003;฀Sauquet฀et฀ al.,฀2003)฀-฀to฀be฀most฀similar฀to฀ the฀Unonopsis฀trnL-F฀sequences.฀The฀assumption฀of฀identical฀divergence฀rates฀ inherent฀in฀BLAST-based฀methods฀do฀not฀make฀this฀a฀reliable฀method฀to฀infer฀ relatedness฀of฀the฀sequences฀(Thornton฀&฀DeSalle,฀2000).฀However,฀this฀would฀ appear฀to฀exclude฀the฀possibility฀of฀the฀Unonopsis฀trnL-Fs฀being฀descendents฀ of฀a฀more฀recent฀common฀ancestor฀with฀chloroplast฀ tr nL-F฀sequences฀from฀ outside฀the฀taxonomic฀scope฀of฀our฀analyses,฀or฀with฀tRNA฀genes฀known฀from฀ other฀genomic฀compartments฀of฀taxa฀such฀as฀Arabidopsis. Reconstruction฀ of฀ the฀ phylogeny฀ of฀ the฀ entire฀ tr nL-F฀ region฀ using฀ both฀Bayesian฀inference฀and฀MP฀resulted฀in฀topologies฀congruent฀with฀that฀ presented฀ in฀ Fig.฀ 1.฀ Alignment฀ length,฀ numbers฀ of฀ included,฀ variable฀ and฀ parsimony฀informative฀characters฀for฀the฀(entire)฀trnL-F฀matrix,฀CI/RI฀based฀ on฀an฀arbitrary฀MP฀topology,฀and฀best-fitting฀model฀are฀presented฀in฀Table฀1.฀ The฀nodes฀defining฀the฀position฀of฀the฀Unonopsis฀pseudtr nL-F฀sequences฀as฀ monophyletic฀sister฀group฀to฀the฀rest฀of฀the฀Annonaceae฀(A,฀B฀and฀C;฀Fig.฀1)฀ were฀subject฀in฀all฀cases฀to฀strong฀BS฀and฀100%฀PP.฀ ฀ Ta b l e 1 : Details of phylogenetic analysis of the entire trnL-F region Alignment length of which excluded Variable characters Pars� inf� characters �ree length CI/RI Best fitting model 1222 139 359 208 561 0�772/0�881 K81uf+Γ A paralogue of tr�L-F in Annonaceae • 75 Explaining฀the฀incongruent฀position฀of฀Unonopsis฀sequences฀in฀the฀trnLF฀ phylogeny฀ according฀ to฀ errors฀ in฀ the฀ analyses฀ appears฀ unlikely.฀ In฀ order฀ to฀test฀possible฀sensitivity฀of฀the฀result฀to฀alignment฀ClustalX฀(Thompson฀et฀ al.,฀1997)฀was฀applied฀with฀default฀multiple฀alignment฀parameters.฀The฀entire,฀ unedited,฀resulting฀alignment฀was฀analysed฀using฀MP฀and฀Bayesian฀inference,฀ which฀recovered฀nodes฀A,฀B,฀and฀C฀with฀moderate฀to฀ strong฀BS฀and฀>95%฀ PP฀(data฀not฀shown).฀Long฀branch฀attraction฀cannot฀have฀affected฀the฀result,฀ as฀it฀can฀only฀be฀manifested฀through฀the฀attraction฀of฀long฀branches฀(Siddall฀ &฀Whiting,฀1999).฀In฀the฀Annonaceae฀phylogeny฀this฀might฀be฀expected฀to฀be฀ seen฀in฀the฀erroneous฀placement฀of฀highly฀divergent฀sequences฀as฀sister฀groups฀ to฀one฀of฀the฀more฀derived฀LBC฀taxa,฀certainly฀not฀as฀sister฀to฀the฀Annonaceae฀ clade.฀Finally,฀applying฀ModelTest฀with฀and฀without฀non-Annonaceae฀sequences฀ resulted฀in฀the฀same฀best฀fitting฀substitution฀model฀(K81uf+Γ).฀ A paralogue of trnL-F in Annonaceae The฀ PCR-based฀ approach฀ employed฀ here฀ resulted฀ in฀ the฀ amplification฀ and฀sequencing฀of฀trnL-F฀in฀Unonopsis฀(forming฀a฀monophyletic฀group฀with฀ Bocageopsis฀and฀Onychopetalum฀in฀the฀SBC),฀and฀of฀pseudtrnL-F฀in฀accessions฀ of฀Bocageopsis,฀Cremastosper ma,฀Malmea,฀Onyc hopetalum฀and฀Oxandra:฀all฀ SBC฀taxa฀of฀the฀SAC฀clade฀forming฀a฀monophyletic฀group฀with฀Unonopsis฀ pseudtr nL-F฀sequences฀(Fig.฀4).฀Copy-specific฀amplification฀was฀not฀always฀ successful:฀in฀some฀taxa฀of฀the฀SAC฀clade฀(e.g.฀Pseudoxandra),฀and฀accessions฀of฀ further฀SBC฀taxa฀(such฀as฀Annickia฀and฀Polyalthia),฀the฀pseudtrnL-F฀primers฀ instead฀amplified฀the฀original฀trnL-F฀copy.฀When฀applied฀to฀accessions฀of฀LBC฀ or฀basal฀grade฀taxa,฀no฀amplified฀product฀was฀produced.฀It฀is฀possible฀that฀the฀ small฀length฀differences฀(indels)฀used฀as฀targets฀for฀the฀copy฀specific฀primers฀ are฀ only฀ present฀ in฀ the฀ SAC฀ clade,฀ either฀ representing฀ synapomorphies,฀ or฀ sympleisiomorphies฀secondarily฀lost฀in฀the฀other฀clades฀sampled.฀This฀result฀ offers฀no฀direct฀evidence฀for฀the฀presence฀or฀otherwise฀of฀pseudtrnL-F฀in฀other฀ clades฀in฀Annonaceae. The฀ location฀ of฀ a฀ duplication฀ event฀ on฀ a฀ gene฀ tree฀ represents฀ its฀ most฀ recent฀ possible฀ age฀ (Thornton฀ &฀ DeSalle,฀ 2000),฀ which฀ in฀ the฀ case฀ of฀ the฀ divergence฀ of฀ tr nL-F฀ and฀ pseudtr nL-F฀ is฀ more฀ recent฀ than฀ the฀ MRCA฀ of฀ Annonaceae฀and฀Eupomatiaceae,฀but฀prior฀to฀the฀generally฀accepted฀MRCA฀ of฀Annonaceae฀as฀represented฀by฀the฀stem฀lineage฀of฀Anaxagorea.฀This฀latter฀ node฀has฀been฀estimated฀at฀57฀or฀69฀mya฀(Doyle฀et฀al.,฀2004)฀and฀64.9฀or฀76.8฀ mya฀ (Richardson฀ et฀ al.,฀ 2004),฀ using฀ molecular฀ dating฀ with฀ different฀ fossil฀ calibrations.฀Similar฀analyses฀performed฀here฀primarily฀to฀estimate฀the฀rates฀of฀ evolution฀of฀the฀two฀copies฀of฀tr nL-F฀(see฀below)฀were฀calibrated฀with฀two฀ different฀fossils฀to฀estimate฀the฀age฀of฀the฀divergence฀in฀absolute฀terms.฀This฀ resulted฀in฀ages฀of฀81.18฀mya฀using฀the฀fossil฀Archaeanthus,฀following฀Doyle฀et฀ al.฀(2004)฀and฀Richardson฀et฀al.฀(2004),฀and฀58.30฀mya฀using฀Cronquistiflora฀ 76 • Chapter 4 �ersea�amer��a�a Coe�o�ar�o��pre�ss�� �a�����m�ma��e��ra�ea�a �e�e�er�a�rose���ora L�r�o�e��ro������e�se 0�62 50 0�99 78 �a��o��a��o��s ��pomat�a��e��et�� ��a��ra�esp��ta�a 0�66 1�00 93 �����opeta��m�ama�o����m ��o�ops�s�e�e�a�t�ss�ma 0�98 60 0�83 74 �o�a�eops�s�m��t���ora 0�98 62 1�00 100 ��o�ops�s�p�tt�er� ��o�ops�s�st�p�tata 0�98 80 1�00 91 1�00 100 0�96 66 1�00 98 �a�mea���e�s�a�a �a�mea�sp� �a�mea�s�r��ame�s�s 1�00 84* Cremastosperma��re��pes 0�99 82 Cremastosperma��a�����or�m Cremastosperma�ma�ro�arp�m Cremastosperma��e�op�����m ��a�a�orea�p�aeo�arpa Ca�a��a�o�orata 0�99 80 C�m�opeta��m�sp� 0�98 59 1�00 85 1�00 89 1�00 98 ����et�a���r�sea ��ar�a������a �a�mea�s�r��ame�s�s 1�00 99 1�00 85* 1�00 81 0�96 0�94 57 51 0�65 �a�mea���e�s�a�a �a�mea�sp� �����opeta��m�ama�o����m �o�a�eops�s�m��t���ora ��o�ops�s�e�e�a�t�ss�ma 0�64 ��o�ops�s�p�tt�er� ��o�ops�s�st�p�tata ��a��ra�esp��ta�a 0�90 56 Cremastosperma��a�����or�m 0�99 77 Cremastosperma��e�op�����m 0�97 66 Cremastosperma��re��pes Cremastosperma�ma�ro�arp�m 0.1 Fig. 4: Bayesian 50% majority rule consensus tree and phylogram resulting from analysis of partial trnL-F/pseudtrnL-F. Posterior probabilities are indicated above the nodes, maximum parsimony proefschrift Pirie Chap4 fig 4�a-b�ai mrt 2005 bootstrap support (where >50%) below. Bootstrap support values for conflicting nodes referred to in the text are highlighted in bold and with an asterisk (*) and฀Detrusandria฀following฀Crepet฀et฀ al.฀(2004)฀respectively,฀with฀a฀standard฀ deviation฀ of฀ 7.38.฀ In฀ the฀ same฀ analyses฀ the฀ age฀ of฀ SAC฀ clade฀ crown฀ group฀ was฀also฀estimated,฀as฀represented฀by฀two฀different฀nodes:฀1)฀the฀MRCA฀of฀ all฀ the฀ Annonaceae฀ tr nL-F฀ sequences฀ included฀ and฀ 2)฀ the฀ MRCA฀ of฀ the฀ corresponding฀pseudtrnL-F฀sequences.฀The฀resulting฀ages฀were฀not฀significantly฀ different฀(Table฀2).฀ A paralogue of tr�L-F in Annonaceae • 77 Ta b l e 2 : r8s analyses: trnL-F divergence MRCA SAC clade: MRCA SAC clade: trnL-F pseudtrnL-F 34�25 mya sd=7�68 31�61 mya sd=5�94 Age: calibration 1 81�18 mya sd=7�38 Age: calibration 2 58�30 mya sd=7�38 21�68 mya sd=7�68 18�08 mya sd=5�94 *Rate 0�000612 0�000461 0�000744 estimated: mean: 0�000709 0�000490 0�000907 sd: 0�000085 0�000106 0�00014 -1 -1 * changes site million years Phylogenetic utility of Annonaceae trnL-F copies Duplicated฀genes฀have฀proved฀useful฀in฀rooting฀phylogenetic฀analyses,฀in฀ particular฀those฀in฀which฀there฀are฀no฀extant฀outgroups฀(an฀extreme฀example฀ being฀ that฀ of฀ the฀ tree฀ of฀ life)฀ or฀ where฀ outgroups฀ are฀ too฀ distant฀ (such฀ as฀ in฀angiosperms;฀Donoghue฀&฀Mathews,฀1998;฀Mathews฀&฀Donoghue,฀1999).฀ Character฀ polarisation฀ in฀ Annonaceae฀ is฀ uncontroversial:฀ Anaxagorea฀ is฀ supported฀as฀first฀branching฀lineage฀in฀cladistic฀analyses฀of฀both฀morphological฀ and฀of฀molecular฀data฀(Doyle฀&฀Le฀Thomas,฀1996;฀Doyle฀et฀al.,฀2000;฀Sauquet฀ et฀ al.,฀2003).฀However,฀were฀the฀root฀of฀the฀Annonaceae฀to฀be฀misplaced฀this฀ could฀ provide฀ an฀ explanation฀ (where฀ no฀ other฀ is฀ yet฀ forthcoming)฀ for฀ the฀ increased฀ rate฀ of฀ evolution฀ (of฀ cpDNA฀ markers)฀ apparent฀ in฀ the฀ LBC.฀ No฀ such฀ conclusion฀ can฀ be฀ drawn฀ from฀ analyses฀ performed฀ here:฀ phylogenetic฀ analysis฀of฀trnL-F฀and฀pseudtrnL-F฀sequences฀resulted฀in฀a฀gene฀tree฀in฀which฀ this฀rooting฀is฀clearly฀not฀disrupted฀(see฀Figs฀1฀&฀4).฀Relationships฀between฀ taxa฀represented฀twice฀were฀mostly฀congruent฀(Fig.฀4),฀with฀any฀apparently฀ conflicting฀nodes฀subject฀to฀PP฀<95%฀and/or฀BS฀<75%.฀Analysis฀of฀rbcL,฀matK฀ and฀psbA-trnH฀sequences฀separately฀also฀revealed฀no฀supported฀conflict.฀These฀ partitions฀were฀therefore฀included฀in฀a฀combined฀analysis฀which฀resulted฀in฀a฀ topology฀congruent฀with฀those฀of฀trnL-F฀and฀pseudtrnL-F,฀with฀two฀apparent฀ exceptions฀(Fig.฀5a,฀compare฀with฀Fig.฀4):฀1)฀relationships฀between฀species฀of฀ Malmea,฀which฀were฀in฀conflict฀with฀those฀revealed฀by฀tr nL-F,฀and฀2)฀those฀ between฀species฀of฀Cremastosperma,฀which฀were฀in฀conflict฀with฀those฀revealed฀ by฀pseudtr nL-F฀(BS฀values฀of฀corresponding฀nodes฀are฀bold฀and฀highlighted฀ with฀asterisks฀in฀Figs฀4฀and฀5).฀ Combined฀ analysis฀ resulted฀ in฀ (generally)฀ higher฀ support฀ for฀ a฀ more฀ resolved฀topology฀(Fig.฀5b),฀with฀the฀exception฀of฀lower฀support฀for฀relationships฀ between฀species฀of฀Malmea. Proportions฀of฀variable฀and฀parsimony฀informative฀characters฀and฀CI/RI฀ (based฀on฀the฀topology฀derived฀from฀combined฀analysis)฀are฀presented฀in฀Table฀3.฀ Of฀the฀five฀markers฀compared,฀the฀roughly฀600฀bp฀long฀fragment฀of฀pseudtrnL78 • Chapter 4 A Coe�o�ar�o��pre�ss�� �a��o��a��o��s �a�mea�sp� 1�00 100 1�00 90* �a�mea���e�s�a�a �a�mea�s�r��ame�s�s ��a��ra�esp��ta�a 1�00 100 Cremastosperma��re��pes 1�00 98 �0�50 68 Cremastosperma��a�����or�m 0�98 76* Cremastosperma��e�op�����m Cremastosperma�ma�ro�arp�m ��o�ops�s�e�e�a�t�ss�ma 1�00 98 ��o�ops�s�p�tt�er� ��o�ops�s�st�p�tata 1�00 95 0�87 69 B �����opeta��m�ama�o����m �o�a�eops�s�m��t���ora Coe�o�ar�o��pre�ss�� �a��o��a��o��s 1�00 100 1�00 68* 0�92 57 1�00 100 1�00 100 1�00 75 0�99 81 �a�mea�sp� �a�mea���e�s�a�a �a�mea�s�r��ame�s�s Cremastosperma��re��pes Cremastosperma�ma�ro�arp�m Cremastosperma��a�����or�m Cremastosperma��e�op�����m ��a��ra�esp��ta�a 0�98 72 ��o�ops�s�e�e�a�t�ss�ma 1�00 100 1�00 98 0�93 77 1�00 81 ��o�ops�s�p�tt�er� ��o�ops�s�st�p�tata �����opeta��m�ama�o����m �o�a�eops�s�m��t���ora F i g . 5 : Bayesian / Maximum parsimony bootstrap consensus trees of a) combined chloroplast DNA markers rbcL, matK and psbA-trnH and b) rbcL, matK, psbA-trnH, trnL-F and pseudtrnL-F. Bootstrap support values for conflicting nodes referred to in the text are highlighted in bold and with an asterisk (*) F฀amplified฀using฀the฀copy฀specific฀primers฀provided฀the฀highest฀proportion฀of฀ parsimony฀informative฀characters:฀more฀in฀total฀than฀rbcL,฀which฀is฀rather฀more฀ than฀twice฀as฀long฀and฀has฀to฀be฀amplified฀and฀sequenced฀in฀two฀pieces.฀The฀ phylogenetic฀ utility฀ of฀ pseudtr nL-F฀ would฀ also฀ appear฀ clear฀ comparing฀ the฀ resolution฀and฀support฀values,฀particularly฀within฀Cremastosperma,฀as฀presented฀ in฀Fig.฀4.฀The฀limited฀conflict฀apparent฀in฀the฀topologies฀(see฀Figs฀4฀and฀5)฀is฀ A paralogue of tr�L-F in Annonaceae • 79 not฀consistently฀between฀one฀partition฀and฀the฀others.฀It฀may฀be฀a฀result฀of฀ either฀the฀small฀numbers฀of฀informative฀characters฀involved,฀or฀the฀very฀limited฀ taxon฀sampling,฀or฀both.฀However,฀in฀comparing฀the฀two฀tr nL-F฀paralogues฀ with฀the฀single฀corresponding฀copy฀found฀in฀other฀plants,฀two฀further฀levels฀ of฀ homology฀ can฀ be฀ considered:฀ functional฀ and฀ positional.฀These฀ may฀ have฀ implications฀for฀the฀patterns฀of฀evolution฀displayed฀which฀might฀lead฀us฀to฀treat฀ this฀new฀potential฀marker฀with฀some฀caution.฀ Ta b l e 3 : Comparison of markers use in combined analyses Marker Approx. Variable Pars. inf. sequence characters: characters: length No. / % No. / % 600 74 / 12�3 27 / 4,5 0�94/0�88 600 89 / 14�8 50 / 8�3 0�92/0�92 rbcL 1500 122 / 8�1 44 / 2�9 0�83/0�70 matK 850 136 / 16 60 / 7�1 0�95/0�92 (partial) CI/RI trnL-F (partial) p s e u d trnL-F psbA-trnH 430 61 / 14�2 10 / 2�3 0�93/0�79 combined 3980 482 / 12�1 191 / 4�8 0�91/0�86 CI/RI calculated using single most parsimonious topology of combined analysis. Functional homology of Annonaceae trnL-F copies Although฀ examples฀ of฀ paralogues฀ of฀ chloroplast฀ genes฀ have฀ been฀ documented฀ where฀ function฀ in฀ one฀ copy฀ (often฀ to฀ be฀ found฀ in฀ a฀ different฀ genomic฀compartment)฀appears฀to฀have฀been฀lost฀entirely฀(see฀below),฀genomewide฀studies฀suggest฀that฀the฀vast฀majority฀of฀redundant฀genes฀still฀experience฀ purifying฀selection฀(Wagner,฀2002).฀Loss฀of฀function฀in฀protein฀coding฀genes฀ can฀ in฀ many฀ cases฀ be฀ demonstrated,฀ for฀ example฀ where฀ mutations฀ result฀ in฀ disruption฀of฀the฀reading฀frame฀or฀the฀appearance฀of฀‘stop’฀codons฀within฀it.฀ The฀ function฀ of฀ the฀ transfer฀ RNA฀ for฀ which฀ the฀ tr nL฀ gene฀ itself฀ codes฀ is฀ related฀ to฀ its฀ secondary฀ structure฀ (and฀ also฀ that฀ of฀ the฀ intron฀ found฀ within฀ it).฀We฀therefore฀attempted฀to฀assess฀the฀functionality฀of฀copies฀of฀tr nL-F฀in฀ Annonaceae฀by฀comparison฀with฀plant฀tr nL฀introns฀and฀3’฀exons฀for฀which฀ secondary฀structures฀have฀been฀proposed฀(by฀Borsch฀et฀ al.,฀2003;฀Quandt฀et฀ al.,฀2004;฀Quandt฀&฀Stech,฀2004). Borsch฀et฀ al.฀(2003)฀found฀across฀basal฀angiosperms฀only฀20%฀of฀the฀95฀ positions฀corresponding฀to฀proposed฀stem฀structures฀in฀the฀tr nL฀intron฀to฀be฀ variable,฀attesting฀to฀the฀conservation฀of฀its฀secondary฀structure.฀The฀structure฀ 80 • Chapter 4 A A A A U AUU G G G C A A U C Q Oxandra, Malmea G A A G C U G A A A G G U C C C A A A P G A G A C C A U G R1 U U G A A G A A A A U A U G C C C G U A Unonopsis U stipitata A G A U G A U A U A C G A A P6 G C A A A A A U A A C A AG C A G U U A U A C G A A G G A C AAG G G U A U A U G U U G IGS U A C G A G G U AA A UUU AU A G A U A G A G G G S A U A G C C G A A C C GUGAGA U A A DELETIONS U U G C G A U U A U G C G C A A A P9 A C A G A CA U G U C A AU A C A U C U U A C R2 C G A DELETION Malmea C G and M. surinamensis U A G C A U G C A A A C G C G C U A U A G C A C CA U G U U C A U A A G P8 263-270 bp A, AA or AAA F i g . 6 : Secondary structure of the trnL intron sequence of Cremastosperma brevipes. Conserved sequence motifs are outlined and labelled following Cech et al. (1994). Selected differences between this representative trnL-F sequence and all or specified pseudtrnL-F sequences are outlined and labelled, with deletions filled in grey. A paralogue of tr�L-F in Annonaceae • 81 presented฀in฀Fig.฀6฀(following฀Cech฀et฀al.฀(1994),฀based฀on฀the฀secondary฀structure฀ of฀group฀I฀introns฀modelled฀by฀Michel฀&฀Westhof,฀(1990)),฀is฀that฀of฀the฀trnL฀ intron฀sequence฀of฀Cremastosperma฀brevipes฀(see฀table฀of฀samples฀in฀Appendix฀A).฀฀ The฀ conserved฀ sequence฀ motifs,฀ as฀ described฀ by฀ Quandt฀ et฀ al.฀ (2004),฀ and฀ selected฀differences฀between฀this฀and฀pseudtrnL-F฀sequences฀are฀indicated. The฀greater฀number฀of฀parsimony฀informative฀characters฀in฀pseudtrnL-F฀ sequences฀in฀comparison฀to฀trnL-F฀sequences฀is฀reflected฀in฀a฀higher฀number฀ of฀changes฀occurring฀in฀regions฀inferred฀to฀represent฀stem฀structures฀as฀opposed฀ to฀loops฀in฀Nymphaea.฀Excluding฀the฀P6฀and฀P8฀regions฀(for฀which฀secondary฀ structures฀in฀both฀copies฀appeared฀to฀differ฀significantly฀from฀that฀of฀Nymphaea),฀ 11฀single฀base฀changes฀are฀observed฀in฀pseudtr nL-F฀putative฀stem฀regions,฀in฀ comparison฀to฀5฀in฀the฀same฀taxa฀for฀trnL-F.฀No฀compensatory฀changes฀were฀ apparent฀in฀either฀tr nL-F฀or฀pseudtr nL-F฀sequences฀analysed.฀The฀few฀base฀ changes฀apparent฀in฀the฀3’฀exon฀relative฀to฀that฀of฀Nymphaea฀are฀limited฀to฀ within฀loop฀structures฀in฀both฀copies.฀More฀significantly,฀a฀9฀bp฀long฀deletion฀ is฀observed฀in฀the฀pseudtr nL-F฀sequences฀of฀two฀of฀the฀Malmea฀accessions฀ (representing฀a฀synapomorphy฀for฀a฀clade฀also฀supported฀by฀rbcL,฀psbA-tr nH฀ and฀matK฀data฀but฀apparently฀contradicted฀by฀trnL-F).฀This฀deletion฀includes฀ most฀of฀the฀conserved฀R1฀and฀R2฀sequence฀motif฀(as฀described฀by฀Quandt฀et฀ al.,฀2004;฀see฀Fig.฀6),฀and฀thus฀seems฀likely฀to฀represent฀a฀significant฀change฀in฀ the฀secondary฀structure.฀ Rate of evolution of Annonaceae trnL-F copies The฀role฀of฀selection฀can฀be฀investigated฀by฀estimating฀relative฀and฀absolute฀ rates฀of฀sequence฀divergence฀for฀different฀branches฀in฀the฀gene฀family฀tree.฀If฀ mutation฀ rate฀ is฀ constant,฀ differences฀ in฀ divergence฀ rates฀ represent฀ strength฀ of฀selection฀(Thornton฀&฀DeSalle,฀2000).฀Under฀the฀best฀fitting฀substitution฀ model฀ (K81uf+ Γ),฀ the฀ likelihood฀ of฀ the฀ partial฀ tr nL-F฀ data฀ given฀ one฀ of฀ the฀ MP฀ topologies฀ was฀ 2022.27.฀ Enforcing฀ a฀ molecular฀ clock฀ resulted฀ in฀ a฀ significantly฀ different฀ likelihood฀ of฀ 2074.00฀ (P=<0.01,฀ chi฀ squared฀ test,฀ 31฀ degrees฀of฀freedom),฀and฀the฀clock฀hypothesis฀was฀thus฀rejected.฀We฀therefore฀ used฀the฀penalized฀likelihood฀method฀of฀Sanderson฀(2002a)฀to฀estimate฀rates฀of฀ evolution฀in฀different฀branches฀of฀the฀trnL-F฀gene฀tree,฀applying฀a฀bootstrapping฀ technique฀ to฀ assess฀ error฀ according฀ to฀ character฀ sampling.฀The฀ rate฀ at฀ SAC฀ crown฀node฀was฀estimated฀to฀be฀0.000744฀(sd=0.000146)฀changes฀per฀position฀ per฀million฀years฀for฀pseudtrnL-F,฀significantly฀higher฀than฀that฀estimated฀for฀ trnL-F:฀0.000461฀(sd=0.000106). Should฀we฀be฀comparing฀two฀chloroplast฀encoded฀markers,฀the฀difference฀ observed฀might฀thus฀be฀interpreted฀to฀suggest฀less฀stringent฀selection฀acting฀on฀ pseudtrnL-F.฀However,฀although฀one฀of฀the฀two฀copies฀of฀trnL-F฀in฀Annonaceae฀ is฀presumably฀to฀be฀found฀in฀the฀expected฀position฀in฀the฀chloroplast฀genome,฀ the฀position฀of฀the฀other฀(whichever฀that฀might฀be)฀is฀unknown.฀Relative฀rate฀ 82 • Chapter 4 differences฀ can฀ be฀ attributable฀ to฀ other฀ evolutionary฀ or฀ population฀ genetic฀ phenomena฀(Small฀et฀al.,฀1998),฀some฀of฀which,฀such฀as฀background฀mutational฀ processes,฀drift,฀and฀rates฀of฀recombination,฀differ฀across฀genomic฀compartments.฀ The฀ positional฀ homology฀ of฀ both฀ copies฀ is฀ thus฀ critical฀ to฀ interpreting฀ this฀ higher฀rate฀of฀change.฀It฀also฀determines฀their฀mode฀of฀inheritance฀and฀is฀thus฀ critical฀for฀the฀interpretation฀of฀the฀phylogenetic฀signal฀they฀contain. Positional homology in Annonaceae trnL-F copies There฀ are฀ numerous฀ examples฀ of฀ duplicated฀ chloroplast฀ genes฀ in฀ the฀ literature,฀the฀position฀of฀many฀of฀which฀has฀been฀demonstrated.฀A฀pseudogene฀ of฀r pl23฀is฀present฀as฀in฀the฀chloroplasts฀of฀many฀members฀of฀Caryophyllales฀ (Clegg฀&฀Zurawski,฀1992),฀and฀partial฀or฀complete฀pseudogenes฀of฀trnF฀have฀ been฀ observed฀ as฀ insertions฀ in฀ the฀ tr nL-F฀ spacer฀ in฀ Microser is฀ (Asteraceae,฀ Vijverberg฀&฀Bachmann,฀1999)฀and฀particular฀lineages฀of฀Brassicaceae฀(Dobes฀ et฀ al.,฀ 2004;฀ Koch฀ et฀ al.,฀ 2005).฀Ayliffe฀ et฀ al.฀ (1998)฀ identified฀ sequences฀ of฀ plastid฀ homology฀ in฀ the฀ nuclear฀ genome฀ of฀ various฀ angiosperms,฀ as฀ did฀ Millen฀et฀ al.฀(2001),฀demonstrating฀transfers฀of฀infA฀from฀the฀chloroplast฀to฀ the฀nucleus.฀ Gene฀ content฀ of฀ the฀ mitochondrial฀ genome฀ is฀ considered฀ particularly฀ dynamic฀ and฀ flexible฀ (Nakazono฀ &฀ Hirai,฀ 1993).฀ Cummings฀ et฀ al.฀ (2003)฀ reported฀hundreds฀of฀successful฀transfers฀of฀rbcL฀from฀the฀chloroplast฀to฀the฀ mitochondrion฀ in฀ flowering฀ plants:฀ of฀ the฀ five฀ examined,฀ all฀ had฀ disrupted฀ reading฀ frames.฀ Nakazono฀ &฀ Hirai฀ (1993)฀ compared฀ mitochondrial฀ and฀ chloroplast฀gene฀content฀in฀rice:฀16฀chloroplast฀sequences฀between฀32฀bases฀and฀ 6.8฀kb฀long฀(equivalent฀in฀total฀to฀19%฀of฀rice฀total฀chloroplast฀genome)฀were฀ discovered฀ in฀ the฀ mitochondrion.฀The฀ nine฀ intact฀ and฀ three฀ defunct฀ tRNA฀ genes฀ discovered฀ included฀ the฀ 3’฀ tr nL฀ exon฀ and฀ tr nF.฀With฀ the฀ publication฀ of฀the฀complete฀rice฀mitochondrial฀genome,฀Notsu฀et฀al.฀(2002)฀discovered฀a฀ total฀of฀17฀tRNA฀genes฀and฀five฀pseudo฀tRNA฀sequences฀of฀chloroplast฀origin.฀ They฀additionally฀identified฀nuclear฀sequences฀of฀chloroplast฀origin฀positioned฀ adjacent฀to฀sequences฀of฀mitochondrial฀origin,฀suggesting฀transfer฀to฀have฀been฀ via฀the฀mitochondrion. Further approaches to locating copies of trnL-F in Annonaceae A฀number฀of฀possible฀approaches฀might฀identify฀the฀positional฀homology฀ of฀trnL-F฀copies฀in฀Annonaceae.฀Regions฀surrounding฀trnL-F฀and฀pseudtrnLF฀might฀contain฀adjacent฀sequences฀of฀identifiable฀origin,฀or฀those฀which฀might฀ suggest฀a฀mechanism฀by฀which฀a฀duplication฀could฀have฀occurred,฀such฀as฀mobile฀ genetic฀elements฀(Thornton฀&฀DeSalle,฀2000).฀However,฀either฀region฀might฀ represent฀part฀of฀a฀significantly฀larger฀duplicated฀fragment,฀and฀this฀approach฀ would฀in฀any฀case฀be฀likely฀to฀suffer฀either฀from฀similarity฀(amplification฀of฀ A paralogue of tr�L-F in Annonaceae • 83 the฀ wrong฀ copy)฀ or฀ parallel฀ divergence฀ (failure฀ of฀ amplification)฀ of฀ primer฀ annealing฀sequences. It฀ might฀ be฀ possible฀ to฀ determine฀ the฀ location฀ of฀ the฀ paralogues฀ using฀ southern฀blotting฀techniques฀with฀mitochondrial฀or฀nuclear฀DNA฀preparations฀ and฀PCR฀products฀as฀a฀probe.฀However,฀the฀quality฀of฀DNA฀samples฀available฀ for฀SBC฀taxa฀is฀relatively฀low฀(very฀few฀are฀known฀from฀cultivation),฀which฀ might฀ present฀ challenges฀ for฀ isolating฀ mitochondrial,฀ chloroplast฀ or฀ nuclear฀ DNA.฀ Fresh฀ material฀ of฀ other฀Annonaceae฀ taxa฀ is฀ available฀ from฀ cultivated฀ specimens฀ (mostly฀ LBC฀ taxa฀ such฀ as฀ Annona฀ or฀ Monodora).฀ However,฀ pseudtrnL-F฀sequences,฀if฀present฀(not฀subject฀to฀secondary฀loss)฀in฀LBC฀taxa,฀ might฀be฀highly฀divergent฀in฀comparison฀to฀those฀of฀the฀SBC฀or฀the฀other฀ tr nL-F฀copy.฀Similarity฀less฀than฀70%฀might฀result฀in฀failure฀of฀the฀gel฀blot฀ (Fulton฀et฀al.,฀2002). Conclusions The฀discovery฀of฀a฀paralogous฀copy฀of฀trnL-F฀in฀Annonaceae฀poses฀at฀the฀ same฀ time฀ theoretical฀ and฀ practical฀ problems฀ for฀ phylogeny฀ reconstruction.฀ When฀ using฀ chloroplast฀ markers฀ in฀ Annonaceae,฀ single฀ regions฀ on฀ their฀ own฀(as฀defined฀by฀the฀primers฀available฀to฀amplify฀and฀sequence฀them,฀and฀ generally฀ of฀ less฀ than฀ 2kb฀ length)฀ almost฀ always฀ provide฀ insufficient฀ total฀ numbers฀of฀informative฀characters฀to฀arrive฀at฀supported฀topologies.฀However,฀ the฀ most฀ densely฀ sampled฀ phylogenetic฀ studies฀ of฀Annonaceae฀ to฀ date฀ (i.e.฀ Mols฀ et฀ al.,฀ 2004;฀ Richardson฀ et฀ al.,฀ 2004)฀ rely฀ on฀ the฀ tr nL-F฀ region฀ for฀ a฀large฀proportion฀of฀the฀informative฀characters฀analysed฀(the฀rest฀provided฀ by฀rbcL).฀Paralogy฀of฀tr nL-F฀sequences฀for฀Unonopsis฀with฀respect฀to฀other฀ available฀Annonaceae฀ tr nL-F฀ sequences,฀ might฀ mean฀ a฀ large฀ source฀ of฀ data฀ would฀be฀excluded฀from฀analyses฀in฀which฀the฀positions฀of฀its฀species฀are฀of฀ interest฀(i.e.฀monophyly฀of฀Bocageopsis฀and฀Onychopetalum฀or฀phylogeny฀of฀ the฀ SBC).฀ More฀ worryingly,฀ although฀ no฀ further฀‘hard’฀ conflicts฀ have฀ been฀ observed฀ in฀ phylogeny฀ reconstruction฀ in฀Annonaceae,฀ the฀ known฀ existence฀ of฀at฀least฀one฀paralogous฀copy฀of฀tr nL-F,฀in฀combination฀with฀the฀limited฀ resolution฀provided฀by฀currently฀available฀alternative฀markers฀(for฀most฀taxa฀ rbcL฀alone),฀might฀then฀call฀into฀question฀results฀derived฀from฀using฀tr nL-F฀ alone฀and฀perhaps฀also฀in฀combined฀analyses฀in฀which฀it฀provides฀the฀majority฀ of฀informative฀characters. The฀ presence฀ of฀ these฀ two฀ copies฀ of฀ tr nL-F฀ in฀Annonaceae฀ is฀ not฀ be฀ a฀ problem฀ for฀ phylogeny฀ reconstruction:฀ pseudtr nL-F฀ is฀ identifiable฀ on฀ phylogenetic฀analysis฀as฀forming฀a฀monophyletic฀sister-group฀to฀the฀Annonaceae฀ trnL-F฀clade,฀and฀primers฀developed฀during฀this฀study฀allow฀the฀amplification฀ 84 • Chapter 4 of฀ tr nL-F฀ in฀ Unonopsis.฀ The฀ phylogenetically฀ informative฀ characters฀ in฀ pseudtrnL-F฀sequences฀may฀be฀an฀unforeseen฀benefit฀of฀this฀phenomenon.฀This฀ marker฀may฀have฀limited฀taxonomic฀scope฀(thus฀far฀amplification฀has฀only฀been฀ successful฀in฀a฀limited฀number฀of฀closely฀related฀taxa),฀but฀contains฀a฀relatively฀ large฀amount฀of฀information฀for฀the฀length฀of฀sequence.฀Its฀phylogenetic฀signal฀ appears฀to฀be฀complementary฀to฀that฀of฀chloroplast฀markers,฀though฀further฀ sampling฀of฀taxa฀is฀necessary฀to฀test฀this฀further.฀Should฀both฀copies฀of฀trnL-F฀ be฀found฀in฀the฀chloroplast,฀or฀one฀transferred฀to฀the฀mitochondrion,฀maternal฀ inheritance฀should฀result฀in฀congruent฀gene฀trees.฀However,฀if฀one฀copy฀has฀ been฀transferred฀to฀the฀nucleus,฀its฀different฀mode฀of฀inheritance฀might฀result฀ in฀a฀conflicting฀gene฀phylogeny.฀This฀itself฀could฀offer฀valuable฀insight฀into฀the฀ evolutionary฀history฀of฀clades฀in฀Annonaceae. Results฀presented฀here฀serve฀to฀confirm฀the฀caution฀with฀which฀datasets฀ dominated฀by฀characters฀from฀one฀marker฀should฀be฀interpreted.฀Even฀widely฀ used฀ cpDNA฀ markers฀ such฀ as฀ the฀ tr nL-F฀ region฀ can฀ result฀ in฀ a฀ gene฀ tree฀ at฀ odds฀ with฀ those฀ of฀ other฀ markers.฀ In฀ the฀ case฀ of฀ tr nL-F฀ in฀Annonaceae,฀ the฀paralogous฀copies฀discovered฀appear฀to฀have฀a฀higher฀rate฀of฀change฀than฀ those฀ orthologous฀ with฀ Annonaceae฀ tr nL-F฀ sequences฀ previously฀ known.฀ This฀difference,฀and฀possibly฀also฀the฀positions฀of฀these฀changes฀(in฀particular฀ the฀ deletion฀ common฀ to฀ two฀ Malmea฀ accessions฀ in฀ a฀ conserved฀ region฀ of฀ the฀ intron฀ secondary฀ structure),฀ may฀ indicate฀ relaxed฀ selectional฀ constraint.฀ Alternatively,฀it฀could฀reflect฀the฀higher฀ambient฀mutation฀rates฀characteristic฀ of฀ the฀ nuclear฀ genome.฀ Interpretation฀ of฀ the฀ functional฀ homology฀ of฀ the฀ trnL-F฀and฀pseudtrnL-F฀regions,฀as฀well฀as฀that฀of฀their฀phylogenetic฀signal,฀is฀ therefore฀dependent฀on฀their฀positional฀homology,฀i.e.฀within฀the฀Annonaceae฀ genome,฀and฀worthy฀of฀further฀study. A paralogue of tr�L-F in Annonaceae • 85 86 • Chapter 4 Chapter 5 R ev i s i o n a n d p hy l o g e ny o f Cremastosperma (Annonaceae) Michael D. Pirie, Sanna Kankainen & Paul J. M. Maas Nat i onaal Her bar i um Neder l and, Uni ver s i t ei t Ut r ec ht br anc h, Hei del ber gl aan 2, 3584 CS Ut r ec ht , The Net her l ands Revision and phylogeny of Cremastosperma • 87 Abstract A฀ taxonomic฀ revision฀ and฀ phylogeny฀ based฀ on฀ DNA฀ sequence฀ data฀ is฀ presented฀for฀the฀Neotropical฀genus฀Cremastosperma฀(Annonaceae).฀Twentynine฀species฀are฀recognised,฀the฀majority฀occurring฀in฀lowland฀to฀premontane฀ wet฀ forest฀ in฀ areas฀ surrounding฀ the฀ Andean฀ mountain฀ chain.฀ Clades฀ are฀ identified฀corresponding฀to฀species฀distributed฀either฀west฀or฀east฀of฀the฀Andes,฀ suggesting฀the฀Andean฀orogeny฀to฀represent฀an฀important฀vicariance฀event฀in฀ the฀evolution฀of฀the฀group. Keywords:฀Annonaceae,฀Cremastosperma,฀Neotropics,฀phylogeny฀ reconstruction,฀taxonomy. Introduction The฀genus Cremastosperma฀R.E.Fr.฀can฀most฀easily฀be฀distinguished฀from฀ other฀Neotropical฀Annonaceae฀with฀apocarpous,฀stipitate฀fruits฀by฀its฀midrib,฀ which฀ is฀ raised฀ on฀ the฀ upper฀ side฀ with฀ a฀ mostly฀ conspicuous฀ longitudinal฀ groove.฀The฀individual฀distributions฀of฀its฀species฀are฀restricted฀to฀four฀(disjunct)฀ areas฀of฀lowland฀to฀premontane฀tropical฀forest:฀1.฀the฀Chocó/Darién/western฀ Ecuador฀region฀(the฀narrow฀tropical฀zone฀to฀the฀west฀of฀the฀Andean฀mountain฀ chain฀on฀the฀Pacific฀Ocean฀side฀of฀north-western฀South฀America)฀plus฀central฀ America,฀ 2.฀ the฀ tropical฀Andes฀ (including฀ forests฀ on฀ the฀ eastern฀ side฀ of฀ the฀ Andes฀ extending฀ from฀ Colombia฀ through฀ eastern฀ Ecuador฀ and฀ Peru฀ as฀ far฀ south฀as฀Bolivia),฀3.฀coastal฀Venezuela฀and฀4.฀French฀Guiana. Ta x o n o m i c h i s t o r y o f C r e m a s t o s p e r m a Cremastosperma฀was฀described฀by฀Robert฀E.฀Fries฀(1930)฀based฀on฀a฀species฀ originally฀described฀by฀Diels฀(1906)฀under฀the฀genus฀Aberemoa:฀A.฀pedunculata฀ Diels,฀ which฀ thus฀ became฀ the฀ type฀ species฀ Cremastosperma฀ pedunculatum฀ (Diels)฀ R.E.Fr.฀ Further฀ work฀ by฀ Fries฀ up฀ to฀ 1950฀ increased฀ the฀ number฀ of฀ species฀of฀Cremastosperma฀to฀17.฀In฀1931฀he฀described฀C.฀cauliflorum฀R.E.Fr.,฀ C.฀ gracilipes฀R.E.Fr.,฀C.฀ leiophyllum฀R.E.Fr.,฀and฀C.฀ megalophyllum฀R.E.Fr.,฀ and฀ made฀ four฀ new฀ combinations฀ transferring฀ species฀ from฀ Cymbopetalum฀ (C.฀ monospermum฀ (Rusby)฀ R.E.Fr.),฀ Guatteria฀ (C.฀ pendulum฀ (Ruiz฀ &฀ Pav.)฀ R.E.Fr.฀ and฀ C.฀ poiteaui฀ (Diels)฀ R.E.Fr.),฀ and฀ Unonopsis฀ (C.฀ polyphlebum฀ (Diels)฀R.E.Fr.).฀In฀1934฀he฀described฀four฀new฀species:฀C.฀longicuspe฀R.E.Fr.,฀ C.฀ peruvianum฀ R.E.Fr.,฀ C.฀ guianense฀ R.E.Fr.,฀ and฀ C.฀ williamsii฀ R.E.Fr.;฀ in฀ 1937฀ one:฀ C.฀ juruense฀ R.E.Fr.฀ (in฀ which฀ publication฀ he฀ also฀ transferred฀ C.฀ guianense,฀C.฀polyphlebum฀and฀C.฀williamsii฀to฀the฀genus฀Pseudoxandra);฀and฀ in฀ 1939,฀ one:฀ C.฀ microcarpum฀ R.E.Fr.฀ plus฀ a฀ variety฀ of฀ C.฀ monospermum:฀ 88 • Chapter 5 C.฀monospermum฀(Rusby)฀R.E.Fr.฀var.฀brachypodum฀R.E.Fr.฀In฀the฀latter฀paper฀ he฀ also฀ made฀ a฀ new฀ combination:฀ C.฀ brevipes฀ (DC.)฀ R.E.Fr.,฀ under฀ which฀ he฀brought฀both฀Guatteria฀brevipes฀DC.฀and฀Cremastosperma฀poiteaui฀(Diels)฀ R.E.Fr.฀ into฀ synonymy.฀ In฀ 1948฀ and฀ 1950฀ Fries฀ described฀ his฀ last฀ five฀ new฀ species฀of฀Cremastosperma:฀ C.฀ anomalum฀R.E.Fr.,฀C.฀ killipii฀R.E.Fr.,฀and฀C.฀ oblongum฀R.E.Fr.,฀followed฀by฀the฀first฀species฀of฀the฀genus฀from฀the฀Pacific฀ coast฀of฀Colombia:฀C.฀novogranatense฀R.E.Fr.฀and฀C.฀pacificum฀R.E.Fr. After฀ the฀ death฀ of฀ Fries฀ in฀ 1966,฀ taxonomic฀ work฀ on฀ Cremastosperma฀ was฀resumed฀subsequent฀to฀the฀establishment฀of฀the฀international฀Annonaceae฀ project฀ (Maas,฀ 1983).฀ C.฀ anomalum฀ was฀ transferred฀ by฀ Maas฀ (in฀ Maas฀ et฀ al.,฀ 1986)฀to฀Malmea฀and฀subsequently฀by฀Chatrou฀(1998)฀to฀the฀newly฀described฀ Klarobelia,฀ along฀ with฀ a฀ number฀ of฀ species฀ formerly฀ placed฀ under฀ Malmea.฀ Maas฀(1986)฀further฀described฀one฀species฀from฀Panama฀(C.฀panamense฀Maas)฀ and฀one฀from฀Venezuela฀(C.฀macrocarpum฀Maas).฀ The฀increase฀in฀the฀numbers฀of฀collections฀available฀for฀study฀in฀the฀last฀ fifty฀ years฀ has฀ revealed฀ a฀ substantial฀ proportion฀ of฀ previously฀ undescribed฀ species฀in฀a฀number฀of฀recently฀revised฀Neotropical฀genera฀of฀Annonaceae.฀The฀ most฀extreme฀example฀may฀be฀that฀of฀฀Pseudoxandra,฀revised฀by฀Maas฀&฀Westra฀ (2003),฀in฀which฀80%฀of฀species฀treated฀were฀new฀or฀very฀recently฀described.฀ However,฀Klarobelia฀now฀comprises฀12฀species,฀of฀which฀only฀3฀were฀described฀ prior฀to฀the฀treatment฀of฀Chatrou฀(1998)฀(i.e.฀75%฀new).฀In฀preparation฀for฀a฀ revision฀of฀Cremastosperma฀for฀this฀thesis,฀a฀further฀13฀species,฀45%฀of฀the฀29฀ recognised฀here,฀have฀been฀described.฀ Pirie฀&฀Zapata฀(2004)฀described฀three฀species฀from฀the฀Marañon฀basin฀in฀ northern฀Peru,฀and฀Pirie฀(in฀Chatrou฀&฀Pirie,฀2005)฀increased฀the฀number฀of฀ species฀of฀Cremastosperma฀known฀from฀Venezuela฀to฀two.฀However,฀the฀greatest฀ underestimation฀of฀species฀diversity฀appeared฀to฀be฀represented฀by฀collections฀ from฀ the฀ Chocó/Darién/western฀ Ecuador฀ region.฀The฀ general฀ increase฀ in฀ numbers฀of฀collections฀of฀Cremastosperma฀made฀since฀the฀time฀of฀Fries฀has฀been฀ restricted฀in฀areas฀of฀north-western฀South฀America.฀Far฀fewer฀were฀available฀ than฀ for฀ comparable฀ areas฀ on฀ the฀Amazonian฀ side฀ of฀ the฀Andean฀ mountain฀ chain.฀Nevertheless,฀the฀degree฀of฀morphological฀differentiation฀clear฀from฀the฀ previously฀largely฀undetermined฀collections฀that฀were฀available฀led฀Pirie฀(2005)฀ to฀describe฀eight฀new฀species฀from฀extra-Amazonian฀Colombia฀and฀Ecuador฀ plus฀ Panama.฀ One฀ additional฀ new฀ species,฀ endemic฀ to฀Amazonian฀ Ecuador,฀ published฀in฀the฀latter฀paper฀brought฀the฀total฀number฀of฀recognised฀species฀of฀ Cremastosperma฀to฀31฀(including฀two฀species,฀C.฀killipii฀and฀C.฀juruense,฀which฀ are฀ synonymised฀ in฀ this฀ chapter฀ under฀ C.฀ longicuspe฀ and฀ C.฀ monospermum฀ respectively). Revision and phylogeny of Cremastosperma • 89 F i g . 1 . Flowers of Cremastosperma. Fig. 11 from van Heusden (1992). a. C. microcarpum: longitudinal section; b. C. oblongum R.E.Fr.: longitudinal section; c. C. monospermum (Rusby) R.E.Fr.: bud; d. C. gracilipes R.E.Fr.: flower; e. & f. C. cauliflorum R.E.Fr.: longitudinal section (e) and stamen (f); g. C. microcarpum: carpel (a: Maas et al. 6281; b: Maas et al. 4592; c: Nelson 763; d: Luteyn et al. 4890; e & f: Holm-Nielsen et al. 21501; g: Prance et al. 3527) 90 • Chapter 5 Position of Cremastosperma within Annonaceae Various฀morphological฀characters฀have฀been฀emphasised฀by฀different฀authors฀ in฀placing฀Cremastosperma฀in฀more,฀or฀less,฀formal฀classifications฀of฀Annonaceae.฀ Van฀ Heusden฀ (1992)฀ defined฀ her฀ Cremastosperma฀ group฀ (also฀ including฀ the฀ Neotropical฀ genera฀ Ephedranthus,฀ Malmea,฀ Oxandra,฀ Pseudephedranthus,฀ Pseudoxandra,฀ and฀ Ruizodendron)฀ on฀ a฀ combination฀ of฀ floral฀ characters:฀ imbricate,฀often฀ciliate฀sepals฀and฀petals;฀small฀sepals฀(rarely฀over฀4฀mm฀long);฀ usually฀whitish฀or฀greenish,฀sometimes฀yellowish฀flowers;฀and฀one฀basal,฀lateral฀ or฀apical฀ovule฀(Fig.฀1฀g).฀The฀broader฀ Malmea฀group฀of฀Walker฀(1971)฀also฀ included฀the฀Neotropical฀genera฀Unonopsis,฀Bocageopsis,฀and฀Onychopetalum฀ and฀the฀African฀genus฀Annickia.฀It฀was฀characterised฀by฀solitary,฀medium฀to฀ large,฀sulcate฀pollen฀grains.฀Group฀4฀of฀van฀Setten฀&฀Koek-Noorman฀(1992),฀ defined฀by฀transversely฀grooved฀or฀pitted฀seeds฀lacking฀arils,฀included฀most฀of฀ the฀ Cremastosperma฀ group฀ genera฀ plus฀ a฀ number฀ of฀Asian฀ taxa฀ (but฀ neither฀ Unonopsis,฀Bocageopsis,฀and฀Onychopetalum,฀nor฀Annickia).฀Amongst฀these฀taxa฀ Cremastosperma฀was฀noted฀to฀be฀exceptional฀in฀the฀combination฀of฀a฀pitted฀ seed฀ wall฀ (Fig.฀ 2฀ e,f)฀ with฀ spiniform฀ ruminations฀ -฀ a฀ condition฀ shared฀ with฀ Pseudoxandra฀ and฀ Malmea฀ sensu฀ Chatrou฀ (1998),฀ i.e.฀ not฀ including฀ species฀ formerly฀ included฀ in฀ Malmea฀ and฀ now฀ representing฀ the฀ genera฀ Klarobelia,฀ Mosannona,฀and฀Pseudomalmea.฀ Molecular฀phylogenetic฀studies฀of฀Annonaceae฀using฀DNA฀sequence฀data฀ (Mols฀et฀al.,฀2004;฀Richardson฀et฀al.,฀2004;฀this฀thesis฀Chapters฀2,฀3฀&฀4)฀have฀ shown฀strong฀support฀for฀a฀clade฀including฀all฀the฀above฀mentioned฀taxa฀plus฀ a฀large฀number฀of฀Asian฀species:฀the฀‘short฀branch฀clade’฀(the฀SBC).฀Results฀ presented฀in฀Chapter฀3฀of฀this฀thesis฀identified฀a฀clade฀including฀all฀SBC฀genera฀ with฀distributions฀centred฀in฀South฀America฀(the฀SAC฀clade),฀corresponding฀ to฀the฀Cremastosperma฀group฀of฀by฀van฀Heusden฀(1992),฀plus฀a฀clade฀including฀ Unonopsis,฀Bocageopsis,฀and฀Onychopetalum.฀Relationships฀within฀the฀SAC฀clade฀ were฀not฀all฀subject฀to฀high฀support.฀In฀particular,฀the฀relationship฀between฀the฀ latter฀Unonopsis฀clade฀and฀clades฀representing฀the฀Cremastosperma฀group฀were฀ effectively฀unresolved.฀However,฀sister฀group฀relationships฀were฀demonstrated฀ between฀Cremastosperma฀and฀Pseudoxandra,฀and฀between฀this฀clade฀and฀Malmea.฀ The฀combination฀of฀seed฀characters฀identified฀by฀van฀Setten฀&฀Koek-Noorman฀ (1992;฀see฀above)฀for฀these฀genera฀may฀thus฀represent฀synapomorphies฀for฀this฀ clade. Phylogeny and biogeographic history of Cremastosperma That฀ no฀ one฀ species฀ of฀ Cremastosperma฀ is฀ found฀ on฀ both฀ sides฀ of฀ the฀ Andean฀mountain฀chain฀suggests฀that฀the฀Andes฀represents฀a฀current฀barrier฀ to฀dispersal.฀The฀distribution฀of฀those฀species,฀in฀areas฀surrounding฀the฀Andes,฀ may฀ further฀ suggest฀ that฀ the฀Andean฀ orogeny฀ played฀ an฀ important฀ role฀ in฀ Revision and phylogeny of Cremastosperma • 91 F i g . 2 . Fruits and seeds of Cremastosperma. Adapted from plate 8 from van Setten & KoekNoorman (1992). a. Cremastosperma megalophyllum R.E.Fr.; b. & e. C. cauliflorum R.E.Fr.; c. C. microcarpum R.E.Fr.; d. C. macrocarpum Maas; f. C. monospermum (Rusby) R.E.Fr. (a: Brandbyge & Asanza C. 30017; b, e: Prance et al. 24094; c: Gentry et al. 32153; d: Wingfield & van der Werff 6751; f: Sperling et al. 6198) the฀diversification฀in฀the฀genus.฀Phylogenetic฀analysis฀of฀DNA฀sequence฀data฀ suggest฀ that฀ the฀ species฀ of฀ Cremastosperma฀ comprise฀ a฀ monophyletic฀ group฀ (this฀thesis,฀Chapters฀2฀&฀3),฀and฀results฀of฀molecular฀dating฀(Chapter฀3)฀suggest฀ diversification฀ in฀ this฀ group฀ occurred฀ during฀ the฀ timeframe฀ of฀ the฀Andean฀ orogeny.฀A฀cladistic฀biogeographic฀approach฀could฀be฀employed฀to฀determine฀ whether฀Andean-centred฀distributions฀found฀not฀only฀in฀Cremastosperma,฀but฀ 92 • Chapter 5 also฀in฀other฀clades฀in฀Annonaceae฀(such฀as฀Klarobelia,฀Malmea,฀and฀Mosannona,฀ see฀Chapter฀3)฀may฀be฀the฀result฀of฀a฀common฀biogeographic฀history.฀However,฀ relationships฀ particularly฀ between฀ species฀ of฀ Cremastosperma฀ remain฀ as฀ yet฀ largely฀unresolved.฀ In฀ this฀ chapter฀ a฀ taxonomic฀ revision฀ of฀ Cremastosperma฀ is฀ presented.฀ Twenty-nine฀species฀are฀recognised฀and฀two฀further฀putative฀species฀described฀ unofficially,฀ awaiting฀ further฀ collections.฀ Identification฀ keys฀ are฀ provided.฀ Results฀of฀phylogenetic฀analyses฀using฀DNA฀sequence฀data฀are฀presented.฀ Phylogeny reconstruction using DNA sequence data Materials and methods Taxon฀sampling:฀This฀study฀largely฀utilised฀previously฀unpublished฀sequence฀ data,฀as฀well฀as฀published฀sequences฀(Mols฀et฀al.,฀2004;฀Richardson฀et฀al.,฀2004;฀ this฀thesis,฀Chapters฀2,฀3฀&฀4;฀see฀appendix).฀Nine฀SBC฀outgroup฀taxa฀were฀ selected฀following฀the฀results฀of฀Chapter฀3,฀including฀two฀accessions฀each฀of฀ the฀most฀closely฀related฀genera฀Pseudoxandra฀and฀Malmea.฀The฀33฀accessions฀ of฀Cremastosperma฀sampled฀included฀19฀of฀the฀29฀species฀recognised฀here,฀from฀ across฀the฀entire฀geographical฀distribution,฀plus฀C.฀spec.฀A฀and฀C.฀spec.฀B฀(some฀ species฀were฀thus฀represented฀by฀multiple฀accessions). Character฀sampling:฀For฀all฀42฀accessions฀the฀cpDNA฀markers฀rbcL,฀matK,฀ trnT-F฀ and฀ psbA-trnH฀ were฀ sampled,฀ following฀ the฀ PCR฀ and฀ sequencing฀ protocols฀ described฀ in฀ Chapters฀ 2฀ &฀ 3.฀Amplification฀ and฀ sequencing฀ of฀ a฀ further฀cpDNA฀marker,฀ndhF,฀and฀of฀pseudtrnL-F฀(see฀Chapter฀4)฀was฀only฀ successful฀in฀27฀accessions฀of฀Cremastosperma฀(representing฀18฀species),฀and฀in฀ pseudtrnL-F฀only฀two฀outgroups฀(both฀species฀of฀Malmea).฀ Phylogenetic฀analysis:฀DNA฀sequences฀were฀edited฀in฀SeqMan฀4.0฀(DNAStar฀ Inc.,฀Madison,฀WI)฀and฀aligned฀manually.฀Areas฀of฀the฀alignments฀where฀the฀ assessment฀of฀homology฀was฀ambiguous฀were฀excluded฀from฀the฀analyses.฀Gaps฀ in฀the฀alignments฀were฀coded฀as฀present/absent฀characters฀where฀they฀could฀ be฀coded฀unambiguously,฀following฀Simmons฀and฀Ochoterena฀(2000)฀simple฀ gap฀ coding฀ principles.฀ Phylogeny฀ was฀ inferred฀ under฀ maximum฀ parsimony฀ (MP)฀and฀using฀Bayesian฀inference.฀MP฀‘full’฀heuristic฀searches฀(1000฀iterations,฀ TBR,฀ saving฀ 50฀ trees฀ per฀ iteration)฀ were฀ performed฀ for฀ the฀ combined฀ data฀ and฀MP฀bootstrap฀support฀was฀estimated฀using฀PAUP*฀(Swofford,฀2000),฀for฀ the฀markers฀individually฀and฀combined,฀as฀described฀in฀Chapter฀3.฀Combined฀ analyses฀under฀Bayesian฀inference฀were฀performed฀using฀MrBayes฀version฀3.0฀ (Huelsenbeck,฀2000),฀also฀as฀described฀in฀Chapter฀3. Revision and phylogeny of Cremastosperma • 93 �������a�p��osa �����opeta��m�per�����o �apra�t��s���r�����or�s ��aro�e��a������ata 1�00 100 �p�e�ra�t��s�sp�� 0105 �a�mea���e�s�a�a 1�00 100 �a�mea�sp�� 0197 �se��o�a��ra������a 0�84 60 �se��o�a��ra�sp�r�t�ssa��t� C���re��pes 1�00 63 1�00 93 �rench �uiana Vene�uela 0�93 �50 C��ma�ro�arp�m C���e�e��e�a��m 1�00 100 C��ma��a�e�ae Colombia - Antio�uia C��spe���A� Costa Rica C��pa�ame�se C���estrae 0�93 �50 Panama C���o�o�ra�ate�se Colombia - Choco 1�00 62 C��m��ro�arp�m C��me�a�op�����m� 0493 C���a�����or�m� 0598 C���apoe�se 1�00 79 0�82 �50 C���e�epe�se 1�00 93 0�98 65 C������at�m�1100 C������at�m� 1102 C���a�����or�m� 0312 C��me�a�op�����m� 0490 C��pe������at�m� 754 0�97 62 1�00 71 0�99 62 C���ama�a�ate�se� 0742 C���ama�a�ate�se� 1099 C���ama�a�ate�se� 1101 C��mo�osperm�m �ropical Andes C��pe������at�m� 0760 1�00 89 C��sp�� 0745 C��o��o���m� 1097 1�00 61 C���e�op�����m C��o��o���m�0600 0�69 0�98 �50 59 C��o��o���m� 0739 C��spe��� B 0756� C��spe���B 0758 C��pe�����m� 0749 0�84 �50 C��pe�����m� 0751 C��pe�����m� 1098 F i g . 3 Cremastosperma phylogeny based on rbcL, matK, trnT-F, and psbA-trnH (i.e. without ndhF or pseudtnL-F data): posterior probabilities (above the nodes) and bootstrap percentages (below). 94 • Chapter 5 Results and Discussion MP฀bootstrap฀analysis฀of฀the฀individual฀data฀partitions฀revealed฀no฀supported฀ incongruencies฀(BS฀>75%).฀Data฀were฀thus฀combined฀in฀further฀analyses.฀MP฀ heuristic฀search฀including฀all฀taxa฀but฀neither฀ndhF฀nor฀pseudtrnL-F฀resulted฀ in฀44501฀trees฀of฀539฀steps,฀CI฀=฀0.87,฀RI฀=฀0.83.฀The฀same฀search฀strategy฀ with฀inclusion฀of฀ndhF฀and฀pseudtrnL-F,฀and฀exclusion฀of฀ingroup฀taxa฀for฀ which฀both฀were฀unavailable,฀resulted฀in฀148฀trees฀of฀832฀steps,฀CI฀=฀0.88,฀RI฀ =฀0.83.฀Consensus฀topologies฀(50%฀majority฀rule)฀derived฀under฀MP฀bootstrap฀ and฀Bayesian฀analyses฀were฀congruent.฀Topologies฀with฀posterior฀probabilities฀ (above฀the฀nodes)฀and฀bootstrap฀percentages฀(below)฀are฀presented฀in฀Fig.฀3฀(for฀ analyses฀including฀all฀taxa฀but฀without฀ndhF฀or฀pseudtrnL-F,฀data)฀and฀Fig.฀4฀ (including฀ndhF or฀pseudtrnL-F฀data฀for฀fewer฀taxa).฀ Table 1: DNA sequences: variable and parsimony informative characters. Values are made comparable by the inclusion of only Cremastosperma and Malmea taxa for which ndhF and pseudtrnL-F sequences were available. Marker Approx. Variable Pars. inf. sequence characters: characters length a rbcL 1480 Pars. inf. Indels 22 13 0 matK 850 42 21 0 ndhF 2050 100 33 0 1550-1950 68 34 7 400-450 25 10 1 600 50 29 1 trnT-trnF psbA-trnH pseudtrnL-F a Including 3’ non coding region Use฀of฀multiple฀chloroplast฀markers฀here฀(with฀the฀possible฀exception฀of฀ pseudtrnL-F,฀the฀genomic฀compartment฀of฀which฀is฀unknown:฀see฀Chapter฀4)฀ revealed฀a฀number฀of฀clades฀within฀the฀genus฀ Cremastosperma.฀PseudtrnL-F฀ provided฀ the฀ highest฀ proportion฀ of฀ informative฀ characters฀ for฀ the฀ length฀ of฀ sequence.฀ Given฀ the฀ low฀ numbers฀ of฀ informative฀ characters฀ which฀ cpDNA฀ markers฀appear฀to฀provide฀at฀this฀taxonomic฀level฀in฀Annonaceae฀(Table฀1),฀ this฀is฀probably฀not฀a฀realistic฀strategy฀for฀deriving฀a฀fully฀resolved฀phylogeny฀ of฀the฀genus.฀The฀development฀of฀protocols฀for฀more฀variable฀markers฀(such฀as฀ single฀copy฀nuclear฀genes)฀is฀thus฀of฀importance฀to฀further฀addressing฀problems฀ at฀this฀level. The฀clades฀revealed฀correspond฀largely฀to฀geographic฀areas฀(see฀Figs฀3&4).฀ The฀divergence฀of฀the฀Venezuelan฀and฀Guianan฀lineages฀occurred฀prior฀to฀that฀ leading฀to฀clades฀found฀either฀in฀the฀tropical฀Andes,฀or฀in฀the฀Chocó/Darién/ Revision and phylogeny of Cremastosperma • 95 �������a�p��osa a �apra�t��s���r�����or�s �����opeta��m�per�����o 0�74 55 1�00 100 �p�e�ra�t��s�sp�� 0105 �a�mea���e�s�a�a 1�00 100 �a�mea�sp�� 0197 �se��o�a��ra������a 0�99 82 �se��o�a��ra�sp�r�t�ssa��t� Vene�uela 0�75 �50 ��aro�e��a������ata C���e�e��e�a��m 1�00 100 0�87 84 C��ma�ro�arp�m C���re��pes �rench �uiana Panam a 1�00 84 1�00 96 C��pa�ame�se 1�00 77 C���estrae 0�99 89 C��ma��a�e�ae Colombia C��spe��� A Costa Rica C��m��ro�arp�m C��me�a�op�����m� 0493 0�98 56 C���a�����or�m� 0598 C���apoe�se 1�00 99 0�68 �50 0�98 69 C������at�m� 1100 C������at�m� 1102 C���a�����or�m� 0312 C��me�a�op�����m� 0490 C��pe������at�m� 0760 1�00 97 1�00 91 1�00 88 C���ama�a�ate�se� 1099 C���ama�a�ate�se� 1101 C��mo�osperm�m C��pe�����m� 0751 0�92 52 C��sp� 0745 C��o��o���m� 1097 C��pe�����m�1098 0�75 �50 0�94 �50 96 • Chapter 5 0�97 �50 0�91 �50 C��o��o���m� 0600 C��o��o���m� 0739 C���e�op�����m C��spe��� B 0756 �ropical Andes 1�00 91 �������a�p��osa b �apra�t��s���r�����or�s �����opeta��m�per�����o ��aro�e��a������ata �p�e�ra�t��s�sp��0105 �a�mea���e�s�a�a �a�mea�sp���0197 �se��o�a��ra������a �se��o�a��ra�sp�r�t�ssa��t� C���e�e��e�a��m C���re��pes C��ma�ro�arp�m C��pa�ame�se C���estrae C��ma��a�e�ae C��spe���A C��m��ro�arp�m C��me�a�op�����m�0493 C���a�����or�m� 0598 C� napoense C������at�m� 1100 C������at�m� 1102 C���a�����or�m� 0312 C��me�a�op�����m�0490 C��pe������at�m�0760 C���ama�a�ate�se�1099 C���ama�a�ate�se� 1101 C��mo�osperm�m C��pe�����m� 0751 C��sp�� 0745 C��o��o���m�1097 C�pe�����m 1098 C��o��o���m�0600 C��o��o���m�0739 C���e�op�����m C��spe���B 0756 Fig. 4: 0�01 Cremastosperma phylogeny based on rbcL, matK, trnT-F, psbA-trnH, ndhF, and pseudtrnL-F (taxa for which ndhF or pseudtrnL-F was unavailable are excluded): a. Bayesian 50% consensus topology. Posterior probabilities (above the nodes) and bootstrap percentages (below). b. Phylogram resulting from Bayesian analysis (scale bar indicates 0.01 subst. per site) Revision and phylogeny of Cremastosperma • 97 western฀Ecuador฀region฀or฀Central฀America฀(i.e.฀either฀the฀west฀or฀the฀east฀side฀ of฀the฀Andes฀mountain฀chain).฀The฀tropical฀Andes฀species฀all฀fall฀into฀one฀of฀ two฀clades.฀Relationships฀between฀these฀two฀clades,฀a฀further฀clade฀including฀ the฀Central฀American฀species฀Cremastosperma฀panamense฀and฀C.฀westrae,฀and฀ two฀isolated฀lineages฀corresponding฀to฀accessions฀from฀Costa฀Rica฀(C.฀spec.฀A)฀ and฀the฀Magdalena฀valley฀of฀Colombia฀(C.฀magdalenae)฀are฀unresolved.฀ These฀ results฀ provide฀ further฀ evidence฀ to฀ suggest฀ the฀ importance฀ of฀ the฀Andean฀orogeny฀as฀a฀vicariance฀event฀in฀the฀history฀of฀the฀evolution฀of฀ Cremastosperma.฀Further฀conclusions฀await฀a฀more฀resolved฀phylogeny฀of฀the฀ genus. Systematic treatment Materials and methods Measurements฀are฀based฀mostly฀on฀dried฀material.฀Where฀measurements฀ are฀ derived฀ from฀ material฀ preserved฀ in฀ alcohol฀ (and฀ dimensions฀ therefore฀ often฀greater,฀as฀the฀structures฀are฀not฀subject฀to฀water฀loss)฀these฀are฀indicated฀ between฀accolades฀{}. Indument:฀ unless฀ otherwise฀ indicated,฀ descriptions฀ of฀ the฀ indument฀ of฀ bracts,฀sepals,฀and฀petals฀refers฀to฀that฀on฀the฀outer฀side. Leaf฀apex:฀in฀a฀few฀cases฀we฀make฀a฀distinction฀between฀acuminate฀and฀ cuspidate฀(in฀the฀sense฀of฀abruptly฀acuminate)฀leaf฀apex฀shapes:฀this฀represents฀an฀ arbitrary,฀but,฀the฀first฀author฀believes,฀nevertheless฀useful,฀delimitation฀within฀ a฀pattern฀of฀continuous฀variation.฀An฀illustration฀accompanies฀this฀character฀in฀ the฀key. Inflorescence:฀ according฀ to฀ Chatrou฀ (1998),฀ the฀ inflorescences฀ of฀ all฀ Annonaceae฀can฀be฀considered฀as฀terminal:฀apparently฀axillary฀inflorescences฀ in฀genera฀such฀as฀Guatteria,฀Klarobelia,฀and฀Pseudomalmea฀consists฀of฀a฀short฀ shoot,฀developing฀primarily฀from฀a฀leaf-axillary฀position,฀subtending฀a฀terminal฀ pedicel฀with฀flower.฀The฀distinction฀between฀these฀two฀structures฀is฀in฀most฀cases฀ clear,฀demarcated฀by฀an฀articulation.฀The฀short฀shoot฀bears฀a฀variable฀number฀of฀ bracts฀(referred฀to฀here฀as฀lower฀bracts)฀which฀in฀occasional฀specimens฀are฀larger฀ and฀leaf-like฀in฀appearance.฀In฀two฀species฀of฀Cremastosperma฀(C.฀cauliflorum฀ R.E.Fr.฀ and฀ C.฀ napoense฀ Pirie),฀ a฀ rhipidium฀ is฀ formed฀ by฀ the฀ development฀ of฀ further฀ shoots฀ from฀ the฀ axils฀ of฀ bracts฀ on฀ this฀ short฀ axillary฀ shoot.฀The฀ short฀axillary฀shoot฀is฀described฀here฀under฀the฀convenient฀term฀‘peduncle’,฀ following฀Maas฀&฀Westra฀(2003).฀ 98 • Chapter 5 C r e m a s t o s p e r m a -฀Map฀1 Cremastosperma฀R.E.Fr.฀(1930)฀46,฀f.฀6a-c.฀-฀Type:฀Cremastosperma฀pedunculatum฀฀ (Diels)฀R.E.Fr. Trees฀or฀shrubs฀(0.5-)1.5-20฀m฀tall;฀young฀twigs฀and฀petioles฀glabrous฀to฀ densely฀covered฀with฀appressed฀or฀erect,฀simple,฀whitish฀to฀golden,฀up฀to฀1฀mm฀ long฀hairs.฀Leaves฀distichous,฀simple,฀entire,฀petiolate,฀exstipulate;฀lamina฀elliptic฀ to฀obovate,฀or฀narrowly฀so,฀index฀1.6-5,฀chartaceous฀to฀coriaceous,฀glabrous฀ (rarely฀sparsely฀covered฀with฀appressed฀or฀erect,฀simple,฀up฀to฀1฀mm฀long฀hairs)฀ above,฀glabrous฀to฀densely฀hairy฀(particularly฀at฀the฀base฀and฀on฀veins)฀below,฀ base฀ acute,฀ obtuse,฀ or฀ rounded฀ (subcordate฀ to฀ cordate),฀ apex฀ (obtuse,฀ acute)฀ acuminate฀ (or฀ cuspidate),฀ extreme฀ tip฀ rounded,฀ venation฀ brochidodromous,฀ primary฀vein฀raised฀over฀entire฀leaf฀length฀above฀with฀an฀often฀conspicuous฀ longitudinal฀groove฀particularly฀in฀the฀basal฀half,฀secondary฀veins฀5-20(-30)฀on฀ either฀side฀of฀the฀primary฀vein,฀often฀with฀1-6฀intersecondary฀veins,฀running฀ parallel฀ to฀ primary฀ vein฀ for฀ a฀ short฀ distance,฀ thereafter฀ angles฀ with฀ primary฀ vein฀either฀increasing฀or฀decreasing฀towards฀the฀apex฀(or฀consistent),฀sometimes฀ branching,฀often฀forming฀distinct฀loops,฀smallest฀distance฀between฀loops฀and฀ margin฀1-7฀mm,฀tertiary฀veins฀percurrent฀(or฀reticulate).฀Inflorescence฀of฀single฀ flowers,฀or฀occasionally฀up฀to฀8฀in฀a฀rhipidium,฀pendant,฀clustered฀in฀groups฀of฀ up฀to฀7,฀terminal฀on฀short฀axillary฀shoots฀(i.e.฀peduncles)฀on฀leafy฀or฀leafless฀ twigs,฀ older฀ branches฀ or฀ on฀ the฀ main฀ trunk฀ (then฀ often฀ on฀ brachyblasts).฀ Indument:฀peduncles,฀pedicels,฀outer฀sides฀of฀bracts,฀sepals,฀and฀petals฀glabrous฀ to฀ densely฀ covered฀ with฀ appressed฀ or฀ erect,฀ simple,฀ up฀ to฀ 1฀ mm฀ long฀ hairs,฀ bracts,฀sepals,฀and฀petals฀ciliate.฀1-several฀lower฀bracts,฀deltate฀to฀depressed฀ovate฀ (rarely฀narrowly฀elliptic,฀leafy),฀rounded฀to฀acute,฀caducous฀or฀persistent;฀single฀ upper฀bract฀attached฀to฀pedicel,฀ovate฀to฀deltate,฀acute฀to฀obtuse;฀closed฀flower฀ buds฀(ovoid฀to฀triangular)฀broadly฀to฀depressed฀ovoid.฀Flowers฀actinomorphic,฀ bisexual,฀with฀one฀whorl฀of฀free฀or฀slightly฀connate,฀imbricate,฀sepals฀and฀two฀ whorls฀of฀free,฀imbricate,฀petals,฀green,฀creamy฀or฀yellow฀in฀vivo,฀often฀black฀in฀ sicco;฀sepals฀and฀petals฀thin฀at฀margins,฀occasionally฀with฀prominent฀venation;฀ sepals฀three,฀much฀smaller฀than฀petals;฀petals฀six,฀the฀outer฀ones฀ovate,฀elliptic,฀or฀ broadly฀so,฀the฀inner฀ones฀elliptic฀to฀obovate฀or฀narrowly฀so,฀stamens฀numerous,฀ spirally฀arranged,฀extrorse,฀inserted฀on฀and฀below฀a฀ventral฀ridge฀encircling฀a฀ central฀depression฀in฀the฀receptacle฀in฀which฀the฀carpels฀are฀inserted,฀1-2฀mm฀ long,฀ connective฀ appendage฀ transversely฀ rhombic-hexagonal;฀ carpels฀ 20-40,฀ spirally฀arranged,฀free,฀ovary฀1-locular,฀glabrous฀or฀hairy,฀with฀1฀basal,฀lateral,฀or฀ apical฀ovule,฀stigma฀sessile.฀Fruit฀apocarpous,฀monocarps฀5-40,฀stipitate,฀mostly฀ asymmetrical,฀sometimes฀strongly฀so,฀sometimes฀with฀an฀often฀excentric฀apicule,฀ green฀maturing฀mostly฀through฀red฀to฀brown฀or฀black฀in฀vivo,฀light฀brown฀to฀ black฀in฀sicco.฀Seeds฀1,฀basal,฀lateral฀or฀apical,฀ellipsoid฀to฀globose,฀yellow฀to฀ Revision and phylogeny of Cremastosperma • 99 Map 1: Distribution of Cremastosperma reddish฀brown,฀surface฀deeply฀to฀shallowly฀pitted,฀lacking฀an฀aril,฀with฀a฀raised฀ or฀sunken฀raphe฀encircling฀seed฀longitudinally฀(diagonally),฀regularly฀(or฀more฀ sinuously),฀ruminations฀spiniform. Distribution฀ -฀Twenty-nine฀ species฀ in฀ the฀ Neotropics:฀ from฀ southern฀ Costa฀Rica฀in฀the฀north฀to฀Bolivia฀in฀the฀south.฀Most฀species฀are฀distributed฀ in฀ regions฀ surrounding฀ the฀Andean฀ mountain฀ range,฀ 2฀ in฀ coastal฀Venezuela฀ (Cremastosperma฀macrocarpum฀Maas฀and฀C.฀venezuelanum฀Pirie),฀1฀in฀French฀ Guiana฀(C.฀brevipes฀(DC.)฀R.E.Fr.),฀and฀1฀widespread฀across฀Brazil฀south฀of฀the฀ Amazon฀river฀(C.฀monospermum฀(Rusby)฀R.E.Fr.). Habitat฀ and฀ Ecology฀ -฀ Lowland฀ to฀ premontane฀ tropical฀ wet฀ forest,฀ inundated฀areas,฀and฀terra฀firme.฀At฀elevations฀of฀0-2000฀m. 100 • Chapter 5 Key to species of Cremastosperma A฀ number฀ of฀ species฀ of฀ Cremastosperma฀ cannot฀ always฀ be฀ identified฀ with฀certainty฀on฀the฀basis฀of฀fruiting฀material฀alone.฀Floral฀characters฀are฀of฀ primary฀importance฀from฀lead฀12฀onwards฀in฀the฀identification฀key.฀From฀this฀ point฀fruit฀characters฀are฀still฀included฀where฀they฀are฀also฀informative,฀and฀in฀ combination฀with฀known฀distributions฀of฀the฀species฀may฀still฀allow฀successful฀ determination฀in฀many฀cases฀where฀flowering฀material฀is฀unavailable.฀C.฀spec.฀ A฀(only฀known฀from฀fruits)฀is฀not฀included.฀It฀is,฀however,฀the฀only฀species฀of฀ the฀genus฀known฀from฀Costa฀Rica. 1a.฀ Pedicels฀ >120฀ mm฀ long฀ in฀ flower,฀ >150฀ mm฀ long฀ in฀ fruit.฀฀ Lamina฀35-60฀cm฀long฀-฀(Pacific฀coast฀of฀Colombia฀and฀Ecuador).............. .....................................................................................12.฀C.฀longipes฀ b.฀ Pedicels฀<120฀mm฀long฀in฀flower,฀≤150฀mm฀long฀in฀fruit.฀Lamina฀8-64฀ cm฀long.......................................................................................................2 2a.฀ Lamina฀bullate,฀densely฀covered฀with฀hairs฀1฀mm฀long฀below฀and฀on฀ margin.฀Pedicels฀c.฀100฀to฀120฀mm฀long฀in฀flower,฀110-150฀mm฀long฀in฀fruit฀-฀ (The฀Peruvian฀state฀of฀Amazonas).........................................4.฀C.฀bullatum b.฀ Leaf฀lamina฀not,฀or฀rarely฀slightly,฀bullate,฀sparsely฀to฀densely฀covered฀with฀ hairs฀up฀to฀0.5฀mm฀long฀or฀glabrous฀below.฀Pedicels฀<95฀mm฀long฀in฀flower,฀ <110฀mm฀long฀in฀fruit................................................................................3฀ 3a.฀ Sepals฀7-10฀mm฀long,฀often฀persistent,฀densely฀covered฀with฀hairs฀c.฀0.6฀ mm฀long.฀Stipes฀1.5-4฀mm฀long................................................................4b.฀ Sepals฀≤7฀mm฀long,฀mostly฀caducous,฀sparsely฀to฀densely฀covered฀with฀hairs฀ <0.6฀mm฀long฀or฀glabrous.฀Stipes฀>4฀mm฀long...........................................5 4a.฀ Monocarps฀sparsely฀to฀rather฀densely฀covered฀with฀hairs฀to฀0.2฀mm฀long฀-฀ (Pacific฀Coast฀of฀Colombia).....................................19.฀C.฀novogranatense b.฀ Monocarps฀glabrous฀or฀sparsely฀covered฀with฀hairs฀to฀0.1฀mm฀long฀-฀ (Panama)...............................................................................28.฀C.฀westrae 5a.฀ Leaf฀apex฀obtuse฀to฀acute,฀base฀cordate฀(rarely฀rounded),฀secondary฀veins฀ 15-30฀on฀each฀side฀-฀(N฀Venezuela)...........................27.฀C.฀venezuelanum b.฀ Leaf฀apex฀acuminate฀to฀cuspidate,฀base฀acute฀to฀obtuse฀(rarely฀rounded฀or฀ cordate),฀secondary฀veins฀mostly฀<20฀on฀each฀side......................................6 6a.฀ Inflorescence฀branching,฀of฀multiple฀flowers/fruits................................7 b.฀ Inflorescence฀not฀branching,฀of฀single฀flowers/fruits..............................8 Revision and phylogeny of Cremastosperma • 101 7a.฀ Outer฀side฀of฀sepals฀and฀petals฀densely฀covered฀with฀hairs฀to฀0.4฀mm฀long.฀ Monocarps฀globose฀to฀transversely฀broadly฀ellipsoid,฀rather฀densely฀covered฀ with฀hairs฀to฀0.2฀mm฀long฀-฀(Amazonian฀Brazil,฀Ecuador,฀and฀Peru)................ ..........................................................................................5.฀C.฀cauliflorum b.฀ Outer฀side฀of฀sepals฀and฀outer฀petals฀rather฀densely฀to฀densely฀covered฀with฀ hairs฀to฀0.2฀mm฀long,฀inner฀petals฀largely฀glabrous฀but฀with฀a฀narrow,฀dense,฀ sometimes฀branching฀band฀of฀appressed฀whitish-golden฀hairs฀to฀0.2฀mm฀long฀ extending฀from฀the฀base฀to฀halfway฀towards฀the฀apex.฀Monocarps฀ellipsoid,฀ glabrous฀ -฀ (Amazonian฀ Ecuador).................................18.฀ C.฀ napoense฀ 8a.฀ Lamina฀up฀to฀20฀cm฀long.฀Inflorescence฀on฀main฀trunk.฀Pedicels฀glabrous.฀ Monocarps฀ellipsoid,฀strongly฀asymmetrical฀-฀(Pacific฀coast฀of฀Colombia)....... ............................................................................................7.฀C.฀chococola b.฀ Lamina฀up฀to฀64฀cm฀long.฀Inflorescence฀on฀leafy฀twigs,฀thicker฀branches,฀ or฀ on฀ the฀ main฀ trunk.฀ Pedicels฀ hairy฀ or฀ glabrous.฀ Monocarps฀ globose฀ or฀ ellipsoid,฀mostly฀slightly฀asymmetrical..........................................................9 9a.฀ Monocarps฀hairy฀(if฀without฀fruits,฀go฀to฀lead฀12)..............................10 b.฀ Monocarps฀ glabrous,฀ or฀ appearing฀ so฀ (sparsely฀ covered฀ with฀ hairs฀฀ <0.1฀mm฀long)..........................................................................................12 10a.฀ Leaf฀base฀cordate฀to฀subcordate.฀Pedicels฀c.฀8฀mm฀long.฀Monocarps฀slightly฀฀ longer฀than฀stipes฀-฀(The฀Peruvian฀state฀of฀Amazonas).........6.฀C.฀cenepense b.฀ Leaf฀base฀acute฀to฀obtuse,฀rarely฀rounded.฀Pedicels฀(5-)12-25฀mm฀long.฀ Monocarps฀roughly฀equal฀to฀or฀shorter฀than฀stipes.....................................11 11a.฀ Leaf฀apex฀cuspidate,฀lamina฀drying฀green.฀Flowers฀covered฀with฀hairs฀c.฀ 0.2฀mm฀long.฀Monocarps฀10-15฀mm฀long฀-฀(Amazonian฀Colombia,฀Ecuador,฀ and฀Peru).............................................................................9.฀C.฀gracilipes b.฀ Leaf฀ apex฀ acuminate,฀ lamina฀ drying฀ brown฀ or฀ greyish฀ green.฀ Flowers฀ covered฀with฀hairs฀c.฀0.3฀mm฀long.฀Monocarps฀8-11฀mm฀long฀-฀(Amazonian฀ Brazil,฀Colombia,฀and฀Peru).........................................16.฀C.฀microcarpum 12a.฀ Sepals,฀petals,฀and฀pedicels฀hairy..........................................................13 b.฀ Sepals฀and฀petals฀glabrous,฀pedicels฀mostly฀so.......................................18 13a.฀ Sepals฀≥3฀mm฀long.............................................................................14 b.฀ Sepals฀<3฀mm฀long.............................................................................15 102 • Chapter 5 14a.฀฀Leaf฀ axillary฀ buds฀ with฀ conspicuous฀ dense฀ indument.฀ Pedicels฀ >25฀ mm฀ long฀ in฀ flower.฀ Monocarps฀ narrowly฀ ellipsoid,฀ 27-28฀ mm฀ long฀ -฀฀ (The฀Colombian฀state฀of฀Antioquia)...........................8.฀C.฀dolichocarpum b.฀ Leaf฀ axillary฀ buds฀ inconspicuous.฀ Pedicels฀ ≤฀ 25฀ mm฀ long฀ in฀ flower.฀ Monocarps฀ellipsoid,฀up฀to฀15฀mm฀long....................................................11 15a.฀ Pedicels฀<17฀mm฀long฀in฀fruit฀or฀flower฀-฀(Panama)......28.฀C.฀westrae b.฀ Pedicels฀>17฀mm฀long฀in฀fruit฀or฀flower............................................16 16a.฀ Pedicels฀>30฀mm฀long฀in฀flower฀-฀(W฀Ecuador)...................................... ...................................................................................26.฀C.฀stenophyllum฀ b.฀ Pedicels฀<30฀mm฀long฀in฀flower........................................................17 17a.฀ Outer฀side฀of฀petals฀rather฀densely฀(evenly)฀covered฀with฀hairs.฀Monocarps฀ <15฀mm฀long,฀shorter฀than฀stipes,฀strongly฀asymmetrical฀-฀(The฀Colombian฀ state฀of฀Antioquia)...........................................................1.฀C.฀antioquense฀ b.฀ Outer฀ side฀ of฀ petals฀ densely฀ covered฀ with฀ hairs฀ at฀ base฀ and฀ in฀ a฀ line฀ leading฀ to฀ the฀ petal฀ apex.฀ Monocarps฀ 22-28฀ mm฀ long,฀ longer฀ than฀ stipes,฀ slightly฀asymmetrical฀-฀(Pacific฀coast฀of฀Colombia฀and฀Ecuador)...................... ..............................................................................................2.฀C.฀awaense 18a.฀ Flower฀ buds฀ depressed฀ ovoid฀ or฀ broadly฀ ovoid-triangular,฀ remaining฀ closed฀in฀development...............................................................................19 b.฀ Flower฀buds฀depressed฀ovoid,฀opening฀(loosely)฀in฀development.........22 19a.฀ Flower฀buds฀depressed฀ovoid..............................................................20 b.฀ Flower฀buds฀broadly฀ovoid-triangular.................................................21 ฀ 20a.฀ Pedicels฀>28฀mm฀long฀in฀flower฀and฀fruit฀-฀(Central฀Peru).................... ........................................................................................24.฀C.฀pendulum b.฀ Pedicels฀ <20฀ mm฀ long฀ in฀ flower฀ and฀ fruit฀ -฀ (The฀ Peruvian฀ state฀฀ of฀Amazonas)............................................................29.฀C.฀yamayakatense 21a.฀ Lamina฀≤฀35฀cm฀long.฀Pedicels฀40-50฀mm฀long฀in฀flower.฀Stipes฀8-15฀mm฀ long฀-฀(Amazonian฀Bolivia,฀Brazil,฀and฀Peru)............17.฀C.฀monospermum b.฀ Lamina฀>฀35฀cm฀long.฀Pedicels฀18-20฀mm฀long฀in฀flower.฀Stipes฀20-40฀mm฀ long฀-฀(The฀Peruvian฀state฀of฀Amazonas).......................25.฀C.฀peruvianum Revision and phylogeny of Cremastosperma • 103 22a.฀ Petals฀opening฀loosely฀in฀bud฀development,฀mostly฀covering฀reproductive฀ organs฀on฀drying........................................................................................23 b.฀ Petals฀opening฀widely฀in฀bud฀development,฀reproductive฀organs฀mostly฀ exposed฀on฀drying.....................................................................................27 23a.฀ Pedicels฀>30฀mm฀long.......................................................................24 b.฀ Pedicels฀<30฀mm฀long.......................................................................25 24a.฀ Sepals฀up฀to฀2฀mm฀long.฀Monocarps฀18-24฀mm฀long฀-฀(N฀Venezuela)..... ..................................................................................13.฀C.฀macrocarpum b.฀ Sepals฀ longer฀ than฀ 2฀ mm.฀ Monocarps฀ 12-17฀ mm฀ long฀ -฀ (N฀ Peru,฀฀ Ecuador)...................................................................23.฀C.฀pedunculatum 25a.฀ Sepals฀≥฀4฀mm฀long...........................................................................26 b.฀ Sepals฀<฀4฀mm฀long฀-฀(Pacific฀coast฀of฀Colombia)......21.฀C.฀pacificum 26a.฀ Leaf฀lamina฀20-28฀cm฀long.฀Pedicels฀<10฀mm฀long฀in฀flower.฀Monocarps฀ globose฀-฀(The฀Colombian฀state฀of฀Antioquia)............................................... ......................................................................................14.฀C.฀magdalenae b.฀ Leaf฀lamina฀13-57฀cm฀long.฀Pedicel฀>10฀mm฀long฀in฀flower.฀Monocarps฀ ellipsoid฀to฀broadly฀so฀-฀(Amazonian฀Colombia,฀Ecuador,฀and฀Peru).............. ................................................................................15.฀C.฀megalophyllum 27a.฀ Pedicels฀>20฀mm฀long฀in฀flower........................................................28 b.฀ Pedicels฀<20฀mm฀long฀in฀flower........................................................29 28a.฀ Pedicels฀ up฀ to฀ 34฀ mm฀ long฀ in฀ flower.฀ Monocarps฀ drying฀ blackish,฀ asymmetrical,฀the฀stipes฀thickening฀somewhat฀where฀they฀meet฀the฀monocarps฀ -฀(Bolivia)......................................................................10.฀C.฀leiophyllum b.฀ Pedicels฀up฀to฀68฀mm฀long฀in฀flower.฀Monocarps฀drying฀light฀to฀dark฀ brown฀or฀blackish,฀the฀stipes฀not฀markedly฀thickening฀where฀they฀meet฀the฀ monocarps฀-฀(S฀Peru)..............................................................31฀C.฀spec฀B 29a.฀ Lamina฀coriaceous,฀up฀to฀45฀cm฀long.฀Sepals฀≤2฀mm฀long฀-฀(The฀Brazilian฀ state฀of฀Acre฀and฀C฀and฀S฀Peru).......................................20.฀C.฀oblongum b.฀ Lamina฀chartaceous,฀up฀to฀40฀cm฀long.฀Sepals฀mostly฀>2฀mm฀long........ .................................................................................................................30 30a.฀ Leaf฀apex฀cuspidate฀-฀(N฀Peru).................................11.฀C.฀longicuspe b.฀ Leaf฀apex฀acuminate...........................................................................31 104 • Chapter 5 31a.฀ Lamina฀ 18-39฀ cm฀ long,฀ often฀ drying฀ reddish฀ on฀ veins฀ below.฀ Sepals฀฀ 3-4฀mm฀long.฀Monocarps฀ellipsoid฀-฀(French฀Guiana)............3.฀C.฀brevipes b.฀ Lamina฀ up฀ to฀ 22฀ cm฀ long,฀ drying฀ greenish.฀ Sepals฀ 1.5-3฀ mm฀ long.฀ Monocarps฀more฀or฀less฀globose฀-฀(Panama)....................22.฀C.฀panamense Synoptical Key Numbers฀ cited฀ refer฀ to฀ those฀ used฀ in฀ the฀ key฀ above฀ and฀ descriptions.฀ Those฀cited฀under฀more฀than฀one฀lead฀are฀indicated฀in฀bold.฀Where฀a฀character฀ state฀is฀unknown฀for฀a฀given฀species฀the฀corresponding฀number฀is฀omitted. ฀ ฀ ฀ 1.฀Leaf฀axillary฀buds฀indument ฀ a)฀Conspicuously฀hairy฀(versus฀inconspicuous฀and฀glabrous). ฀ ฀ 8-19 ฀ 2.฀Leaf฀length ฀ a)฀<25฀cm฀ ฀ ฀ 1 -2 -3 -4 -5 -6-7-8-9 -1 0 -1 1 -1 3 -1 4 -1 5 -1 6 -1 7 -1 8 -2 0 -2 1 -฀ ฀ 22-23-24-29-30-31 ฀ b)฀25-35฀cm ฀ ฀ 1 -2 -3 -4 -5 -9 -1 0 -1 1 -1 3 -1 4 -1 5 -1 6 -1 7 -1 8 -1 9 -2 0 -2 1 -2 3 -฀ ฀ 24-26-27-28-30-31 ฀ c)฀>35฀cm ฀ ฀ 3-5-12-15-18-19-20-21-25-27-28-30 3.฀Leaf฀base ฀ a)฀Cordate,฀subcordate฀or฀rounded฀(versus฀obtuse฀or฀acute) ฀ ฀ 4-6-19-27 ฀ 4.฀Leaf฀apex ฀ a)฀Acute฀or฀obtuse ฀ ฀ 27 ฀ b)฀Acuminate ฀ ฀ 1-2-3-4-5-6-7-8-10-11-12-13-14-15-16-17-18-19-20-21-฀ ฀ 22-23-24-25-26-28-29-30-31 ฀ c)฀Cuspidate ฀ ฀ 9-11-17-19 5.฀Leaf฀appearance ฀ a)฀Bullate฀(versus฀not฀bullate) ฀ ฀ 4-(15) Revision and phylogeny of Cremastosperma • 105 6.฀Secondary฀veins ฀ a)฀≥18฀(versus฀<18) ฀ ฀ 4-15-20-25-27 ฀ ฀ 7.฀Inflorescence฀position ฀ a)฀On฀main฀trunk ฀ ฀ 1-4-5-7-15-(20)-23-27-30 ฀ b)฀On฀thicker฀branches฀and฀leafless฀twigs ฀ ฀ 1-2-3-4-5-8-10-12-15-16-17-18-19-20-21-22-23-25-26-฀ ฀ 27-28-29-30-31 ฀ c)฀On฀leafy฀twigs ฀ ฀ 1-2-4-6-8-9-11-13-14-15-16-17-19-20-22-23-24-25-28-฀ ฀ 29-31 8.฀Inflorescence ฀ a)฀Branching฀(versus฀simple) ฀ ฀ 5-18 ฀ 9.฀Pedicel฀length฀(in฀flower) ฀ a)฀<10฀mm฀ ฀ ฀ 14-(16)-20-29 ฀ b)฀10-20฀mm ฀ ฀ 1-3-5-9-10-11-15-16-19-20-21-22-25-27 ฀ c)฀21-80฀mm ฀ ฀ 1-2-5-8-9-10-13-(15)-16-17-18-23-24-26-31 ฀ d)฀>80฀mm ฀ ฀ 4-12-(23) ฀ 10.฀Pedicel฀length฀(in฀fruit) ฀ a)฀<20฀mm฀ ฀ ฀ 3-5-6-9-10-11-14-15-16-19-20-22-25-27-28-29-30 ฀ b)฀20-39฀mm ฀ ฀ 1-2-3-5-7-9-10-11-14-15-16-17-18-19-20-21-22-23-฀ ฀ 25-27-30 ฀ c)฀40-100฀mm ฀ ฀ 1-2-5-7-8-13-(15)-17-(20)-23-24 ฀ d)฀>100฀mm ฀ ฀ 4-12-(23) 106 • Chapter 5 11.฀Pedicel฀indument ฀ a)฀Hairy ฀ ฀ 1-2-3-4-5-6-8-9-11-12-16-18-19-20-21-23-26-28 ฀ b)฀Glabrous ฀ ฀ 3-7-10-11-13-14-15-17-21-22-23-24-25-27-29-30-31 12.฀Flower฀buds ฀ a)฀Opening฀widely ฀ ฀ 3-5-10-18-20-22 ฀ b)฀Opening฀loosely ฀ ฀ 1-2-8-9-11-12-13-14-15-16-19-21-23-26-27-28 ฀ c)฀Remaining฀closed ฀ ฀ 4-17-24-25-29 13.฀Closed฀flower฀bud฀shape ฀ a)฀Broadly฀ovoid-triangular฀(versus฀depressed฀ovoid) ฀ ฀ 4-17-25 14.฀Sepals฀length ฀ a)฀≤2.5฀mm ฀ ฀ 1-2-10-13-(14)-17-18-20-22-23-24-26-27-28 ฀ b)฀2.6-5฀mm ฀ ฀ 3-5-8-9-10-11-12-14-15-16-17-18-21-22-23-25-29-31 ฀ c)฀5.1-7.5 ฀ ฀ 4-14-15-19 ฀ d)฀>7.5 ฀ ฀ 19 15.฀Sepals฀indument ฀ a)฀Hairy ฀ ฀ 1-2-4-5-8-9-12-16-18-19-(20)-26-28 ฀ b)฀Glabrous ฀ ฀ 3-10-11-13-14-15-17-20-21-22-23-24-25-27-29-31 16.฀Sepals ฀ a)฀Mostly฀persistent ฀ ฀ 14 ฀ b)฀Occasionally฀and/or฀partially฀persistent ฀ ฀ 8-15-19-22-28 ฀ c)฀Caducous ฀ ฀ 1-2-3-4-5-9-10-11-12-13-16-17-18-20-21-23-24-25-27-29฀ ฀ 30-31 Revision and phylogeny of Cremastosperma • 107 17.฀Monocarp฀shape ฀ a)฀Globose฀to฀transversely฀broadly฀ellipsoid ฀ ฀ 5-14-22 ฀ b)฀Ellipsoid ฀ ฀ 1-2-3-4-6-7-8-9-10-11-12-13-15-16-17-18-19-20-21-23฀ ฀ 24-25-27-28-29-30-31 ฀ c)฀Narrowly฀ellipsoid ฀ ฀ 8 18.฀Monocarp฀symmetry ฀ a)฀Strongly฀asymmetrical฀(versus฀slightly฀so) ฀ ฀ 1-7 ฀ 19.฀Monocarp฀length ฀ a)฀<12฀mm฀ ฀ ฀ 3-5-9-16-17-22-24-31 ฀ b)฀12-18฀mm ฀ ฀ 1-3-4-5-6-7-9-10-11-14-18-19-20-21-22-23-24-25-28฀ ฀ 29-30-31 ฀ c)฀18.1-24฀mm ฀ ฀ 2-12-13-15-18-19-20-25-27-28 ฀ d)฀>24฀mm ฀ ฀ 2-8 20.฀Monocarps ฀ a)฀Hairy฀(versus฀(appearing)฀glabrous) ฀ ฀ 4-5-6-9-16-19 ฀ 21.฀Stipe฀length ฀ a)฀<4฀mm฀ ฀ ฀ 19 ฀ b)฀4-10฀mm ฀ ฀ 3-5-6-9-11-13-14-15-16-17-19-20-21-22-24-28-31 ฀ c)฀10.1-20฀mm ฀ ฀ 1-2-3-4-5-7-8-9-10-11-13-15-16-17-18-20-21-22-23฀ ฀ 24-25-27-28-29-30-31 ฀ d)฀20.1-30฀mm ฀ ฀ 2-5-10-12-15-18-22-23-25 ฀ e)฀>30฀mm฀ ฀ ฀ ฀ (5)-25 108 • Chapter 5 22.฀Distribution ฀ a)฀French฀Guiana:฀ ฀ ฀ ฀ ฀ 3 ฀ b)฀N฀Venezuela:฀ ฀ ฀ ฀ ฀ 13-27 ฀ c)฀Central฀America฀(Costa฀Rica,฀Panama):฀ 22-28 ฀ d)฀Pacific฀coast฀of฀Colombia฀and฀Ecuador:฀ 2-7-12-19-21-26 ฀ e)฀The฀Colombian฀state฀of฀Antioquia:฀ ฀ 1-8-14 ฀ f)฀Amazonian฀Brazil:฀ ฀ ฀ ฀ ฀ 5-16-17-20 ฀ g)฀Amazonian฀Colombia:฀ ฀ ฀ ฀ 9-15-16 ฀ h)฀Amazonian฀Ecuador:฀ ฀ ฀ ฀ 5-9-15-18-23 ฀ i)฀Peru:฀ ฀ 4-5-6-9-15-16-17-23-24-25-29-31 ฀ j)฀N฀Peru:฀ ฀ 4-5-6-9-15-16-17-23-25-29 ฀ k)฀The฀Peruvian฀state฀of฀Amazonas:฀ ฀ 4-6-25-29 ฀ l)฀S฀and฀Central฀Peru:฀฀ ฀ ฀ ฀ 24-25-31 ฀ m)฀Bolivia:฀ ฀ ฀ ฀ ฀ 10-17 1. Cremastosperma antioquense฀ Pirie฀-฀Fig.฀5;฀Map฀2 Cremastosperma฀antioquense฀Pirie฀(2005)฀43,฀f.฀1.฀-฀Type:฀Soejarto฀3586฀(holo฀COL;฀iso฀ F,฀GH,฀HUA,฀MO),฀Colombia,฀Antioquia:฀Mun.฀Anorí,฀Corregimiento฀Providencia,฀ Buenos฀Aires,฀4฀km฀from฀Providencia,฀500-700฀m,฀10฀Dec.฀1972. Tree฀ c.฀ 5฀ m฀ tall;฀ young฀ twigs฀ and฀ petioles฀ sparsely฀ covered฀ with฀ appressed฀ brown฀hairs฀up฀to฀0.2฀mm฀long฀or฀glabrous.฀ Leaves:฀petioles฀7-10฀by฀2-3฀mm;฀ lamina฀narrowly฀elliptic,฀16-27฀by฀6-9.5฀cm฀(index฀2.3-2.8),฀chartaceous,฀drying฀ to฀a฀mosaic฀of฀brown฀and฀lighter฀green฀on฀both฀sides,฀glabrous฀on฀both฀sides,฀base฀ obtuse,฀apex฀acuminate฀(acumen฀10-15฀mm฀long),฀primary฀vein฀grooved฀in฀the฀ basal฀half,฀1-1.5฀mm฀wide฀at฀widest฀point,฀secondary฀veins฀8-11,฀intersecondary฀ veins฀occasional,฀distance฀between฀from฀10฀mm฀at฀the฀base฀to฀50฀mm฀closer฀to฀ the฀apex,฀angles฀with฀primary฀vein฀from฀50°฀at฀the฀base฀to฀70°฀closer฀to฀the฀apex,฀ forming฀distinct฀loops,฀smallest฀distance฀between฀loops฀and฀margin฀3-4฀mm,฀tertiary฀ veins฀more฀or฀less฀percurrent.฀Inflorescence฀of฀single฀flowers,฀axillary฀on฀leafy฀twigs฀or฀ from฀main฀trunk,฀then฀solitary฀or฀clustered฀in฀groups฀of฀at฀least฀two฀on฀brachyblasts;฀ peduncles฀c.฀2฀by฀1.5฀mm฀(in฀flower),฀2-3฀by฀1.5-2฀mm฀(in฀fruit);฀pedicels฀20-28฀ by฀c.฀1฀mm฀diam.฀at฀the฀base,฀1.5-2฀mm฀diam.฀at฀the฀apex฀(in฀flower),฀20-40฀by฀c.฀2฀ mm฀diam.฀at฀the฀base,฀c.฀3฀mm฀diam.฀at฀the฀apex฀(in฀fruit),฀peduncles฀and฀pedicels฀ sparsely฀ covered฀ with฀ appressed฀ whitish-golden฀ hairs฀ to฀ 0.2฀ mm฀ long;฀ 2฀ lower฀ bracts,฀deltate,฀c.฀1฀mm฀long,฀obtuse,฀caducous;฀upper฀bract฀attached฀around฀halfway฀ along฀pedicel,฀deltate,฀1฀mm฀long,฀obtuse,฀outer฀side฀of฀upper฀and฀lower฀bracts฀ rather฀densely฀to฀densely฀covered฀with฀appressed฀whitish-golden฀hairs฀to฀0.2฀mm฀ long;฀closed฀flower฀buds฀not฀seen;฀flowers฀light฀green,฀stamens฀and฀carpels฀yellowish฀ or฀pinkish฀in฀vivo,฀petals฀dark฀brown,฀contrasting฀to฀lighter฀colour฀of฀sepals฀and฀ Revision and phylogeny of Cremastosperma • 109 F i g . 5 . Cremastosperma antioquense Pirie. a. leaf and flower; b. fruit (a: Soejarto 2798; b: Soejarto 3586) pedicels฀in฀sicco;฀sepals฀fused฀at฀base,฀deltate,฀appressed,฀2-2.5฀by฀2-2.5฀mm,฀acute,฀ caducous,฀sparsely฀to฀rather฀densely฀covered฀with฀appressed฀whitish-golden฀hairs฀ to฀0.2฀mm฀long;฀outer฀petals฀elliptic,฀c.฀12฀by฀8฀mm,฀inner฀petals฀elliptic,฀10-12฀by฀ 5-6฀mm,฀outer฀side฀of฀outer฀and฀inner฀petals฀rather฀densely฀covered฀with฀appressed฀ whitish-golden฀hairs฀to฀0.2฀mm฀long;฀receptacle฀depressed฀ovoid;฀androecium฀5-7฀ mm฀diam.,฀stamens฀c.฀1฀mm฀long,฀connective฀appendage฀roughly฀rhombic,฀0.5-0.7฀ mm฀wide,฀glabrous;฀gynoecium฀1-1.5฀mm฀diam.,฀carpels฀c.฀1.5฀mm฀long,฀glabrous.฀ Monocarps,฀stipes,฀and฀receptacle฀glabrous,฀monocarps฀c.฀10,฀ellipsoid฀to฀broadly฀so,฀ strongly฀asymmetrical,฀13-14฀by฀c.฀11฀mm,฀orange฀to฀deep฀red,฀maturing฀to฀black฀in฀ vivo,฀dark฀reddish฀brown฀in฀sicco,฀with฀an฀excentral฀apicule;฀stipes฀orange฀to฀deep฀ red฀in฀vivo,฀c.฀20฀by฀1.5฀mm;฀fruiting฀receptacle฀depressed฀ovoid,฀c.฀6฀mm฀diam.฀ Seeds฀ellipsoid,฀reddish฀brown฀with฀dark฀pits฀each฀surrounded฀by฀a฀raised฀rim,฀c.฀12฀ by฀9฀mm,฀raphe฀sunken,฀regular. 110 • Chapter 5 Distribution฀-฀Colombia฀(Antioquia). Habitat฀and฀Ecology฀-฀Primary฀forest.฀At฀elevations฀of฀500-700฀m.฀Flowering:฀ February,฀fruiting:฀December. Notes฀-฀Cremastosperma฀antioquense฀is฀similar฀to฀C.฀awaense,฀particularly฀in฀the฀ appearance฀of฀the฀flowers.฀However,฀the฀fruits฀of฀the฀two฀species฀are฀more฀distinct:฀ in฀contrast฀to฀C.฀awaense,฀the฀monocarps฀of฀C.฀antioquense฀are฀smaller,฀shorter฀than฀ the฀stipes,฀strongly฀asymmetrical฀and฀entirely฀glabrous.฀In฀addition,฀none฀of฀the฀ collections฀of฀C.฀awaense฀display฀cauliflory,฀a฀condition฀found฀in฀both฀of฀the฀two฀ collections฀of฀C.฀antioquense. 2. Cremastosperma awaense฀ Pirie฀-฀Fig.฀6;฀Map฀3 Cremastosperma฀awaense฀Pirie฀(2005)฀45,฀f.฀2.฀-฀Type:฀Aulestia฀842฀(holo฀QCNE;฀iso฀U),฀ Ecuador,฀Carchi:฀Maldonado,฀parish฀of฀Tobar฀Donoso,฀Ethnic฀Reserve฀Awá,฀Sabalera,฀ 900฀m,฀22฀Nov.฀1992.฀ Tree฀4-15(-20)฀m฀tall,฀8-25฀cm฀diam.;฀young฀twigs฀and฀petioles฀sparsely฀to฀ rather฀densely฀covered฀with฀appressed฀golden฀hairs฀to฀0.3฀mm฀long.฀Leaves:฀petioles฀ 4-11(15)฀by฀1.5-3฀mm;฀lamina฀narrowly฀elliptic฀to฀slightly฀obovate,฀17-33฀by฀5.513฀cm฀(index฀2.2-3.6),฀chartaceous,฀brown/grey฀green฀above,฀darker฀below,฀veins฀ on฀underside฀dark฀brown,฀glabrous฀above,฀veins฀sparsely฀to฀rather฀densely฀covered฀ with฀appressed฀golden฀hairs฀to฀0.3฀mm฀long฀below,฀base฀obtuse฀to฀acute,฀apex฀ accuminate฀(acumen฀10-20฀mm฀long),฀primary฀vein฀1-3฀mm฀wide฀at฀widest฀point,฀ secondary฀veins฀7-11,฀intersecondary฀veins฀occasional,฀distance฀between฀from฀10฀ mm฀at฀the฀base฀to฀up฀to฀60฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀from฀ 45-50°฀at฀the฀base฀to฀55-60°฀closer฀to฀the฀apex,฀forming฀loops฀in฀the฀apical฀half,฀ smallest฀ distance฀ between฀ loops฀ and฀ margin฀ 1-3฀ mm,฀ tertiary฀ veins฀ with฀ some฀ reticulation.฀Inflorescence฀of฀single,฀solitary฀flowers,฀axillary฀on฀leafy฀or฀leafless฀twigs;฀ peduncles฀c.฀1.5฀by฀1฀mm฀(in฀flower),฀1.5-3฀by฀1-2฀mm฀(in฀fruit);฀pedicels฀2728฀by฀c.฀1฀mm฀(in฀flower),฀35-60฀by฀1-2฀mm฀(in฀fruit),฀peduncles฀and฀pedicels฀ rather฀densely฀to฀densely฀covered฀with฀appressed฀golden฀hairs฀to฀0.3฀mm฀long;฀ single฀lower฀bract,฀broadly฀elliptic,฀1-2฀by฀1-1.5฀mm,฀obtuse,฀caducous,฀outer฀side฀ densely฀covered฀with฀appressed฀golden฀hairs฀to฀0.3฀mm฀long;฀upper฀bract฀attached฀ around฀midway฀along฀the฀pedicel,฀broadly฀elliptic,฀1-2.5฀by฀1-2฀mm,฀obtuse,฀rather฀ densely฀to฀densely฀covered฀with฀appressed฀golden฀hairs฀to฀0.3฀mm฀long;฀flower฀ buds฀depressed฀ovoid;฀flowers฀green฀or฀cream฀in฀vivo,฀blackish฀in฀sicco;฀sepals฀free,฀ deltate,฀reflexed฀(appressed฀in฀bud),฀2-2.5฀by฀2-2.5฀mm,฀obtuse,฀caducous,฀outer฀ side฀ rather฀ densely฀ to฀ densely฀ covered฀ with฀ appressed฀ golden฀ hairs฀ to฀ 0.3฀ mm฀ long;฀outer฀petals฀elliptic฀to฀broadly฀elliptic,฀10-15฀by฀8-9฀mm,฀inner฀petals฀elliptic,฀ 10-15฀by฀5-6฀mm,฀appearing฀glabrous฀but฀sparsely฀to฀rather฀densely฀covered฀with฀ appressed฀golden฀hairs฀to฀0.2฀mm฀long฀on฀the฀outer฀side,฀denser฀at฀the฀base฀and฀ in฀a฀band฀leading฀from฀the฀base฀to฀the฀apex฀of฀the฀petals;฀stamens฀1-1.5฀mm฀long,฀ connective฀appendage฀c.฀1฀mm฀wide;฀gynoecium฀c.฀2฀mm฀diam.,฀carpels฀30-40,฀ Revision and phylogeny of Cremastosperma • 111 Fig. 6. Cremastosperma awaense Pirie. a. flower; b. leaf and fruit (a: Aulestia 842; b: Van der Werff 12045) 1-2฀mm฀long,฀sparsely฀covered฀with฀golden,฀<฀0.1฀mm฀long฀hairs.฀Monocarps฀1012(-20),฀ellipsoid,฀slightly฀asymmetrical,฀22-28฀by฀12-17฀mm,฀brown฀in฀sicco,฀with฀ an฀excentral฀apicule฀or฀rarely฀a฀nipple-like฀protuberance;฀stipes฀11-24฀by฀1-1.5฀ mm;฀fruiting฀receptacle฀depressed฀ovoid,฀3.5-8฀mm฀diam.,฀monocarps,฀stipes฀and฀ receptacle฀very฀sparsely฀to฀sparsely฀covered฀with฀appressed฀white฀hairs,฀<0.1฀mm฀ long.฀Seeds฀ellipsoid,฀asymmetrical,฀yellow-orange฀with฀shallow฀pits,฀c.฀19฀by฀11฀ mm,฀raphe฀sunken,฀regular. 112 • Chapter 5 Distribution฀-฀Pacific฀coast฀of฀Ecuador฀(Esmeraldas฀and฀Carchi)฀and฀Colombia฀ (Nariño฀and฀Chocó). Habitat฀ and฀ Ecology฀ -฀ Primary฀ humid฀ to฀ pre-montane฀ tropical฀ forest.฀At฀ elevations฀of฀0-2,000฀m.฀Flowering:฀January,฀September,฀and฀November;฀fruiting:฀ January,฀February฀and฀June฀to฀September. Vernacular฀ names฀ -฀ Colombia:฀ Guasca฀ negra.฀ Ecuador:฀ Cargadera฀ negra,฀ Castaña฀negro,฀Huasca฀negra,฀Teuug฀teiug. Note฀-฀Cremastosperma฀awaense฀can฀be฀distinguished฀by฀the฀unique฀pattern฀of฀ indument฀on฀the฀outer฀sides฀of฀the฀petals:฀denser฀at฀base฀and฀in฀a฀line฀leading฀to฀the฀ petal฀apex.฀The฀sparse฀indument฀of฀very฀short฀(<0.1mm)฀hairs฀on฀the฀monocarps฀ and฀stipes฀are฀not฀visible฀without฀magnification,฀and฀the฀fruits฀appear฀glabrous.฀This฀ character฀is฀also฀exhibited฀by฀some฀specimens฀of฀C.฀westrae.฀C.฀awaense฀can฀easily฀be฀ distinguished฀from฀both฀C.฀westrae฀and฀the฀geographically฀closer฀C.฀stenophyllum฀ Pirie฀on฀the฀basis฀of฀the฀length฀of฀the฀pedicel.฀That฀of฀C.฀westrae฀is฀shorter฀(not฀ exceeding฀17฀mm)฀and฀that฀of฀C.฀stenophyllum฀longer฀(c.฀45฀mm฀in฀comparison฀to฀ 27-28฀mm฀in฀flower). 3. Cremastosperma brevipes (DC.)฀R.E.Fr.฀-฀Fig.฀7;฀Map฀4;฀App.฀2,฀Fig.฀2 Cremastosperma฀brevipes฀(DC.)฀R.E.Fr.฀(1939)฀538.฀-฀Guatteria฀brevipes฀DC.฀in฀Dunal฀ (1817)฀ 126.฀ -฀Type:฀ Martin฀ s.n.฀ (lecto฀ G,฀ iso฀ BM,฀ K,฀ S),฀ French฀ Guiana,฀ without฀ location. Cremastosperma฀poiteaui฀(Diels)฀R.E.Fr.฀(1931)฀328.฀-฀Guatteria฀poiteaui฀Diels฀(1931)฀ 74.฀-฀Type:฀Poiteau฀s.n.฀(holo฀G;฀iso฀B,฀F,฀G),฀French฀Guiana,฀without฀location,฀18191821. Tree฀4-20฀m฀tall,฀3-20฀cm฀diam.;฀young฀twigs฀and฀petioles฀glabrous฀to฀sparsely฀ covered฀with฀appressed฀white/yellowish฀hairs฀to฀0.2฀mm฀long.฀Leaves:฀petioles฀4-9(12)฀by฀1.5-4฀mm;฀lamina฀elliptic฀to฀obovate฀(or฀narrowly฀so),฀18-39฀by฀7-15฀cm฀(index฀ 2-3.5),฀chartaceous,฀brown/green฀with฀a฀reddish฀tinge฀on฀both฀sides฀(particularly฀ on฀the฀veins฀on฀the฀underside),฀darker฀above,฀glabrous฀on฀both฀sides,฀base฀obtuse฀to฀ rounded฀(rarely฀acute),฀apex฀acuminate฀(acumen฀5-20฀mm฀long),฀primary฀vein฀1.53฀mm฀wide฀at฀widest฀point,฀secondary฀veins฀9-15,฀intersecondary฀veins฀0-2,฀distance฀ between฀from฀5-10฀mm฀at฀the฀base฀to฀35฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀ vein฀from฀45-60°฀at฀the฀base฀to฀55-70°฀closer฀to฀the฀apex,฀not฀branching,฀forming฀ mostly฀distinct฀loops,฀smallest฀distance฀between฀loops฀and฀margin฀1-3฀mm,฀tertiary฀ veins฀percurrent.฀Inflorescence฀of฀single฀flowers,฀solitary฀or฀clustered฀in฀groups฀of฀2,฀ on฀leafless฀twigs;฀peduncles฀1-2฀by฀1-2฀mm฀(in฀flower),฀2-5฀by฀c.฀2฀mm฀(in฀fruit),฀ sparsely฀to฀rather฀densely฀covered฀with฀appressed฀white/yellowish฀c.฀0.1฀mm฀long฀ hairs;฀pedicels฀15-20฀by฀1-2฀mm฀at฀the฀base฀(in฀flower),฀18-23฀by฀c.฀2฀mm฀(in฀fruit),฀ sparsely฀covered฀with฀appressed฀white/yellowish฀hairs฀c.฀0.1฀mm฀long฀or฀glabrous;฀2฀ lower฀bracts,฀deltate,฀basal฀to฀0.2฀mm฀long,฀apical฀0.3-0.5฀mm฀long,฀obtuse,฀mostly฀ Revision and phylogeny of Cremastosperma • 113 a b F i g . 7 . Cremastosperma brevipes (DC.) R.E.Fr. a. fruiting specimen; b. flower buds (a: Boom 10812; b: Prévost 3446) 114 • Chapter 5 persistent,฀rather฀densely฀covered฀with฀appressed฀white/yellowish฀hairs฀c.฀0.1฀mm฀ long;฀upper฀bract฀mostly฀attached฀midway฀along฀pedicel,฀deltate฀to฀broadly฀ovate,฀ 1-2฀by฀c.฀1฀mm,฀obtuse฀or฀emarginate,฀sparsely฀covered฀with฀appressed฀white/ yellowish฀hairs฀c.฀0.1฀mm฀long;฀closed฀flower฀buds฀depressed฀ovoid,฀opening฀in฀ development;฀flowers฀green,฀sometimes฀tinged฀with฀red฀around฀margins฀of฀petals฀ or฀creamy฀yellow฀in฀vivo,฀reddish฀or฀dark฀brown฀in฀sicco,฀sepals฀and฀petals฀glabrous;฀ sepals฀connate฀for฀0.5-1฀mm,฀broadly฀ovate,฀recurved฀(rarely฀appressed),฀3-4{-5}฀by฀ 3.5-5฀mm,฀obtuse,฀caducous;฀outer฀petals฀elliptic,฀12-22฀by฀7-12฀mm,฀rounded,฀inner฀ petals฀narrowly฀obovate฀to฀narrowly฀elliptic,฀10-24฀by฀4-7฀mm,฀obtuse;฀androecium฀ c.฀ 6฀ mm฀ diam.;฀ stamens฀ 1.4-1.6฀ mm฀ long,฀ connective฀ appendage฀ 0.5-0.7฀ mm฀ wide;฀gynoecium฀c.฀1฀mm฀diam.;฀carpels฀c.฀25,฀c.2.3฀mm฀long,฀glabrous.฀Monocarps฀ 7-17,฀ellipsoid,฀slightly฀asymmetrical,฀11-17฀by฀9-11฀mm,฀green฀maturing฀to฀red,฀ reddish฀brown,฀dark฀purple฀or฀black฀in฀vivo,฀blackish฀or฀reddish฀brown฀in฀sicco,฀ with฀an฀excentric฀apicule;฀stipes฀green฀maturing฀to฀red฀in฀vivo,฀7-14฀by฀1-1.5(3)฀ mm;฀fruiting฀receptacle฀depressed฀ovoid,฀4-8฀mm฀diam.,฀monocarps,฀stipes,฀and฀ receptacle฀glabrous.฀Seeds฀broadly฀ellipsoid฀to฀globose,฀yellowish฀or฀orange-brown,฀ lightly฀pitted,฀c.฀8฀by฀7-8฀mm,฀raphe฀raised฀within฀a฀sunken฀groove,฀somewhat฀ irregular. Distribution฀-฀French฀Guiana:฀region฀of฀Saül฀and฀Nouragues. Habitat฀and฀Ecology฀-฀Primary฀moist฀forest.฀At฀elevations฀of฀200฀-฀800฀m.฀ Flowering:฀February,฀March,฀May,฀and฀October;฀fruiting:฀May฀-฀October,฀December,฀ and฀January. Vernacular฀names฀-฀French฀Guiana:฀Apélému,฀Maman฀yawé. Notes฀-฀Cremastosperma฀brevipes฀is฀the฀only฀species฀of฀the฀genus฀found฀in฀the฀ Guianas.฀The฀leaves,฀when฀dried,฀have฀a฀characteristic฀reddish฀tinge฀(particularly฀on฀ the฀underside).฀This฀species฀is฀similar฀to฀C.฀venezuelanum,฀but฀differing฀in฀particular฀ by฀ the฀ acuminate฀ as฀ opposed฀ to฀ acute฀ or฀ obtuse฀ leaf฀ apex฀ of฀ C.฀ venezuelanum,฀ smaller฀sepals฀and฀monocarps฀and฀shorter฀stipes.฀ Its฀use฀as฀fish฀bait฀has฀been฀reported.฀As฀of฀many฀Annonaceae,฀the฀bark฀of฀฀ C.฀brevipes฀has฀been฀described฀as฀aromatic:฀on฀collection฀Riera฀668฀a฀peppery฀smell฀ is฀reported. 4. Cremastosperma bullatum฀ Pirie฀-฀Fig.฀8;฀Map฀5;฀App.฀2,฀Fig.฀3 Cremastosperma฀bullatum฀Pirie฀in฀Pirie฀&฀Zapata฀(2004)฀8,฀f.฀2,฀3-5.฀-฀Type:฀Pirie฀et฀al.฀ 71฀(holo฀U;฀iso฀AAU,฀AMAZ,฀CUZ,฀E,฀F,฀HAO,฀HUT,฀K,฀MO,฀MOL,฀NY,฀US,฀USM,฀ WU),฀Peru,฀Amazonas:฀Bagua,฀District฀Imaza,฀community฀Yamayakat,฀trail฀to฀Putuim,฀ 420฀m,฀22฀Nov.฀2003.฀ Tree฀2-10฀m฀tall;฀young฀twigs฀and฀petioles฀densely฀covered฀with฀mainly฀erect฀ golden฀hairs฀up฀to฀1฀mm฀long.฀Leaves:฀petioles฀3-7฀mm฀by฀2.5-3฀mm;฀lamina฀elliptic฀ or฀narrowly฀so฀to฀slightly฀obovate,฀17-28฀by฀6-11฀cm฀(index฀2.4-3.5),฀chartaceous,฀ Revision and phylogeny of Cremastosperma • 115 F i g . 8 . Cremastosperma bullatum Pirie a. leaf; b. leaf base; c. flower; d. fruit (a-c: Vásquez et al. 24891; d: Jaramillo, N. et al. 972) mid฀ brown,฀ occasionally฀ slightly฀ grey฀ above฀ (immature฀ leaves฀ drying฀ black),฀ sparsely฀covered฀with฀mainly฀erect฀golden฀hairs฀up฀to฀1฀mm฀long฀or฀glabrous฀above,฀ densely฀so฀on฀edge฀of฀lamina฀and฀on฀all฀veins฀below,฀base฀rounded฀to฀subcordate,฀ apex฀acuminate฀(acumen฀5-20฀mm฀long),฀primary,฀secondary฀and฀tertiary฀veins฀ sunken฀in฀depressions฀in฀leaf฀surface,฀primary฀vein฀1.5-2฀mm฀wide฀at฀widest฀point,฀ densely฀covered฀with฀mainly฀erect฀golden฀hairs฀up฀to฀1฀mm฀long฀above฀and฀below,฀ 116 • Chapter 5 secondary฀veins฀15-20฀(intersecondary฀veins฀rare),฀distance฀between฀from฀6฀mm฀at฀ the฀base฀to฀16฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀consistently฀around฀ 60-70°,฀occasionally฀branching,฀forming฀distinct฀loops,฀smallest฀distance฀between฀ loops฀and฀margin฀1-1.5฀mm;฀tertiary฀veins฀largely฀percurrent.฀Inflorescences฀of฀ single,฀successively฀produced,฀flowers,฀axillary฀on฀leafy฀branches,฀on฀leafless฀branches฀ and฀produced฀from฀the฀main฀trunk฀(then฀on฀brachyblasts);฀peduncles฀and฀pedicels฀ sparsely฀to฀rather฀densely฀covered฀with฀mainly฀erect฀golden฀hairs฀up฀to฀1฀mm฀long,฀ peduncles฀17-20฀mm฀by฀1-1.5฀mm฀diam.฀(in฀flower),฀18-25฀mm฀by฀c.฀1.5฀mm฀(in฀ fruit);฀pedicels฀c.100฀to฀120฀mm฀by฀c.฀1฀mm฀at฀the฀base฀(in฀flower),฀110-150฀by฀ c.฀1.5฀mm฀(in฀fruit);฀bracts฀densely฀covered฀with฀mainly฀erect฀golden฀hairs฀up฀to฀ 1฀mm฀long,฀single฀lower฀bract,฀elliptic฀to฀ovate,฀2.5฀mm฀long,฀1฀mm฀wide,฀acute,฀ persistent฀or฀partially฀caducous;฀upper฀bract฀within฀central฀third฀of฀pedicel฀length,฀ elliptic฀to฀ovate,฀2-3฀mm฀long,฀1-2฀mm฀wide,฀acute;฀closed฀flower฀buds฀depressed฀ ovoid฀developing฀to฀ovoid฀before฀opening,฀green฀in฀vivo,฀brown฀in฀sicco;฀flowers฀ green,฀maturing฀to฀yellow฀with฀a฀basal฀orange฀patch฀on฀the฀outside฀of฀the฀outer฀ petals฀in฀vivo,฀golden฀brown฀in฀sicco,฀outer฀sides฀and฀apical฀portion฀of฀the฀inner฀ sides฀of฀petals฀and฀outer฀sides฀of฀sepals฀densely฀covered฀in฀appressed฀golden฀hairs฀ up฀to฀1฀mm฀long,฀inner฀sides฀otherwise฀glabrous;฀sepals฀basally฀connate,฀5-7฀by฀c.฀ 6฀mm,฀deltate,฀acute,฀caducous,฀outer฀petals฀broadly฀ovate,฀c.฀18฀by฀15฀mm,฀inner฀ petals฀ovate,฀concave,฀c.฀25฀by฀12฀mm฀wide;฀androecium฀c.฀5฀mm฀diam.,฀stamens฀ 1-1.5฀mm฀long,฀connective฀appendage฀of฀inconsistent฀and฀irregular฀shape,฀c.฀0.5฀ mm฀wide;฀gynoecium฀{c.฀2.5}฀mm฀diam.,฀glabrous.฀Monocarps,฀stipes,฀and฀receptacle฀ sparsely฀to฀moderately฀densely฀covered฀with฀erect฀golden฀hairs฀up฀to฀0.2฀mm฀long,฀ monocarps฀ 8-10,฀ dark฀ brown฀ in฀ sicco,฀ ellipsoid,฀ slightly฀ asymmetrical,฀ c.฀ 15฀ by฀ 11฀mm฀diam.,฀often฀with฀an฀excentric฀apicule;฀stipes฀14-16฀by฀1.5฀mm;฀fruiting฀ receptacle฀5-6฀mm฀in฀diam.฀Seeds฀ellipsoid,฀orange-brown,฀shallowly฀pitted,฀c.13฀by฀ 10฀mm,฀raphe฀raised,฀regular. Distribution฀-฀Peru฀(Amazonas).฀ Habitat฀and฀Ecology฀-฀Primary฀forest฀on฀red฀clay.฀At฀elevations฀of฀300-500฀m.฀ Flowering:฀February฀and฀November;฀fruiting:฀November฀and฀June.฀ Notes฀-฀Cremastosperma฀bullatum฀can฀easily฀be฀distinguished฀from฀all฀other฀species฀ of฀Cremastosperma฀by฀any฀one฀of฀the฀number฀of฀unique฀and฀striking฀characteristics฀ it฀displays.฀The฀leaf฀blade฀has฀a฀blistered฀or฀bubbled฀(bullate)฀appearance,฀both฀in฀the฀ field฀and฀when฀pressed,฀which฀is฀due฀to฀the฀deeply฀sunken฀nature฀of฀the฀primary,฀ secondary,฀and฀tertiary฀venation.฀The฀indument฀present฀on฀many฀of฀its฀parts฀is฀ far฀longer฀than฀in฀any฀other฀species฀in฀the฀genus,฀and,฀also฀uniquely฀in฀the฀genus,฀ densely฀inserted฀in฀a฀halo-like฀formation฀around฀the฀leaf฀margin.฀Other฀notable฀ characteristics฀are฀the฀unusually฀long฀pedicel,฀the฀orange฀colouring฀of฀the฀base฀of฀ the฀outer฀petals฀of฀mature฀flowers,฀the฀inner฀petals฀considerably฀longer฀than฀the฀ outer฀petals,฀and฀the฀rounded฀to฀subcordate฀shape฀of฀the฀leaf฀base. Revision and phylogeny of Cremastosperma • 117 5. Cremastosperma cauliflorum฀ R.E.Fr.฀-฀Fig.฀2฀b,฀e,฀9;฀Map฀5;฀App.฀2,฀Fig.฀4 Cremastosperma฀cauliflorum฀R.E.Fr.฀(1931)฀330.฀-฀Type:฀Klug฀902฀(holo฀B;฀iso฀F,฀NY,฀ US),฀Peru,฀Loreto:฀Mishuyacu,฀near฀Iquitos,฀100฀m,฀Feb.-Mar.฀1930. Tree฀2-20฀m฀tall,฀4-25฀cm฀diam.;฀young฀twigs฀and฀petioles฀glabrous฀to฀rather฀ densely฀ covered฀ with฀ appressed฀ or฀ erect฀ golden฀ hairs฀ to฀ 0.5฀ mm฀ long.฀ Leaves:฀ petioles฀ 4-12(-16)฀ by฀ 2-4(-6)฀ mm;฀ lamina฀ elliptic฀ to฀ obovate฀ or฀ narrowly฀ so,฀฀ (14-)฀20-61฀by฀5-14(-22)฀cm฀(index฀2.3-3.7),฀chartaceous,฀olive/brown฀green฀above,฀ darker฀below,฀glabrous฀above฀except฀for฀base฀of฀primary฀vein฀sparsely฀covered฀with฀ appressed฀or฀erect฀hairs฀to฀0.3฀mm฀long,฀base,฀primary฀and฀secondary฀veins฀sparsely฀ to฀rather฀densely฀covered฀with฀appressed฀or฀erect฀golden฀hairs฀to฀0.5฀mm฀long฀ below,฀base฀acute฀to฀obtuse,฀apex฀acuminate฀(acumen฀5-45฀mm฀long),฀primary฀ vein฀verrucose฀(particularly฀at฀the฀base),฀deeply฀grooved฀for฀most฀of฀length,฀1.53.5(-5)฀ mm฀ wide฀ at฀ widest฀ point,฀ secondary฀ veins฀ (6-)10-17,฀ occasionally฀ 1-2฀ intersecondary฀veins,฀distance฀between฀from฀4฀mm฀at฀the฀base฀to฀up฀to฀40฀mm฀ closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀from฀45-70°฀at฀the฀base฀to฀45-60°฀ closer฀to฀the฀apex,฀not฀branching,฀forming฀mostly฀distinct฀loops,฀smallest฀distance฀ between฀loops฀and฀margin฀1-5฀mm,฀tertiary฀veins฀percurrent.฀Inflorescence฀of฀1-5฀ flowers,฀branching,฀solitary฀or฀clustered฀in฀groups฀of฀up฀to฀7,฀on฀thick฀leafless฀twigs฀ or฀on฀main฀trunk฀(then฀often฀on฀brachyblasts);฀peduncles฀3-12(-15)฀by฀1-1.5(-3)฀ mm฀(in฀flower),฀3-15฀by฀1-3฀mm฀(in฀fruit);฀pedicels฀10-45฀by฀1-3฀mm฀at฀the฀base฀ (in฀flower),฀15-45฀by฀1-3฀mm฀(in฀fruit),฀peduncles฀and฀pedicels฀rather฀densely฀to฀ densely฀covered฀with฀mainly฀erect฀golden฀hairs฀c.฀0.3฀mm฀long,฀often฀with฀hairs฀ more฀densely฀covering฀the฀articulation฀point฀between฀shoot฀and฀pedicel;฀single฀ lower฀bract฀(from฀the฀axil฀of฀which฀short฀shoots฀subtending฀new฀flowers฀develop),฀ deltate,฀1.5-2฀mm฀long,฀acute,฀caducous,฀densely฀covered฀with฀mostly฀appressed฀ golden฀hairs฀to฀0.3฀mm฀long;฀upper฀bract฀attached฀around฀midway฀along฀pedicel,฀ broadly฀to฀very฀broadly฀ovate฀or฀deltate,฀2-4฀mm฀long,฀obtuse฀or฀acute,฀outer฀side฀ densely฀covered฀with฀appressed฀or฀erect฀golden฀hairs฀to฀0.3฀mm฀long;฀closed฀flower฀ buds฀depressed฀ovoid,฀opening฀in฀development;฀flowers฀(pale)฀green,฀creamy฀white,฀ greenish฀yellow฀or฀yellow฀in฀vivo,฀brownish฀yellow฀or฀brown฀with฀orange,฀dark฀ brown฀or฀black฀base฀in฀sicco,฀outer฀side฀of฀sepals฀and฀petals฀densely฀covered฀with฀ erect฀or฀appressed฀golden฀hairs฀to฀0.4฀mm฀long,฀inner฀side฀of฀sepals฀and฀petals฀ sparsely฀to฀rather฀densely฀covered฀with฀erect฀hairs฀to฀0.4฀mm฀long฀or฀glabrous,฀ base฀glabrous;฀sepals฀free,฀broadly฀to฀very฀broadly฀ovate-deltate,฀mostly฀recurved,฀ 3-5฀by฀4-6฀mm,฀obtuse,฀caducous;฀outer฀petals฀elliptic฀to฀broadly฀elliptic,฀10-25(32)฀by฀9-17฀mm,฀inner฀petals฀elliptic,฀11-21(-32)฀by฀6-11฀mm;฀androecium฀7-10฀ mm฀diam.,฀stamens฀1.7-2.1฀mm฀long,฀connective฀appendage฀roughly฀rhombic฀to฀ diamond-shaped,฀0.7-1฀mm฀wide;฀gynoecium฀2-3฀mm฀diam.,฀carpels฀c.฀40,฀2-2.3฀ mm฀long,฀sparsely฀to฀rather฀densely฀covered฀with฀mostly฀appressed฀golden฀hairs฀ to฀0.2฀mm฀long.฀Monocarps฀9-41,฀globose฀to฀transversely฀broadly฀ellipsoid,฀slightly฀ asymmetrical,฀8-13฀by฀10-14฀mm,฀green฀maturing฀to฀orange,฀red,฀brown฀and฀black฀ 118 • Chapter 5 a b F i g . 9 . Cremastosperma cauliflorum R.E.Fr. a. fruiting specimen; b. inflorescence (a: Prance et al. 24094; b: Vásquez & Jaramillo 11423) in฀vivo,฀blackish฀brown฀or฀brown฀in฀sicco,฀sometimes฀with฀an฀apicule฀at฀or฀near฀ the฀apex,฀monocarps,฀stipes฀and฀receptacle฀rather฀densely฀covered฀with฀erect฀golden฀ hairs฀to฀0.2฀mm฀long;฀stipes฀7-23(-32)฀by฀1-2฀mm;฀fruiting฀receptacle฀depressed฀ Revision and phylogeny of Cremastosperma • 119 ovoid,฀4-11฀mm฀diam.฀Seeds฀broadly฀ellipsoid฀to฀globose,฀orange,฀pitted,฀9-10฀by฀ 9-10฀mm,฀raphe฀sunken,฀regular. Distribution฀ -฀ Amazonian฀ Colombia,฀ Ecuador,฀ Peru,฀ and฀ Brazil฀ (Acre,฀ Amazonas).฀ Habitat฀and฀Ecology฀-฀Moist฀primary฀forest,฀mostly฀non-inundated฀areas,฀on฀ clayey฀or฀lateritic฀soil฀or฀white฀sand.฀At฀elevations฀of฀100฀-฀500฀m.฀Flowering:฀June,฀ August,฀October-February;฀fruiting:฀May-February. Vernacular฀ names฀ -฀ Colombia:฀ Espintana฀ blanca.฀ Ecuador:฀ Mantach,฀ Moncapatahue,฀Piton,฀Uñetahue,฀Uñitague.฀Peru:฀Bara฀caspi,฀Espintana. Notes฀-฀Cremastosperma฀cauliflorum฀is฀one฀of฀the฀only฀two฀species฀of฀the฀genus฀ displaying฀a฀branched฀inflorescence.฀It฀can฀be฀discerned฀from฀the฀other,฀C.฀napoense,฀ by฀ the฀ presence฀ of฀ indument฀ on฀ the฀ monocarps฀ and฀ stipes฀ and฀ by฀ the฀ greater฀ length฀and฀density฀of฀hairs฀on฀the฀inflorescences.฀ The฀wood฀is฀aromatic,฀flowers฀reported฀as฀vanilla฀scented. 6. Cremastosperma cenepense฀ Pirie฀&฀Zapata฀-฀Fig.฀10;฀Map฀5 Cremastosperma฀cenepense฀Pirie฀&฀Zapata฀(2004)฀13,฀f.฀2,฀9.฀-฀Type:฀Rojas฀et฀al.฀269฀(holo฀ U;฀iso฀AMAZ,฀HUT,฀MO,฀USM);฀Peru,฀Amazonas:฀prov.฀Condorcanqui,฀Río฀Cenepa฀ region,฀community฀Mamayaque,฀400฀m,฀11฀Aug.฀1997.฀ Tree฀ c.฀ 10฀ m฀ tall;฀ young฀ twigs฀ and฀ petioles฀ sparsely฀ (axillary฀ buds฀ densely)฀ covered฀with฀appressed฀golden฀hairs฀c.฀0.1฀mm฀long.฀Leaves:฀petioles฀4-7฀by฀1-2฀ mm;฀lamina฀narrowly฀elliptic,฀12-22฀by฀4-8฀cm฀(index฀2.7-3),฀chartaceous,฀greyyellow฀green฀above,฀light฀brown฀or฀yellowish฀green฀below,฀glabrous฀on฀both฀sides,฀ base฀ cordate฀ to฀ subcordate,฀ apex฀ acuminate฀ (acumen฀ 8-10฀ mm฀ long),฀ primary฀ vein฀1-1.5฀mm฀wide฀at฀widest฀point,฀secondary฀veins฀7-12,฀intersecondary฀veins฀ occasional,฀distance฀between฀from฀2-5฀mm฀at฀the฀base฀to฀15-25(35)฀mm฀closer฀to฀ the฀apex,฀angles฀with฀primary฀vein฀from฀90-80°฀at฀the฀base฀to฀60-50°฀closer฀to฀ the฀apex,฀forming฀distinct฀loops,฀smallest฀distance฀between฀loops฀and฀margin฀2-5฀ mm,฀tertiary฀veins฀more฀or฀less฀percurrent.฀Inflorescence฀of฀single,฀solitary฀flowers,฀ axillary฀on฀leafy฀twigs;฀peduncles฀c.฀2฀mm฀by฀2฀mm฀(in฀fruit);฀pedicels฀c.฀8฀mm฀by฀ 2฀mm฀at฀the฀base฀(in฀fruit),฀peduncles฀and฀pedicels฀sparsely฀covered฀with฀appressed฀ golden฀hairs฀c.฀0.1฀mm฀long;฀2฀lower฀bracts,฀caducous;฀upper฀bract฀attached฀midway฀ along฀pedicel,฀caducous;฀flowers฀not฀observed.฀Monocarps฀8-10,฀blackish฀brown฀in฀ sicco,฀ellipsoid,฀slightly฀asymmetric,฀14-15฀by฀9-11฀mm,฀with฀an฀excentric฀apicule,฀ monocarps,฀stipes,฀and฀receptacle฀rather฀densely฀covered฀with฀appressed฀golden฀ hairs฀c.฀0.1฀mm฀long;฀stipes฀7-8฀mm฀by฀c.฀1.5฀mm;฀fruiting฀receptacle฀4-7฀mm฀ diam.฀Seeds฀ellipsoid,฀golden฀brown฀shallowly฀wrinkled฀(immature),฀c.฀12฀mm฀by฀7฀ mm,฀raphe฀sunken,฀regular. Distribution฀-฀Peru฀(Amazonas),฀in฀the฀area฀of฀the฀Cenepa฀River฀(a฀tributary฀ of฀the฀Marañon฀River). 120 • Chapter 5 F i g . 1 0 . Cremastosperma cenepense Pirie & Zapata. A. fruiting twig; B. leaf base; C. fruit (A, C: Rojas 269; B: Kayap 1078) Habitat฀and฀Ecology฀-฀Primary฀forest.฀At฀elevations฀of฀250-400฀m.฀Fruiting:฀ July฀and฀August. Notes฀-฀Cremastosperma฀cenepense฀is฀similar฀to฀C.฀yamayakatense฀and฀C.฀gracilipes.฀ It฀differs฀in฀the฀shape฀of฀the฀leaf฀base฀(cordate฀or฀subcordate฀as฀opposed฀to฀acute฀in฀C.฀ yamayakatense฀and฀C.฀gracilipes),฀the฀indument฀on฀the฀fruits฀(rather฀dense฀as฀opposed฀ to฀almost฀always฀absent฀in฀C.฀yamayakatense),฀and฀lengths฀of฀the฀pedicel฀(shorter฀ than฀that฀of฀C.฀gracilipes)฀and฀stipes฀(shorter฀than฀those฀of฀C.฀yamayakatense). Revision and phylogeny of Cremastosperma • 121 7. Cremastosperma chococola฀ Pirie฀-฀Fig.฀11;฀Map฀2 Cremastosperma฀chococola฀Pirie฀(2005)฀47,฀f.฀3.฀-฀Type:฀Gentry฀&฀Forero฀7286฀(holo฀MO;฀ iso฀COL),฀Colombia,฀Chocó:฀Alto฀de฀Buey,฀500-1200฀m,฀8฀Jan.฀1973. Tree฀c.฀5฀m฀tall;฀young฀twigs฀and฀petioles฀glabrous.฀Leaves:฀petioles฀5-8฀by฀1.5-2฀ mm;฀lamina฀narrowly฀elliptic,฀11-20฀by฀4-5.5฀cm฀(index฀3.7-4),฀chartaceous,฀dark/ olive฀brown,฀shiny฀above,฀lighter฀pinkish฀brown,฀matt฀below,฀glabrous฀above฀and฀ below,฀base฀acute฀to฀cuneate,฀apex฀acuminate฀(acumen฀7-10฀mm฀long),฀primary฀ F i g . 1 1 . Cremastosperma chococola Pirie. a. leaf and fruit (Gentry & Forero 7286) 122 • Chapter 5 vein฀c.฀1฀mm฀wide฀at฀widest฀point,฀verrucose฀below,฀secondary฀veins฀8-10,฀no฀ intersecondary฀veins,฀distance฀between฀from฀5฀mm฀at฀the฀base฀to฀29฀mm฀closer฀to฀ the฀apex,฀angles฀with฀primary฀vein฀from฀c.฀60°฀at฀the฀base฀to฀60-70°฀closer฀to฀the฀ apex,฀forming฀distinct฀loops,฀smallest฀distance฀between฀loops฀and฀margin฀2.5-3.5฀ mm,฀tertiary฀veins฀reticulate.฀Inflorescence฀of฀single฀flowers,฀solitary฀or฀clustered฀in฀ groups฀of฀at฀least฀two,฀on฀brachyblasts฀on฀the฀main฀trunk;฀peduncles,฀2-3฀by฀11.5฀mm฀(in฀fruit);฀pedicels฀38-42฀by฀1฀mm฀diam.฀at฀the฀base,฀1฀mm฀diam.฀at฀the฀ apex฀(in฀fruit),฀peduncles฀and฀pedicels฀glabrous;฀lower฀bract(s)฀not฀observed;฀upper฀ bract฀attached฀within฀basal฀half฀of฀pedicel,฀ovate,฀c.฀1฀by฀0.7฀mm,฀obtuse,฀glabrous;฀ flowers฀ not฀ observed.฀ Monocarps฀ 10-13,฀ ellipsoid,฀ strongly฀ asymmetrical฀ (stipes฀ inserted฀within฀basal฀half฀of฀longest฀axis),฀13-14฀by฀10-11฀mm,฀with฀an฀excentral,฀ to฀0.2฀mm฀long,฀apicule,฀green฀maturing฀through฀red฀to฀dark฀blue฀in฀vivo,฀dark฀ brown฀in฀sicco,฀glabrous;฀stipes฀15-18฀by฀c.฀1฀mm฀increasing฀to฀1.5฀diam.฀when฀ mature,฀glabrous;฀fruiting฀receptacle฀depressed฀ovoid,฀4-5฀mm฀diam.,฀glabrous.฀Seeds฀ ellipsoid,฀orange-brown,฀pitted,฀9-11฀by฀6-8฀mm,฀raphe฀sunken,฀regular. Distribution฀-฀Pacific฀coast฀of฀Colombia฀(Chocó). Habitat฀and฀Ecology฀-฀Tropical฀wet฀forest.฀At฀elevations฀of฀0-1200฀m.฀Fruiting:฀ January฀and฀June Notes฀-฀The฀strongly฀asymmetric฀monocarps฀of฀Cremastosperma฀ chococola฀ resemble฀those฀of฀C.฀antioquense,฀and฀collections฀of฀both฀species฀display฀cauliflory฀ (though฀not฀exclusively฀so฀in฀C.฀antioquense)฀with฀inflorescences฀inserted฀on฀similar฀ brachyblasts.฀However,฀C.฀chococola฀can฀easily฀be฀distinguished฀by฀its฀small,฀narrowly฀ elliptic฀ leaves฀ with฀ typical฀ pinkish-brown฀ colour฀ on฀ the฀ underside,฀ and฀ by฀ the฀ absence฀of฀hairs฀on฀the฀pedicels.฀ 8. Cremastosperma dolichocarpum฀ Pirie฀-฀Fig.฀12;฀Map฀3 Cremastosperma฀dolichocarpum฀Pirie฀(2005)฀49,฀f.฀4.฀-฀Type:฀Sánchez฀et฀al.฀415฀(holo฀U;฀ iso฀COL),฀Colombia,฀Antioquia:฀Frontino,฀Nutibara,฀upper฀watershed฀of฀Río฀Cuevas,฀ 15฀Jul.฀1986. Tree฀6-18฀m฀tall,฀15-22฀cm฀diam.;฀young฀twigs฀and฀petioles฀sparsely฀covered฀ with฀white-yellow฀appressed฀hairs฀0.3-0.5฀mm฀long.฀Leaves:฀petioles฀3-6฀by฀1.5-2฀ mm,฀often฀with฀warts฀extending฀up฀primary฀vein;฀axillary฀buds฀densely฀covered฀ with฀white-yellow฀appressed฀hairs฀0.3-0.5฀mm฀long;฀lamina฀elliptic฀to฀narrowly฀ so,฀14-24.5฀by฀6-10฀cm฀(index฀1.6-3.1),฀chartaceous฀or฀subcoriaceous,฀mid-dark฀ brown฀above,฀lighter฀below,฀glabrous฀above,฀sparsely฀covered฀with฀white-yellow฀ appressed฀hairs฀0.3-0.5฀mm฀long฀(particularly฀on฀veins)฀below,฀base฀obtuse-acute฀ (narrowly฀cuneate),฀apex฀acuminate฀(acumen฀4-15฀mm฀long),฀primary฀vein฀not฀ conspicuously฀grooved,฀1-2฀mm฀wide฀at฀widest฀point;฀secondary฀veins฀(5-)7-9(-11),฀ intersecondary฀veins฀occasional,฀distance฀between฀from฀5฀mm฀at฀the฀base฀to฀30฀mm฀ closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀40-50°฀at฀the฀base฀to฀50-70°฀closer฀ Revision and phylogeny of Cremastosperma • 123 F i g . 1 2 . Cremastosperma dolichocarpum Pirie. a. flower; b. leaf and fruit (a: Callejas 3110; b: Sánchez 415) to฀the฀apex,฀not฀branching,฀forming฀distinct฀loops฀in฀the฀apical฀half-two฀thirds,฀ smallest฀distance฀between฀loops฀and฀margin฀1.5-4฀mm;฀tertiary฀veins฀percurrent฀ with฀significant฀reticulation.฀Inflorescences฀of฀single฀flowers฀solitary฀or฀clustered฀in฀ groups฀of฀2฀(or฀more),฀produced฀from฀leafy฀twigs฀or฀leafless฀branches;฀peduncles฀ of฀two฀internodes,฀the฀second฀1.2-4฀mm฀long,฀c.฀1฀mm฀diam.฀(in฀flower),฀approx.฀ 2฀mm฀long,฀2฀mm฀diam.฀(in฀fruit);฀pedicels฀28-47฀mm฀long,฀c.฀1฀mm฀diam.฀at฀the฀ base,฀1-1.5฀mm฀diam.฀at฀the฀apex฀(in฀flower),฀40-55฀mm฀long,฀1.5-2฀mm฀diam.฀ 124 • Chapter 5 at฀the฀base,฀2-2.5฀mm฀diam.฀at฀the฀apex฀(in฀fruit);฀peduncles฀and฀pedicels฀rather฀ densely฀covered฀with฀white-yellow฀appressed฀hairs฀0.3-0.5฀mm฀long;฀two฀lower฀ bracts฀(one฀on฀each฀internode),฀the฀apical฀one฀persisting฀later฀into฀flowering,฀1-1.5฀ by฀0.7-1฀mm,฀broadly฀ovate,฀obtuse,฀caducous,฀rather฀densely฀covered฀with฀whiteyellow฀appressed฀hairs฀0.3-0.5฀mm฀long;฀upper฀bract฀broadly฀to฀narrowly฀ovate,฀ 1-3.5฀by฀1-2฀mm,฀obtuse,฀persistent,฀densely฀covered฀with฀white-yellow฀appressed฀ hairs฀0.3-0.5฀mm฀long;฀flower฀buds฀depressed฀ovoid;฀flowers฀green฀maturing฀to฀ yellow฀with฀formaline-like฀scent฀in฀vivo,฀brown฀outside฀and฀black฀inside฀in฀sicco;฀ sepals฀free,฀ovate,฀appressed,฀3-4฀mm฀long,฀obtuse,฀occasionally฀persistent฀on฀less฀ mature฀fruit,฀densely฀covered฀with฀white-yellow฀appressed฀hairs฀0.3-0.5฀mm฀long;฀ outer฀petals฀ovate฀to฀broadly฀so,฀10-15฀by฀9-11฀mm,฀inner฀petals฀ovate,฀10-16฀by฀ 7-8฀ mm฀ densely฀ covered฀ with฀ white-yellow฀ appressed฀ hairs฀ 0.3-0.5฀ mm฀ long;฀ receptacle฀ovoid฀to฀depressed฀ovoid;฀androecium฀3-5฀mm฀diam.,฀stamens฀1-1.2฀mm฀ long,฀connective฀appendage฀c.฀1฀mm฀wide,฀glabrous;฀gynoecium฀1.5-2฀mm฀diam.,฀ carpels฀0.5-0.6฀mm฀long,฀glabrous.฀Monocarps฀10-20฀(fully฀ripe฀fruit฀not฀observed),฀ black฀in฀sicco,฀ellipsoid฀or฀narrowly฀so,฀27-28฀mm฀long,฀11-12฀mm฀diam.,฀with฀an฀ excentric฀apicule฀(obvious฀only฀in฀immature฀fruit);฀stipes฀17-19฀mm฀by฀1.5-2฀mm;฀ monocarps฀and฀stipes฀sparsely฀covered฀with฀golden฀appressed฀hairs฀<฀0.1mm฀long฀ or฀glabrous;฀fruiting฀receptacle฀broadly฀ovoid,฀4-5฀mm฀diam.,฀often฀sparsely฀covered฀ with฀white-yellow฀appressed฀hairs฀0.3-0.5฀mm฀long.฀Seeds฀ellipsoid฀to฀narrowly฀ so,฀c.฀17฀mm฀long฀and฀7฀mm฀diam.,฀dark฀brown฀and฀wrinkled,฀raphe฀impressed,฀ encircling฀seed฀diagonally. Distribution฀-฀Colombia฀(Antioquia):฀northern฀and฀western฀foothills฀of฀the฀ Cordillera฀Occidental. Habitat฀and฀Ecology฀-฀At฀elevations฀of฀1200-1500฀m.฀Flowering:฀May฀and฀ December;฀fruiting:฀May,฀July,฀and฀December. Notes฀-฀Cremastosperma฀dolichocarpum฀can฀be฀distinguished฀from฀other฀species฀ of฀Cremastosperma฀by฀the฀unique฀long-ellipsoid฀monocarps฀after฀which฀the฀species฀ is฀named฀and฀identified฀even฀when฀sterile฀by฀the฀conspicuous฀axillary฀buds฀with฀ dense฀indument฀(otherwise฀only฀observed฀in C.฀novogranatense,฀where฀they฀ are฀less฀conspicuous). 9. Cremastosperma gracilipes฀ R.E.Fr.฀-฀Fig.฀1฀d,฀13;฀Map฀4 Cremastosperma฀gracilipes฀R.E.Fr.฀(1931)฀325.฀t.฀26.฀-฀Type:฀Tessmann฀4748฀(holo฀B;฀iso฀ S),฀Peru,฀Loreto:฀upper฀Río฀Marañon,฀Puerto฀Melendez,฀below฀Pongo฀de฀Manseriche,฀ 155฀m,฀13฀Dec.฀1924. Tree฀ or฀ shrub฀ 0.5-10฀ m฀ tall;฀ young฀ twigs฀ and฀ petioles฀ glabrous฀ to฀ sparsely฀ covered฀with฀appressed฀brown฀hairs฀to฀0.4฀mm฀long.฀Leaves:฀petioles฀2-8฀by฀1-2.5฀ mm;฀lamina฀elliptic฀to฀obovate฀or฀narrowly฀so,฀11-28฀by฀3-10฀cm฀(index฀2-4(-4.7)),฀ chartaceous,฀(pale)฀greyish฀or฀brownish฀green฀on฀both฀sides,฀often฀more฀greyish฀ Revision and phylogeny of Cremastosperma • 125 a b F i g . 1 3 . Cremastosperma gracilipes R.E.Fr. a. fruiting specimen; b. flower (a: Hurtado 3019; b: Palacios 1651) above,฀glabrous฀on฀both฀sides,฀base฀acute฀to฀obtuse฀or฀rounded,฀apex฀cuspidate฀ (cusp฀10-35฀mm฀long),฀primary฀vein฀lightly฀grooved฀for฀basal฀third,฀1-1.5฀mm฀ wide฀at฀widest฀point,฀more฀or฀less฀verrucose฀below,฀secondary฀veins฀8-17,฀often฀ 1-3฀intersecondary฀veins,฀distance฀between฀from฀c.฀5฀mm฀at฀the฀base฀to฀up฀to฀ 25(-30)฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀rather฀variable,฀from฀ 126 • Chapter 5 45-80°฀at฀the฀base฀to฀60-80°฀closer฀to฀the฀apex,฀forming฀distinct฀loops,฀smallest฀ distance฀between฀loops฀and฀margin฀2-6.5฀mm,฀tertiary฀veins฀percurrent.฀Inflorescence฀ of฀single,฀solitary฀flowers,฀on฀leafy฀twigs;฀peduncles฀1-4฀by฀c.฀1฀mm฀(in฀flower),฀2-5฀ by฀1.5-2฀mm฀(in฀fruit);฀pedicels฀(12-)15-25฀by฀1฀mm฀diam.฀at฀the฀base฀(in฀flower),฀ 14-30฀by฀1-1.5฀mm฀(in฀fruit),฀peduncles฀and฀pedicels฀rather฀densely฀covered฀with฀ more฀or฀less฀erect฀brown฀hairs฀0.2฀mm฀long;฀2฀lower฀bracts,฀deltate,฀c.฀1฀mm฀long,฀ caducous,฀ rather฀ densely฀ covered฀ with฀ more฀ or฀ less฀ erect฀ brown฀ hairs฀ 0.2฀ mm฀ long;฀ upper฀ bract฀ attached฀ around฀ midway฀ along฀ the฀ pedicel,฀ ovate฀ or฀ broadly฀ so,฀1-3฀by฀c.฀1฀mm,฀obtuse฀or฀acute,฀outer฀side฀sparsely฀to฀rather฀densely฀covered฀ with฀appressed฀or฀erect฀whitish฀hairs฀to฀0.2฀mm฀long;฀flower฀buds฀depressed฀ovoid;฀ flowers฀green฀to฀greenish฀yellow,฀pale฀yellow฀or฀cream฀in฀vivo,฀dark฀brown฀with฀ a฀lighter฀brown฀calyx฀in฀sicco,฀outer฀sides฀of฀sepals฀and฀petals฀sparsely฀to฀rather฀ densely฀covered฀with฀erect฀or฀appressed฀whitish฀hairs฀to฀0.2฀mm฀long,฀inner฀sides฀ glabrous฀to฀sparsely฀covered฀with฀appressed฀whitish฀hairs฀to฀0.2฀mm฀long฀(or฀inner฀ petals฀papillate);฀sepals฀free,฀broadly฀ovate฀to฀deltate,฀recurved,฀3-4฀by฀2.5-4฀mm,฀ obtuse,฀caducous;฀outer฀petals฀(broadly)฀elliptic฀to฀ovate,฀9-15฀by฀7-12฀mm,฀inner฀ petals฀elliptic,฀obovate,฀or฀narrowly฀so,฀8-16฀by฀4-7฀mm;฀androecium฀c.฀5฀mm฀diam.,฀ stamens฀1.2-1.5฀mm฀long,฀connective฀appendage฀roughly฀rhombic฀to฀diamond฀ shaped,฀0.7-0.8฀mm฀wide;฀gynoecium฀c.฀2฀mm฀diam.,฀carpels฀c.฀25,฀c.฀2.2฀mm฀long,฀ sparsely฀covered฀with฀erect฀golden฀hairs฀0.1฀mm฀long.฀Monocarps฀3-23,฀ellipsoid,฀ slightly฀asymmetrical,฀10-15฀by฀7-9฀mm,฀with฀an฀excentric฀apicule,฀green฀maturing฀ to฀pink฀or฀yellow฀through฀to฀red,฀purple฀and฀black฀in฀vivo,฀reddish฀or฀dark฀brown฀ in฀sicco,฀monocarps,฀stipes฀and฀receptacle฀sparsely฀to฀rather฀densely฀covered฀with฀ erect฀whitish฀hairs฀0.1฀mm฀long;฀stipes฀green฀maturing฀to฀pink฀or฀yellow฀to฀red฀in฀ vivo,฀7-17฀by฀1-1.5฀mm,฀increasing฀to฀3฀mm฀diam.฀when฀mature;฀fruiting฀receptacle฀ 3-8฀mm฀diam.฀Seeds฀ellipsoid,฀orange-brown,฀shallowly฀pitted,฀5-8฀by฀3.5-6฀mm,฀ raphe฀sunken,฀regular. Distribution฀-฀Amazonian฀Colombia,฀Ecuador,฀and฀Peru. Habitat฀ and฀ Ecology฀ -฀ Often฀ primary,฀ but฀ also฀ secondary฀ inundated฀ and฀ non-inundated฀forest.฀At฀elevations฀of฀100-500฀m.฀Flowering:฀January฀and฀April฀ -฀August;฀fruiting:฀throughout฀the฀year. Uses฀ -฀ infusions฀ of฀ the฀ leaves฀ are฀ reported฀ as฀ being฀ used฀ to฀ treat฀ stomach฀ pains. Vernacular฀names฀-฀Ecuador:฀Ansuelo฀caspi฀muyo,฀Ayacara,฀Daycabome. Notes฀-฀Cremastosperma฀gracilipes฀most฀closely฀resembles฀C.฀microcarpum.฀The฀ hairs฀ on฀ the฀ flowers฀ are฀ shorter฀ and฀ less฀ dense,฀ which฀ results฀ in฀ their฀ drying฀ a฀ darker฀brown.฀The฀leaves฀are฀further฀distinctive฀in฀the฀shape฀of฀the฀apex฀(markedly฀ cuspidate฀with฀an฀often฀long฀drip-tip)฀and฀in฀the฀green฀colour฀they฀consistently฀ retain฀on฀drying.฀The฀leaves฀of฀C.฀longicuspe฀are฀similar,฀but฀in฀contrast฀to฀C.฀gracilipes฀ both฀flowers฀and฀fruit฀are฀entirely฀glabrous. Revision and phylogeny of Cremastosperma • 127 10. Cremastosperma leiophyllum฀ R.E.Fr.฀-฀Fig.฀14;฀Map฀6;฀App.฀2,฀Fig.฀1a,฀b;฀5 Cremastosperma฀leiophyllum฀R.E.Fr.฀(1931)฀328.฀-฀Guatteria฀leiophylla฀Diels฀(1931)฀77,฀ non฀(Donn.฀Smith)฀Saff.฀-฀Type:฀Buchtien฀705฀(holo฀B;฀iso฀US),฀Bolivia,฀La฀Paz:฀Mapiri,฀ San฀Carlos,฀850฀m,฀2฀Dec.฀1926. Annona฀nitida฀Ruiz฀&฀Pav.฀in฀López฀(1959)฀429,฀t.฀488,฀non฀Martius฀(1841),฀nom.฀ nud. Guatteria฀rusbyi฀J.F.Macbr.฀(1929)฀171.฀-฀Guatteria฀lucida฀Rusby฀(1927)฀245,฀non฀C.฀ Presl.฀-฀Type:฀O.E.฀White฀913฀(holo฀NY),฀Bolivia,฀Beni:฀Covendo฀630฀m,฀26฀Aug.฀ 1921. Shrub฀or฀tree฀3-20฀m฀tall,฀3-18฀cm฀diam.;฀young฀twigs฀and฀petioles฀glabrous.฀ Leaves:฀petioles฀4-12฀by฀1-3(-4)฀mm;฀lamina฀(narrowly)฀obovate฀to฀narrowly฀elliptic,฀ 12-28฀by฀4-9(-12)฀cm฀(index฀2-3.9),฀chartaceous,฀often฀green฀or฀greenish฀brown,฀ greyish฀above฀with฀darker฀or฀reddish฀veins,฀glabrous฀on฀both฀sides,฀base฀acute฀to฀ obtuse,฀apex฀acuminate฀(acumen฀5-15฀mm฀long),฀primary฀vein฀1.5-3฀mm฀wide฀at฀ widest฀point,฀verrucose,฀secondary฀veins฀7-13,฀intersecondary฀veins฀1-6,฀distance฀ between฀from฀14-24฀mm฀at฀the฀base฀to฀10-25฀mm฀closer฀to฀the฀apex,฀angles฀with฀ primary฀vein฀from฀60-80°฀at฀the฀base฀to฀40-50°฀closer฀to฀the฀apex,฀not฀branching฀ forming฀distinct฀loops,฀smallest฀distance฀between฀loops฀and฀margin฀2-4(-6)฀mm,฀ tertiary฀veins฀percurrent.฀Inflorescence฀of฀single฀flowers,฀solitary฀or฀clustered฀in฀ groups฀of฀up฀to฀4,฀on฀older,฀leafless฀twigs;฀peduncles฀1-2฀by฀1-2฀mm฀(in฀flower),฀ 2-4฀by฀1.5-3฀mm฀(in฀fruit),฀sparsely฀covered฀with฀appressed฀golden฀<0.1฀mm฀long฀ hairs฀or฀glabrous;฀pedicels฀18-34฀by฀1-1.5฀mm฀at฀the฀base฀(in฀flower),฀18-34(-43)฀ by฀1-3฀mm฀(in฀fruit),฀glabrous;฀1-3฀lower฀bracts,฀depressed฀ovate,฀c.฀0.5฀by฀1฀mm,฀ obtuse,฀caducous,฀glabrous;฀upper฀bract฀attached฀around฀midway฀along฀pedicel,฀ ovate฀to฀broadly฀so,฀c.฀1.5฀by฀1฀mm,฀obtuse,฀glabrous;฀closed฀flower฀buds฀depressed฀ ovoid,฀opening฀early฀in฀development;฀flowers฀green฀maturing฀to฀yellow฀or฀creamy฀ yellow฀ in฀ vivo,฀ dark฀ yellow,฀ reddish฀ brown,฀ or฀ dark฀ brown฀ in฀ sicco;฀ sepals฀ free,฀ very฀broadly฀ovate-triangular,฀recurved,฀2-3฀by฀2-3฀mm,฀obtuse,฀caducous,฀sepals฀ and฀petals฀glabrous;฀outer฀petals฀elliptic,฀12-15฀by฀8-11฀mm,฀rounded,฀inner฀petals฀ elliptic,฀13-15฀by฀6-8฀mm;฀androecium฀c.฀7฀mm฀diam.,฀pinkish฀in฀vivo,฀stamens฀ 1.4-1.8฀mm฀long,฀connective฀appendage฀c.฀0.8฀mm฀wide;฀gynoecium฀c.฀2฀mm฀ diam.,฀carpels฀2-2.2฀mm฀long,฀glabrous.฀Monocarps,฀stipes,฀and฀receptacle฀glabrous,฀ monocarps฀6-30,฀ellipsoid,฀asymmetrical,฀14-17฀by฀8-9฀mm,฀green฀maturing฀to฀ yellow,฀orange-red฀and฀red฀in฀vivo,฀black฀(reddish฀brown฀when฀immature)฀in฀sicco,฀ with฀an฀excentric฀apicule;฀stipes฀16-26฀by฀1-1.5฀mm;฀fruiting฀receptacle฀4-9฀mm฀ diam.฀Seeds฀ellipsoid,฀light฀brown,฀pitted฀c.฀12฀by฀c.฀7฀mm,฀raphe฀sunken,฀regular. Distribution฀-฀Bolivia฀(Beni,฀Cochabamba,฀La฀Paz,฀Santa฀Cruz). Habitat฀and฀Ecology฀-฀Mostly฀in฀primary฀wet฀or฀moist฀forest,฀also฀in฀mildly฀ disturbed฀areas,฀often฀on฀slopes฀or฀terraces,฀on฀sandstone฀soils.฀At฀elevations฀of฀ 200-1000฀m.฀Flowering:฀February,฀May,฀July฀through฀September,฀November,฀and฀ December;฀fruiting:฀February฀through฀August,฀November,฀and฀December. 128 • Chapter 5 a b F i g . 1 4 . Cremastosperma leiophyllum R.E.Fr. a. fruiting specimen; b. flower (a & b: Seidel & Schulte 2265) Revision and phylogeny of Cremastosperma • 129 Notes฀-฀Cremastosperma฀leiophyllum฀is฀the฀most฀southerly฀distributed฀species฀of฀ the฀genus,฀and฀of฀the฀two฀found฀in฀Bolivia,฀the฀only฀endemic.฀Bud฀development฀ in฀C.฀leiophyllum฀is฀open฀(as฀opposed฀to฀that฀of฀C.฀monospermum).฀It฀most฀closely฀ resembles฀C.฀spec.฀B฀(which฀is฀not฀found฀in฀Bolivia),฀from฀which฀it฀can฀best฀be฀ distinguished฀ by฀ the฀ characteristic฀ shape฀ (asymmetrical,฀ the฀ stipes฀ thickening฀ somewhat฀where฀they฀meet฀the฀monocarps)฀and฀colour฀(blackish)฀of฀the฀mature฀ fruits฀when฀dried.฀ 11. Cremastosperma longicuspe฀ R.E.Fr.฀-฀Fig.฀15;฀Map฀5 Cremastosperma฀longicuspe฀R.E.Fr.฀(1934)฀203.฀-฀Type:฀Poeppig฀s.n.฀(lecto฀LE;฀iso฀S),฀Peru,฀ Loreto:฀Maynas,฀anno฀1831. Cremastosperma฀killipii฀R.E.Fr.฀(1948)฀3,฀P.฀I฀a-b.฀-฀Type:฀Killip฀&฀Smith฀29020฀(holo฀ US),฀Peru,฀Loreto:฀Yurimaguas,฀lower฀Río฀Huallaga,฀135฀m,฀Aug.-Sep.฀1929. Tree฀or฀shrub฀1.5-20฀m฀tall;฀young฀twigs฀and฀petioles฀sparsely฀covered฀with฀ appressed฀whitish฀or฀golden฀hairs฀to฀0.2฀mm฀long.฀Leaves:฀petioles฀4-14฀by฀1-3฀ mm;฀ lamina฀ elliptic,฀ obovate,฀ or฀ narrowly฀ so,฀ 10-27฀ by฀ 3-11฀ cm฀ (index฀ 1.5-5),฀ chartaceous,฀green฀or฀greyish฀green฀above,฀green฀or฀brownish฀green฀below,฀glabrous฀ above,฀very฀sparsely฀covered฀with฀appressed฀whitish฀hairs฀to฀0.2฀mm฀long฀particularly฀ on฀veins฀below,฀base฀acute฀(rarely฀obtuse),฀apex฀cuspidate฀(cusp฀20-35฀mm฀long),฀ primary฀vein฀1-2฀mm฀wide฀at฀widest฀point,฀secondary฀veins฀7-15,฀intersecondary฀ veins฀0-1(-4),฀distance฀between฀from฀7-15฀mm฀at฀the฀base฀to฀9-18฀mm฀closer฀to฀ the฀apex,฀angles฀with฀primary฀vein฀from฀60-80°฀at฀the฀base฀to฀40-50°฀closer฀to฀ the฀apex,฀rarely฀branching,฀forming฀distinct฀loops,฀smallest฀distance฀between฀loops฀ and฀margin฀2-4฀mm,฀tertiary฀veins฀more฀or฀less฀percurrent.฀Inflorescence฀of฀single฀ flowers฀solitary฀(or฀clustered฀in฀groups฀of฀2),฀on฀leafy฀twigs;฀peduncles฀2-5฀by฀c.฀1฀ mm฀(in฀flower),฀4-10฀by฀1-2฀mm฀(in฀fruit),฀sparsely฀to฀rather฀densely฀covered฀with฀ appressed฀golden฀hairs฀to฀0.2฀mm฀long;฀pedicels฀10-14฀by฀1-1.5฀mm฀at฀the฀base฀(in฀ flower),฀11-20฀by฀1-2฀mm฀(in฀fruit),฀sparsely฀covered฀with฀appressed฀golden฀hairs฀to฀ 0.2฀mm฀long฀or฀glabrous;฀2฀lower฀bracts฀of฀unequal฀dimensions,฀basal฀lower฀bract฀ deltate,฀c.฀0.5฀by฀0.5฀mm,฀acute,฀caducous,฀apical฀lower฀bract฀narrowly฀elliptic,฀c.฀1.5฀ by฀0.5฀mm,฀rounded,฀caducous,฀lower฀bracts฀sparsely฀covered฀with฀appressed฀golden฀ hairs฀to฀0.1฀mm฀long฀or฀glabrous;฀upper฀bract฀attached฀near฀base฀or฀midway฀along฀ pedicel,฀ovate,฀c.฀2฀by฀c.฀1฀mm,฀acute,฀sparsely฀covered฀with฀appressed฀golden฀hairs฀ to฀0.1฀mm฀long฀or฀glabrous;฀closed฀flower฀buds฀not฀seen;฀flowers฀yellowish฀in฀vivo,฀ brown฀in฀sicco;฀sepals฀free,฀deltate,฀appressed฀or฀recurved,฀3-4฀by฀3-4฀mm,฀acute,฀ caducous,฀sepals฀and฀petals฀glabrous;฀outer฀petals฀broadly฀elliptic,฀10-12฀by฀9-12฀ mm,฀inner฀petals฀broadly฀elliptic,฀c.฀11฀by฀10฀mm;฀androecium฀not฀seen;฀gynoecium฀ not฀seen.฀Monocarps,฀stipes,฀and฀receptacle฀glabrous,฀monocarps฀6-13(-36),฀ellipsoid,฀ slightly฀asymmetrical,฀12-13฀by฀8-10฀mm,฀white,฀red,฀deep฀red,฀or฀deep฀purple฀in฀ vivo,฀reddish฀brown฀to฀dark฀brown฀or฀black฀in฀sicco,฀with฀an฀excentric฀apicule฀ 130 • Chapter 5 b a F i g . 1 5 . Cremastosperma longicuspe R.E.Fr. a. fruiting specimen; b. flower (a: van der Werff 10161; b: Poeppig s.n.) Revision and phylogeny of Cremastosperma • 131 when฀unripe;฀stipes฀10-19฀by฀c.฀2฀mm;฀fruiting฀receptacle฀4-5(-9)฀mm฀diam.฀Seeds฀ ellipsoid,฀orange฀or฀reddish฀brown,฀8-9฀by฀5-6฀mm,฀raphe฀sunken,฀regular. Distribution฀-฀Peru฀(San฀Martín฀and฀Loreto),฀most฀collections฀found฀in฀the฀ basin฀of฀the฀Río฀Huallaga. Habitat฀and฀Ecology฀-฀Primary฀tropical฀wet฀forest฀on฀sandy฀or฀white฀sand฀soil.฀ At฀elevations฀of฀140-200฀m.฀Flowering:฀August-October;฀fruiting:฀February-June,฀ August,฀September,฀and฀November. Notes฀-฀Cremastosperma฀longicuspe฀most฀closely฀resembles฀C.฀gracilipes,฀particularly฀ in฀the฀shape฀of฀the฀leaf.฀However,฀in฀contrast฀to฀C.฀gracilipes,฀the฀fruits฀and฀flowers฀ are฀glabrous.฀In฀describing฀C.฀killipii,฀Fries฀(1948)฀noted฀its฀similarity฀to฀C.฀longicuspe.฀ The฀leaves฀of฀the฀type฀specimen฀of฀C.฀killipii฀are฀unusually฀broad,฀but฀leaf฀and฀other฀ characters฀otherwise฀fall฀within฀the฀variation฀found฀in฀C.฀longicuspe,฀including฀the฀ notable฀cuspidate฀apex.฀ 12. Cremastosperma longipes฀Pirie฀-฀Fig.฀16;฀Map฀2 Cremastosperma฀longipes฀Pirie฀(2005)฀51,฀f.฀5฀-฀Type:฀Forero฀et฀al.฀6576฀(holo฀COL;฀iso฀ MO),฀Colombia,฀Chocó:฀San฀José฀del฀Palmar,฀mouth฀of฀Río฀Torito฀(tributary฀of฀Río฀ Hábita),฀west฀slope,฀3฀Mar.฀1980. Tree฀4.5-15฀m฀tall;฀young฀twigs฀and฀petioles฀black,฀verrucose,฀sparsely฀to฀rather฀ densely฀covered฀with฀white-golden฀appressed฀hairs฀c.฀0.4฀mm฀long.฀Leaves:฀petioles฀ 10-15฀mm฀long,฀2.5-7฀mm฀diam.;฀lamina฀elliptic฀to฀narrowly฀so,฀35-60฀by฀10-25฀ cm฀(leaf฀index฀2.3-3),฀chartacous฀to฀subcoriacous,฀olive/dark฀brown฀above,฀lighter฀ below,฀glabrous฀above,฀sparsely฀covered฀with฀white-golden฀appressed฀hairs฀c.฀0.4฀ mm฀long฀on฀veins฀below฀(densely฀so฀developing฀leaves),฀base฀acute,฀apex฀acuminate฀ (acumen฀10-15฀mm฀long),฀primary฀vein฀deeply฀grooved฀in฀basal฀half,฀2-6฀mm฀wide฀ at฀widest฀point;฀secondary฀veins฀10-16,฀intersecondary฀veins฀rare,฀distance฀between฀ from฀10฀mm฀at฀the฀base฀to฀80฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀ 45°฀at฀the฀base฀to฀60-70°฀closer฀to฀the฀apex,฀not฀branching,฀forming฀distinct฀loops฀ in฀the฀apical฀half฀to฀third฀of฀the฀leaf,฀smallest฀distance฀between฀loops฀and฀margin฀ 3-4฀mm;฀tertiary฀veins฀mainly฀percurrent.฀Inflorescences฀of฀single,฀pendulous฀flowers,฀ produced฀from฀leafless฀branches;฀peduncles฀5-8฀mm฀long,฀c.฀1฀mm฀diam.฀(in฀flower),฀ c.฀4฀mm฀long,฀2฀mm฀diam.฀(in฀fruit);฀pedicels฀90฀(less฀mature)฀-฀210฀mm฀long,฀c.฀ 1฀mm฀diam.฀at฀the฀base,฀1.5฀mm฀diam.฀at฀the฀apex฀(in฀flower),฀c.฀240฀mm฀long,฀2฀ mm฀diam.฀at฀the฀base,฀3฀mm฀diam.฀at฀the฀apex฀(in฀fruit);฀peduncles฀and฀pedicels฀ sparsely฀to฀rather฀densely฀covered฀with฀white-golden฀appressed฀hairs฀c.฀0.4฀mm฀ long;฀single฀lower฀bract,฀broadly฀elliptic,฀1-2฀by฀c.฀1฀mm,฀acute,฀caducous,฀densely฀ covered฀with฀white-golden฀appressed฀hairs฀c.฀0.4฀mm฀long;฀upper฀bract฀attached฀on฀ lower฀half฀of฀pedicel,฀elliptic,฀1.5-3฀by฀c.฀1฀mm,฀acute,฀densely฀covered฀with฀whitegolden฀appressed฀hairs฀c.฀0.4฀mm฀long;฀closed฀flower฀buds฀not฀seen;฀flowers฀green฀ (immature)฀in฀vivo,฀medium฀brown฀in฀sicco;฀sepals฀free,฀triangular฀to฀broadly฀trullate,฀ 132 • Chapter 5 F i g . 1 6 . Cremastosperma longipes Pirie. a. monocarp (with stipe) and seeds; b. leaf, flower, and flower bud (a: Fernández 8872; b: Forero et al. 6576) 3-4.5฀mm฀long,฀appressed,฀acute,฀caducous,฀rather฀densely฀to฀densely฀covered฀with฀ white-golden฀appressed฀hairs฀c.฀0.4฀mm฀long;฀outer฀petals฀elliptic,฀c.฀22฀by฀12฀mm,฀ inner฀petals฀narrowly฀elliptic,฀c.฀22฀by฀6฀mm,฀sparsely฀to฀rather฀densely฀covered฀with฀ white-golden฀appressed฀hairs฀c.฀0.4฀mm฀long;฀stamens,฀c.฀1.2฀mm฀long,฀connective฀ appendage฀rhombic,฀1฀mm฀wide,฀glabrous.฀Monocarps฀c.฀20,฀black฀in฀sicco,฀ellipsoid,฀ slightly฀asymmetrical,฀c.฀20฀mm฀long,฀12฀mm฀diam.,฀without฀an฀apicule,฀glabrous;฀ stipes฀c.฀25฀mm฀long,฀2฀mm฀diam.,฀glabrous;฀fruiting฀receptacle฀ovoid,฀8฀mm฀in฀ Revision and phylogeny of Cremastosperma • 133 diam,฀sparsely฀covered฀with฀white-golden฀appressed฀hairs฀c.฀0.4฀mm฀long.฀Seeds฀ ellipsoid,฀18-20฀mm฀long,฀8-9฀mm฀diam.,฀orange/brown,฀with฀many฀shallow฀pits,฀ raphe฀slightly฀raised,฀encircling฀seed฀longitudinally. Distribution฀-฀Pacific฀coast฀of฀Ecuador฀(Esmeraldas)฀and฀Colombia฀(Chocó฀ and฀Riseralda). Habitat฀and฀Ecology฀-฀Humid฀lowland฀to฀premontane฀forest.฀At฀elevations฀of฀ 280-1400฀m.฀Flowering:฀January฀and฀March;฀fruiting:฀September. Notes฀-฀Cremastosperma฀longipes฀can฀easily฀be฀distinguished฀from฀other฀species฀ of฀the฀genus฀by฀the฀exceptional฀length฀of฀the฀pedicel,฀after฀which฀the฀species฀is฀ named.฀The฀ flowers฀ and฀ fruits฀ of฀ most฀ species฀ of฀ Cremastosperma฀ are฀ borne฀ on฀ pedicels฀less฀than฀50฀mm฀long,฀with฀rare฀exceptions฀such฀as฀C.฀pedunculatum฀and฀ C.฀bullatum฀never฀exceeding฀150฀mm฀in฀length,฀significantly฀shorter฀than฀those฀of฀฀ C.฀ longipes.฀ In฀ addition,฀ leaves฀ of฀ C.฀ longipes฀ are฀ unusually฀ large,฀ equalling฀ the฀ maximum฀ dimensions฀ observed฀ in฀ C.฀ megalophyllum,฀ a฀ more฀ densely฀ collected฀ species฀from฀Amazonian฀Colombia,฀Ecuador,฀and฀Peru. 13. Cremastosperma macrocarpum฀ Maas฀-฀Fig.฀2฀d,฀17;฀Map฀9 Cremastosperma฀macrocarpum฀Maas฀in฀Maas฀et฀al.฀(1986)฀253,฀f.฀2,฀3b.฀-฀Type:฀van฀der฀Werff฀ &฀Vera,฀Flora฀Falcón฀937฀(holo฀U),฀Venezuela,฀Falcón:฀Sierra฀de฀San฀Luis,฀above฀Santa฀ María,฀1300฀m,฀26฀Jul.฀1979. Tree฀5-10฀m฀tall;฀young฀twigs฀and฀petioles฀glabrous.฀Leaves:฀petioles฀3-10฀by฀ 1-3฀ mm;฀ lamina฀ narrowly฀ elliptic฀ to฀ elliptic,฀ 10-30฀ by฀ 4-9(12)฀ cm฀ (index฀ 2.33.7),฀chartaceous,฀green,฀brownish฀green฀or฀brown฀on฀both฀sides,฀darker฀above,฀ glabrous฀on฀both฀sides,฀base฀obtuse฀to฀rounded,฀apex฀acuminate฀(acumen฀5-10฀mm฀ long),฀primary฀vein฀grooved฀in฀basal฀half,฀1-2฀mm฀wide฀at฀widest฀point,฀verrucose,฀ secondary฀veins฀6-12,฀intersecondary฀veins฀1-2,฀distance฀between฀from฀7-15฀mm฀ at฀the฀base฀to฀7-23฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀from฀60-80°฀ at฀the฀base฀to฀45-60°฀closer฀to฀the฀apex,฀not฀branching,฀forming฀mostly฀distinct฀ loops,฀smallest฀distance฀between฀loops฀and฀margin฀1-5฀mm,฀tertiary฀veins฀mostly฀ percurrent.฀Inflorescence฀of฀single,฀solitary฀flowers,฀on฀leafy฀twigs;฀peduncles฀1฀by฀0.51฀mm฀(in฀flower),฀1-2฀by฀1-2฀mm฀(in฀fruit),฀sparsely฀covered฀in฀appressed฀golden฀ hairs฀to฀0.1฀mm฀long฀or฀glabrous;฀pedicels฀35-45฀by฀0.5-1฀mm฀at฀the฀base฀up฀to฀ 2.5฀mm฀diam.฀at฀the฀apex฀(in฀flower),฀40-65฀by฀1-1.5฀mm฀at฀the฀base฀up฀to฀5฀mm฀ diam.฀at฀the฀apex฀(in฀fruit),฀glabrous;฀single฀lower฀bract,฀depressed฀ovate,฀0.5฀by฀1฀ mm,฀rounded,฀caducous,฀rather฀densely฀covered฀with฀appressed฀golden฀hairs฀to฀0.1฀ mm฀long;฀upper฀bract฀at฀or฀near฀base฀of฀pedicel,฀ovate,฀c.฀1.5฀by฀0.8฀mm,฀obtuse,฀ glabrous;฀ closed฀ flower฀ buds฀ not฀ seen;฀ flowers฀ pale฀ greenish฀ yellow฀ with฀ green฀ base฀or฀cream-coloured฀in฀vivo,฀black฀in฀sicco,฀sepals฀and฀petals฀glabrous;฀sepals฀ free,฀broadly฀to฀depressed฀ovate,฀recurved,฀1.5-2฀by฀1.5-2฀mm,฀obtuse,฀caducous;฀ 134 • Chapter 5 a b F i g . 1 7 . Cremastosperma macrocarpum Maas. a. fruiting specimen; b. flower (a & b: van der Werff & Vera, Flora Falcón 937) outer฀petals฀ovate,฀11-14฀by฀8-9฀mm,฀obtuse,฀inner฀petals฀obovate,฀15-16฀by฀c.฀7฀ mm,฀obtuse;฀androecium฀c.฀6.5฀mm฀diam.,฀gynoecium฀not฀seen.฀Monocarps,฀stipes,฀ and฀receptacle฀glabrous,฀monocarps฀7-18,฀ellipsoid,฀slightly฀asymmetrical,฀18-24฀ by฀12-14฀mm,฀very฀small฀strongly฀excentric฀apicule,฀green฀maturing฀to฀yellowish฀ brown,฀brown฀or฀purple-black฀in฀vivo,฀reddish฀brown฀or฀dark฀brown฀in฀sicco;฀stipes฀ 7-14฀by฀c.฀2฀mm;฀fruiting฀receptacle฀5-8฀mm฀diam.฀Seeds฀ellipsoid,฀orange-brown,฀ shallowly฀pitted,฀c.฀20฀by฀8฀mm,฀raphe฀sunken,฀regular. Revision and phylogeny of Cremastosperma • 135 Distribution฀-฀Venezuela฀(Falcón฀and฀Yaracuy).฀ Habitat฀ and฀ Ecology฀ -฀ Primary฀ or฀ secondary฀ evergreen฀ cloud฀ forest.฀ At฀ elevations฀of฀700-1500฀m.฀Flowering:฀July;฀fruiting:฀March,฀May฀-฀July,฀October,฀ and฀December. Notes฀-฀One฀of฀only฀two฀species฀of฀Cremastosperma฀found฀in฀Venezuela,฀C.฀ macrocarpum฀can฀most฀easily฀be฀distinguished฀from฀C.฀venezuelanum฀by฀its฀smaller฀ leaves฀(10-30฀cm฀as฀opposed฀to฀30-53฀cm฀long)฀and฀longer฀pedicels฀(40-65฀mm฀as฀ opposed฀to฀16-22฀mm฀in฀fruit).฀ 14. Cremastosperma magdalenae฀ Pirie฀-฀Fig.฀18;฀Map฀3 Cremastosperma฀magdalenae฀Pirie฀(2005)฀53,฀f.฀6.฀-฀Type:฀Escobar฀&฀Folsom฀3309฀(holo฀ NY,฀2฀sheets;฀iso฀HUA,฀U),฀Colombia,฀Antioquia:฀San฀Luis,฀Medellín-Bogotá฀highway,฀ 8.1฀km฀E฀of฀bridge฀over฀Río฀Caldera,฀980-1020฀m,฀13฀Mar.฀1983. Tree฀3-7฀m฀tall;฀young฀twigs฀and฀petioles฀slightly฀canaliculate฀and/or฀verrucose,฀ glabrous.฀Leaves:฀petioles฀6-14฀mm฀long,฀2-3฀mm฀diam.;฀lamina฀narrowly฀elliptic,฀ 20-28฀by฀7-9฀cm฀(index฀2.5-3.1),฀chartaceous฀to฀subcoriaceous,฀olive฀to฀more฀lime฀ green฀or฀brown฀above,฀darker฀below,฀glabrous฀on฀both฀sides,฀base฀obtuse฀to฀acute,฀ apex฀acuminate฀(acumen฀5-10฀mm฀long),฀primary฀vein฀deeply฀grooved฀in฀basal฀ 1/2฀-฀3/4,฀occasionally฀verrucose฀below,฀c.฀2฀mm฀wide฀at฀widest฀point,฀glabrous,฀ secondary฀veins฀9-14,฀often฀2฀or฀3฀intersecondary฀veins,฀distance฀between฀from฀4-5฀ mm฀at฀the฀base฀to฀20-35฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀from฀4050°฀at฀the฀base฀to฀70-80°฀closer฀to฀the฀apex,฀occasionally฀branching,฀occasionally฀ forming฀more฀or฀less฀indistinct฀loops฀in฀the฀apical฀half,฀smallest฀distance฀between฀ loops฀and฀margin฀2-4฀mm,฀tertiary฀veins฀rather฀reticulate.฀Inflorescences฀of฀single฀ flowers,฀solitary฀or฀clustered฀in฀groups฀of฀two,฀axillary฀on฀leafy฀twigs;฀peduncles฀2-3฀ mm฀long,฀c.฀2฀mm฀diam.฀(in฀fruit);฀pedicels฀c.฀7฀mm฀long,฀1฀mm฀diam.฀at฀the฀base,฀ 1.5฀mm฀diam.฀at฀the฀apex฀(in฀flower),฀16-20฀mm฀long,฀1.5-2฀mm฀diam.฀at฀the฀base,฀ c.฀3฀mm฀diam.฀at฀the฀apex฀(in฀fruit),฀peduncles฀and฀pedicels฀glabrous,฀two฀lower฀ bracts,฀the฀apical฀one฀depressed฀triangular,฀c.฀1.5฀by฀2฀mm,฀acute,฀persistent,฀glabrous,฀ upper฀bract฀attached฀around฀midway฀along฀pedicel,฀broadly฀ovate฀to฀deltate,฀0.62.5฀by฀0.8-2.5฀mm฀wide,฀actute฀to฀obtuse,฀glabrous;฀closed฀flower฀buds฀not฀seen,฀ flowers฀in฀vivo฀immature฀light฀green,฀black฀in฀sicco,฀sepals฀and฀petals฀glabrous;฀ sepals฀fused฀for฀basal฀1฀mm,฀broadly฀to฀very฀broadly฀ovate,฀appressed,฀(2-)5-7฀by฀ (2-)฀c.฀5฀mm,฀acute,฀mostly฀persistent,฀outer฀petals฀elliptic,฀c.฀12฀by฀7฀mm,฀inner฀ petals฀narrowly฀elliptic,฀c.฀12฀by฀5฀mm;฀androecium฀2.5-2.7฀mm฀diam.,฀stamens฀c.฀ 0.7฀mm฀long,฀connective฀appendage฀rhombic,฀0.3-0.4฀mm฀wide.฀Monocarps฀20-30,฀ globose,฀symmetrical,฀12-13฀mm฀long,฀12-13฀mm฀diam.,฀green฀maturing฀to฀red฀in฀ vivo,฀black฀in฀sicco,฀glabrous,฀with฀a฀slightly฀excentral,฀0.25฀mm฀long฀apicule;฀stipes฀ (immature)฀9-10฀mm฀long,฀1-1.5฀mm฀diam.,฀glabrous;฀fruiting฀receptacle฀depressed฀ 136 • Chapter 5 F i g . 1 8 . Cremastosperma magdalenae Pirie. a. flower; b. fruiting twig and cross section of seed (a: Hernández 251; b: Escobar & Folsom 3309) ovoid,฀7-9฀mm฀diam.฀(only฀immature฀fruits฀seen).฀Seeds฀globose,฀shallowly฀pitted฀ with฀a฀papery฀outer฀layer,฀c.฀13฀by฀11฀mm,฀orange-brown,฀raphe฀neither฀raised฀nor฀ sunken,฀regular. Distribution฀-฀Colombia฀(Antioquia),฀west฀side฀of฀the฀Magdalena฀valley.฀ Habitat฀and฀Ecology฀-฀Disturbed฀primary฀or฀secondary฀forest.฀At฀elevations฀of฀ 670-1200฀m.฀Flowering฀and฀fruiting:฀March. Note฀ -฀ Cremastosperma฀ magdalenae฀ Pirie฀ can฀ be฀ distinguished฀ from฀ other฀ species฀of฀the฀genus฀by฀the฀combination฀of฀globose฀monocarps฀and฀large฀sepals฀ Revision and phylogeny of Cremastosperma • 137 which฀ mostly฀ persist฀ into฀ fruiting฀ (one฀ slightly฀ differing฀ collection,฀ Cárdenas฀ 2899,฀displays฀immature฀fuits฀with฀smaller฀sepals฀only฀persistent฀on฀one฀of฀the฀ two฀duplicates฀studied).฀Also฀noteworthy฀are฀the฀relatively฀short฀pedicels฀and฀the฀ absence฀of฀indument฀on฀all฀parts.฀The฀absence฀of฀hairs฀on฀fruits฀and฀flowers฀reveals฀ the฀blackish฀colour฀typical฀of฀specimens฀of฀Cremastosperma฀upon฀drying.฀Both฀C.฀ panamense฀Maas฀and฀C.฀pacificum฀R.E.Fr.฀(a฀species฀found฀on฀the฀Pacific฀coast฀of฀ Colombia)฀also฀lack฀indument,฀but,฀amongst฀other฀differences,฀the฀sepals฀of฀both฀ species฀are฀much฀smaller฀and฀do฀not฀persist฀into฀fruiting. 15. Cremastosperma megalophyllum฀ R.E.Fr.฀-฀Fig.฀2฀a,฀19;฀Map฀7;฀App.฀2,฀Fig.฀6 Cremastosperma฀megalophyllum฀R.E.Fr.฀(1931)฀329.฀-฀Type:฀Ducke฀RB19620฀(holo฀S;฀iso฀ B,฀RB),฀Peru,฀Loreto:฀Boca฀de฀Pebas,฀23฀Oct.฀1927. Tree฀ or฀ shrub฀ 3-15฀ m฀ tall,฀ 2-15(-35)฀ cm฀ diam.;฀ young฀ twigs฀ and฀ petioles฀ glabrous.฀Leaves:฀petioles฀5-22฀by฀1.5-9฀mm;฀lamina฀obovate฀to฀elliptic,฀or฀narrowly฀ so,฀13-57฀by฀3-26฀cm฀(index฀2-5.1),฀chartaceous฀to฀coriaceous,฀shiny฀on฀both฀sides,฀ secondary฀veins฀often฀impressed฀above฀(giving฀slightly฀bullate฀appearance),฀(dark)฀ greyish฀green฀or฀brown฀above,฀more฀brown฀or฀green฀below,฀glabrous฀on฀both฀sides,฀ base฀obtuse฀to฀rounded,฀rarely฀acute฀or฀decurrent,฀apex฀acuminate฀(acumen฀(5-)1030฀mm฀long),฀primary฀vein฀conspicuously฀grooved฀in฀basal฀half,฀1-5฀mm฀wide฀at฀ widest฀point,฀secondary฀veins฀(5-)8-21,฀intersecondary฀veins฀often฀1-2(-3),฀distance฀ between฀from฀5-25฀mm฀at฀the฀base฀to฀9-35(-43)฀mm฀closer฀to฀the฀apex,฀angles฀ with฀primary฀vein฀from฀(30-)45-70(-80)°฀at฀the฀base฀to฀(30-)40-75°฀closer฀to฀the฀ apex,฀rarely฀branching,฀forming฀distinct฀loops,฀smallest฀distance฀between฀loops฀and฀ margin฀1-5฀mm,฀tertiary฀veins฀percurrent.฀Inflorescence฀of฀single฀flowers,฀solitary฀ or฀clustered฀in฀groups฀of฀up฀to฀3,฀on฀leafy฀or฀leafless฀twigs฀or฀on฀the฀main฀trunk;฀ short฀axillary฀shoot,฀4-7(-8)฀by฀0.7-1฀mm฀(in฀flower),฀4-10฀by฀1.5-4฀mm฀(in฀fruit),฀ glabrous฀or฀sparsely฀covered฀with฀whitish฀appressed฀hairs฀to฀0.1฀mm฀long;฀pedicels฀ 10-20(-32)฀by฀c.฀1฀mm฀at฀the฀base,฀up฀to฀2(-2.5)฀mm฀diam.฀at฀the฀apex฀(in฀flower),฀ 15-30(-40)฀ by฀ 1.5-3฀ mm฀ at฀ the฀ base,฀ up฀ to฀ 4฀ mm฀ diam.฀ at฀ the฀ apex฀ (in฀ fruit),฀ glabrous;฀3฀lower฀bracts,฀deltate,฀c.฀1฀by฀1฀mm,฀acute,฀caducous,฀sparsely฀covered฀ with฀whitish฀appressed฀hairs฀to฀0.1฀mm฀long฀or฀glabrous;฀upper฀bract฀attached฀in฀ apical฀half฀of฀pedicel,฀broadly฀ovate,฀1-3฀by฀1-2฀mm,฀obtuse,฀glabrous;฀closed฀flower฀ buds฀depressed฀ovoid,฀opening฀loosely฀in฀development;฀flowers฀green฀maturing฀to฀ yellow฀in฀vivo,฀black฀in฀sicco,฀sepals฀and฀petals฀glabrous;฀sepals฀free฀or฀connate฀for฀1฀ mm,฀broadly฀ovate฀to฀triangular,฀appressed,฀open฀and฀conspicuous฀whilst฀petals฀still฀ closed฀in฀young฀buds,฀4-6฀by฀4-6฀mm,฀obtuse฀to฀acute,฀sometimes฀briefly฀or฀partly฀ persistent;฀outer฀petals฀broadly฀elliptic,฀11-18฀by฀9-15฀mm,฀obtuse,฀inner฀petals฀ obovate,฀ 10-16฀ by฀ 5-7฀ mm,฀ obtuse;฀ androecium฀ diam.฀ unknown,฀ stamens฀ 1.31.8฀mm฀long,฀connective฀appendage฀0.7-1฀mm฀wide;฀gynoecium฀diam.฀unknown,฀ 138 • Chapter 5 a b F i g . 1 9 . Cremastosperma megalophyllum R.E.Fr. a. fruiting specimen; b. flower buds (a: Zuleta 175; b: Palacios 3270) carpels฀1.5-2฀mm฀long.฀Monocarps,฀stipes,฀and฀receptacle฀glabrous,฀monocarps฀6-32,฀ ellipsoid฀to฀broadly฀ellipsoid,฀(slightly)฀asymmetrical,฀12-20฀by฀9-14฀mm,฀green฀ maturing฀to฀yellow,฀orange,฀purple฀and฀black฀in฀vivo,฀reddish฀or฀dark฀brown฀or฀ Revision and phylogeny of Cremastosperma • 139 black฀in฀sicco,฀often฀with฀an฀excentric฀apicule;฀stipes฀10-30฀by฀1-2฀mm;฀fruiting฀ receptacle฀3-12฀mm฀diam.฀Seeds฀broadly฀ellipsoid,฀reddish-orange,฀pitted,฀c.฀12฀by฀ 9-10฀mm,฀raphe฀raised฀(more฀so฀when฀seeds฀immature),฀regular. Distribution฀-฀Amazonian฀Peru,฀Ecuador,฀and฀Colombia. Habitat฀ and฀ Ecology฀ -฀ Primary฀ and฀ secondary฀ premontane฀ or฀ lowland฀ rainforest,฀sometimes฀inundated,฀on฀red฀(oxisols/lateritic)฀or฀sometimes฀volcanic฀ soils.฀At฀ elevations฀ of฀ 100-1200฀ m.฀ Flowering:฀April-June,฀ September-January;฀ fruiting:฀throughout฀the฀year. Vernacular฀ names฀ -฀ Ecuador:฀ Caramoyu;฀ Oñetahue฀ (the฀ plant฀ itself),฀ Oñetahuemo฀(the฀fruit)฀(Huaorani);฀Anchu฀panga฀cara฀caspi฀(wide-leaved฀bark฀ wood)฀,฀Ayacara,฀Cucha฀casa฀caspi,฀Ichilla฀cara฀caspi฀(small฀bark฀tree),฀Lynshtimoia,฀ Tara฀caspi฀or฀T’zinytala฀(Quechua);฀Mandachi฀(Shuar).฀Peru:฀Bara. Notes฀-฀Despite฀its฀name,฀the฀size฀of฀leaves฀of฀Cremastosperma฀megalophyllum฀฀ varies฀from฀large฀to฀relatively฀small฀with฀respect฀to฀those฀of฀other฀species฀of฀the฀ genus.฀This฀variation฀is฀also฀apparent฀in฀the฀size฀of฀the฀fruits,฀which฀are฀similar฀ to฀those฀of฀C.฀napoense,฀but฀which฀in฀contrast฀are฀never฀borne฀on฀a฀branching฀ inflorescence.฀C.฀megalophyllum฀is฀best฀distinguished฀by฀the฀shape฀of฀the฀glabrous,฀ black-drying฀flowers:฀the฀large฀sepals฀open฀earlier,฀and฀to฀a฀greater฀extent,฀than฀the฀ petals฀(in฀contrast฀to฀those฀of฀C.฀napoense,฀bud฀development฀of฀which฀is฀open฀from฀ an฀early฀stage฀and฀the฀lighter฀colour฀(particularly฀of฀the฀pedicel)฀of฀which฀indicates฀ the฀presence฀of฀indument). 16. Cremastosperma microcarpum฀ ฀ R.E.Fr.฀ -฀ Fig.฀ 1฀ a,฀ g,฀ 2฀ c,฀ 20;฀ Map฀ 7;฀฀ App.฀2,฀Fig.฀7 Cremastosperma฀microcarpum฀R.E.Fr.฀(1939)฀559.฀-฀Type:฀Krukoff฀6151฀(holo฀S;฀iso฀A,฀F,฀ G,฀GB,฀K,฀MICH,฀MO,฀NY,฀U,฀US),฀Brazil,฀Amazonas:฀Mun.฀Humaitá,฀Rio฀Madeira,฀ near฀Tres฀Casas,฀18฀Sep.฀1934. Tree฀or฀shrub฀2-20฀m฀tall,฀2.5-18.5฀cm฀diam.;฀young฀twigs฀and฀petioles฀glabrous฀ to฀rather฀densely฀covered฀with฀appressed฀and/or฀erect฀white฀or฀golden฀hairs฀0.30.4฀mm฀long.฀Leaves:฀petioles฀2-12฀by฀1.5-3฀mm;฀lamina฀narrowly฀elliptic฀(narrowly฀ obovate),฀ 13-31฀ by฀ 4-12฀ cm฀ (index฀ 1.8-3.8(-4.8)),฀ chartaceous,฀ green,฀ greyish฀ or฀brownish฀green฀or฀brown฀on฀both฀sides,฀shiny฀above,฀veination฀below฀often฀ yellowish,฀glabrous฀above,฀glabrous฀to฀rather฀densely฀covered฀with฀appressed฀or฀ erect฀white฀hairs฀to฀0.2฀mm฀long฀at฀the฀base฀and฀on฀primary฀and฀secondary฀veins฀ below,฀base฀acute฀to฀obtuse,฀rarely฀rounded฀or฀narrowly฀cuneate,฀apex฀acuminate฀ (acumen฀10-40฀mm฀long),฀primary฀vein฀1-2฀mm฀wide฀at฀widest฀point,฀more฀or฀less฀ verrucose฀on฀both฀sides,฀lightly฀grooved฀for฀around฀half฀of฀length,฀secondary฀veins฀ 7-15,฀often฀1-4฀intersecondary฀veins,฀distance฀between฀from฀c.฀5฀mm฀at฀the฀base฀to฀ c.฀20฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀mostly฀from฀45-60°฀at฀the฀ base฀to฀60-80°฀closer฀to฀the฀apex,฀not฀branching,฀forming฀distinct฀loops,฀smallest฀ 140 • Chapter 5 c a b F i g . 2 0 . Cremastosperma microcarpum R.E.Fr. a. leaves; b. flowers and flower buds; c. fruit (a & b: Maas et al. 6281; c: Vásquez & Jaramillo 9350) distance฀between฀loops฀and฀margin฀2-7฀mm,฀tertiary฀veins฀percurrent.฀Inflorescence฀ of฀single฀flowers,฀solitary฀or฀clustered฀in฀groups฀of฀up฀to฀3,฀on฀leafy฀or฀leafless฀twigs;฀ peduncles฀(3-)5-15฀by฀c.฀1฀mm฀(in฀flower),฀4-15฀by฀1-2฀mm฀(in฀fruit);฀pedicels฀(5-)฀ 12-24฀by฀c.฀1฀mm฀at฀the฀base฀(in฀flower),฀10-25฀by฀1-2฀mm฀(in฀fruit),฀peduncles฀and฀ pedicels฀sparsely฀to฀rather฀densely฀covered฀with฀appressed฀or฀erect฀whitish฀hairs฀to฀ 0.3฀mm฀long;฀1฀to฀several฀lower฀bract(s),฀the฀basal-most฀small฀and฀scale-like,฀those฀ more฀apical฀mostly฀(long)฀elliptic,฀occasionally฀leaf-like,฀2-6(-60)฀by฀c.1฀mm,฀acute,฀ Revision and phylogeny of Cremastosperma • 141 caducous,฀rather฀densely฀covered฀with฀appressed฀white฀hairs฀to฀0.3฀mm฀long;฀upper฀ bract฀attached฀mostly฀on฀the฀basal฀half฀of฀the฀pedicel,฀ovate฀to฀deltate,฀1.5-2.5฀by฀11.5฀mm,฀obtuse฀or฀acute,฀sparsely฀to฀rather฀densely฀covered฀with฀appressed฀or฀erect฀ golden฀hairs฀to฀0.3฀mm฀long;฀closed฀flower฀buds฀very฀broadly฀to฀depressed฀ovoid,฀ opening฀loosely฀in฀development;฀flowers฀green,฀maturing฀to฀brown,฀(pale)฀yellow,฀ cream฀or฀white฀outside,฀cream฀or฀yellow฀inside,฀calyx฀green฀or฀dark฀brown฀outside,฀ green฀with฀a฀pink฀base฀inside฀in฀vivo,฀pale฀(orange-)฀brown฀or฀brown฀with฀dark฀ or฀reddish฀brown฀base฀in฀sicco,฀sepals฀and฀petals฀rather฀densely฀to฀densely฀covered฀ with฀appressed฀or฀erect฀golden฀hairs฀(whitish฀close฀to฀the฀edges)฀to฀0.3฀mm฀long;฀ sepals฀free฀or฀connate฀for฀0.5฀mm,฀broadly฀ovate฀to฀deltate,฀not฀reflexed,฀3-4{-6}฀by฀ 2.5-4{-6}฀mm,฀obtuse,฀caducous;฀outer฀petals฀ovate฀to฀very฀broadly฀ovate,฀rounded,฀ 11-18{-19}฀by฀10-17฀mm,฀inner฀petals฀elliptic฀to฀(narrowly)฀obovate,฀obtuse,฀1016{-22}฀by฀5-8{-10}฀mm;฀androecium฀c.฀7฀mm฀diam.,฀stamens฀1.3-1.5฀mm฀long,฀ connective฀ appendage฀ 0.6-0.8฀ mm฀ wide,฀ glabrous;฀ gynoecium฀ c.฀ 1฀ mm฀ diam.,฀ carpels฀2-2.5{-2.9}฀long,฀sparsely฀covered฀with฀erect฀whitish฀hairs฀to฀0.1฀mm฀long.฀ Monocarps฀(8-)17-33,฀ellipsoid฀to฀broadly฀ellipsoid,฀asymmetrical,฀8-11฀by฀6-8฀mm,฀ often฀with฀an฀oblique฀longitudinal฀groove฀corresponding฀to฀the฀seed฀raphe,฀green฀ maturing฀to฀pink฀or฀orange฀through฀purple฀or฀brownish฀red,฀brown฀and฀black฀ in฀ vivo,฀ dark฀ or฀ reddish฀ brown฀ in฀ sicco,฀ with฀ an฀ excentric฀ apicule,฀ monocarps,฀ stipes฀and฀receptacle฀sparsely฀to฀rather฀densely฀covered฀with฀erect฀whitish฀hairs฀ c.฀0.1฀mm฀long;฀stipes฀8-16฀by฀c.฀1฀mm;฀fruiting฀receptacle฀4-8฀mm฀diam.฀Seeds฀ broadly฀ellipsoid,฀orange฀brown,฀pitted,฀6-8฀by฀5-6฀mm,฀raphe฀sunken,฀somewhat฀ irregular. Distribution฀-฀Amazonian฀Peru,฀Brazil,฀and฀Colombia. Habitat฀and฀Ecology฀-฀Forest฀inundated฀by฀white฀(várzea)฀or฀black฀(tahuampa)฀ water,฀on฀yellowish,฀lateritic฀soil.฀At฀elevations฀of฀80-200฀m.฀Flowering:฀March,฀July,฀ and฀September;฀fruiting:฀throughout฀the฀year. Notes฀-฀Cremastosperma฀microcarpum฀resembles฀most฀closely฀C.฀gracilipes,฀from฀ which฀it฀differs฀in฀the฀denser,฀longer฀hairs฀on฀the฀flowers฀and฀acuminate฀as฀opposed฀ to฀cuspidate฀leaf฀apex.฀The฀hairy฀flower฀resemble฀somewhat฀those฀of฀C.฀cauliflorum,฀ but฀ which฀ cannot฀ be฀ confused฀ as฀ C.฀ microcarpum฀ never฀ exhibits฀ a฀ branching฀ inflorescence.฀In฀addition,฀the฀monocarps฀of฀C.฀cauliflorum฀are฀larger฀than฀those฀ of฀C.฀microcarpum฀and฀characteristically฀globose฀to฀transversely฀broadly฀ellipsoid฀as฀ opposed฀to฀ellipsoid.฀ 142 • Chapter 5 17. Cremastosperma monospermum฀(Rusby)฀R.E.Fr.฀-฀Fig.฀1฀c,฀21;฀Map฀4;฀ ฀ App.฀2,฀Fig.฀8 Cremastosperma฀monospermum฀(Rusby)฀R.E.Fr.฀(1931)฀193.฀-฀Cymbopetalum฀monospermum฀ Rusby฀(1910)฀505฀(“Symbopetalum”).฀-฀Type:฀R.S.฀Williams฀670฀(holo฀NY;฀iso฀K),฀ Bolivia,฀La฀Paz:฀San฀Buenaventura,฀470฀m,฀12฀Nov.฀1901. Cremastosperma฀juruense฀R.E.Fr.฀(1937)฀282.฀-฀Type:฀Krukoff฀4697฀(holo฀NY;฀iso฀S),฀ Brazil,฀Amazonas:฀Basin฀of฀Rio฀Juruá,฀near฀mouth฀of฀Rio฀Embira฀(tributary฀of฀Rio฀ Tarauacá),฀6฀Jun.฀1933. Cremastosperma฀monospermum฀(Rusby)฀R.E.Fr.฀var.฀brachypodum฀R.E.Fr.฀(1939)฀559.฀ -฀Type:฀Dahlgren฀&฀Sella฀162฀(holo฀S;฀iso฀B,฀F),฀Brazil,฀Pará:฀Boa฀Vista฀on฀the฀Tapajós฀ River,฀5฀May฀1929. Tree฀or฀shrub,฀1-12฀m฀tall,฀4-10฀cm฀diam.;฀young฀twigs฀and฀petioles฀glabrous.฀ Leaves:฀petioles฀5-10฀by฀1-3฀mm;฀lamina฀obovate,฀elliptic฀or฀narrowly฀so,฀(8-)10-35฀ by฀4-12฀cm฀(index฀2-3.5),฀chartaceous,฀green฀to฀brown,฀darker฀above,฀veins฀often฀ reddish฀below,฀glabrous฀on฀both฀sides฀or฀rarely฀sparsely฀covered฀with฀appressed฀ whitish฀to฀0.4฀mm฀long฀hairs฀on฀primary฀vein฀below,฀base฀acute฀to฀obtuse,฀rarely฀ narrowly฀cuneate,฀mostly฀decurrent,฀apex฀acuminate฀to฀cuspidate฀(acumen/cusp฀ 5-30฀mm฀long),฀primary฀vein฀1-2฀mm฀wide฀at฀widest฀point,฀secondary฀veins฀6-10,฀ intersecondary฀veins฀0-3,฀distance฀between฀from฀10-18฀mm฀at฀the฀base฀to฀12-24฀ mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀from฀70-80°฀at฀the฀base฀to฀4050°฀closer฀to฀the฀apex,฀rarely฀branching,฀forming฀mostly฀distinct฀loops,฀smallest฀ distance฀ between฀ loops฀ and฀ margin฀ 2-6฀ mm,฀ tertiary฀ veins฀ mostly฀ reticulate.฀ Inflorescence฀ of฀ single฀ flowers฀ solitary฀ (or฀ clustered฀ in฀ groups฀ of฀ 2),฀ on฀ leafy฀ or฀ leafless฀twigs;฀peduncles฀2-10฀by฀0.5-1฀mm฀(in฀flower),฀2-10฀by฀1.5-2฀mm฀(in฀ fruit),฀sparsely฀covered฀with฀appressed฀to฀erect฀golden฀to฀0.1฀mm฀long฀hairs฀or฀ glabrous;฀pedicels฀40-50(-70)฀by฀0.5-1฀mm฀at฀the฀base฀(in฀flower),฀(12-)22-73฀by฀ 1-1.5฀mm฀(in฀fruit),฀green฀or฀reddish฀in฀vivo,฀glabrous;฀2฀lower฀bracts,฀deltate,฀c.฀ 0.8฀by฀0.8฀mm฀(occasionally฀large฀and฀leafy),฀obtuse,฀caducous,฀sparsely฀covered฀ with฀appressed฀golden฀0.1฀mm฀long฀hairs฀or฀glabrous;฀upper฀bract฀mostly฀attached฀ around฀half฀way฀along฀pedicel,฀broadly฀ovate฀or฀deltate,฀1-2.5฀by฀c.฀1฀mm,฀acute,฀ obtuse,฀rounded฀or฀truncate,฀outer฀side฀sparsely฀covered฀with฀appressed฀golden฀0.1฀ mm฀long฀hairs฀or฀glabrous;฀flower฀buds฀broadly฀ovoid-triangular,฀remaining฀closed฀ (or฀nearly฀so)฀throughout฀development;฀flowers฀green,฀maturing฀to฀creamy฀yellow,฀ yellow฀or฀orange฀in฀vivo,฀dark฀or฀reddish฀brown฀or฀black฀in฀sicco;฀sepals฀and฀petals฀ glabrous,฀sepals฀free฀or฀connate฀for฀c.฀1฀mm,฀broadly฀ovate฀or฀deltate,฀appressed,฀ patent฀or฀recurved,฀2-4฀by฀2-3฀mm,฀acute฀or฀obtuse,฀mostly฀caducous;฀outer฀petals฀ ovate,฀9-14฀by฀6-8฀mm,฀inner฀petals฀elliptic฀to฀ovate,฀or฀narrowly฀so,฀10-13฀by฀4-5฀ mm;฀androecium฀not฀seen;฀gynoecium฀not฀seen.฀Monocarps,฀stipes฀and฀receptacle฀ glabrous,฀monocarps฀10-29,฀ellipsoid฀to฀broadly฀so,฀slightly฀asymmetrical,฀9-11฀by฀ 7-8฀mm,฀green฀maturing฀to฀pink,฀maroon,฀red฀or฀(blue-)฀black฀in฀vivo,฀brown,฀dark฀ or฀reddish฀brown฀or฀black฀in฀sicco,฀with฀an฀excentric฀apicule;฀stipes฀(6-)8-15฀by฀c.฀ Revision and phylogeny of Cremastosperma • 143 a b F i g . 2 1 . Cremastosperma monospermum (Rusby) R.E.Fr. a. fruiting specimen; b flower buds (a: Nuñez & Timaná 12152; b: Cid Ferreira et al. 6301) 1฀mm;฀fruiting฀receptacle฀3-8฀mm฀diam.฀Seeds฀broadly฀ovoid,฀reddish฀brown,฀pitted,฀ pits฀appear฀black฀with฀raised฀rim,฀8-10฀by฀6-7฀mm,฀raphe฀sunken,฀regular. Distribution฀-฀Bolivia฀(Beni,฀La฀Paz,฀Pando),฀Brazil฀(Acre,฀Pará,฀Rondônia),฀ and฀widespread฀across฀Peru. 144 • Chapter 5 Habitat฀and฀Ecology฀-฀Primary฀and฀secondary฀lowland฀forest,฀occasionally฀ on฀poorly฀drained฀soils฀or฀brown฀latosols.฀At฀elevations฀of฀200-500฀m.฀Flowering:฀ April,฀July฀-฀December;฀fruiting:฀throughout฀the฀year.฀ Vernacular฀names:฀Bolivia:฀Yohisi.฀Peru:฀Ayacbara.฀ Notes฀-฀Cremastosperma฀monospermum฀is฀the฀most฀widespread฀species฀of฀the฀ genus฀-฀the฀only฀found฀both฀along฀the฀eastern฀foothills฀of฀the฀Andes฀as฀far฀south฀as฀ Bolivia฀and฀across฀Brazil฀south฀of฀the฀Amazon.฀It฀is฀best฀distinguished฀by฀the฀shape฀ of฀the฀flower฀bud:฀roughly฀triangular฀with฀an฀obtuse฀apex,฀apparently฀remaining฀ closed฀throughout฀development,฀with฀the฀petals฀not฀opening฀fully฀even฀at฀maturity.฀ In฀most฀other฀species฀of฀the฀genus฀the฀flower฀bud฀opens฀during฀development.฀C.฀ pendulum฀and฀C.฀yamayakatense฀also฀exhibit฀glabrous,฀closed฀flower฀buds,฀but฀the฀ shape฀in฀both฀is฀depressed฀ovoid.฀The฀latter฀also฀has฀a฀short,฀sturdy฀pedicel,฀very฀ different฀to฀C.฀monospermum,฀the฀flower฀of฀which฀is฀borne฀on฀a฀slender,฀and฀often฀ long฀(though฀rather฀variable),฀pedicel.฀ The฀authors฀do฀not฀consider฀it฀useful฀to฀recognise฀sub-specific฀taxa฀within฀C.฀ monospermum.฀The฀variation฀in฀stipe฀length฀and฀thickness฀represented฀by฀the฀type฀ of฀var.฀brachypodum,฀described฀by฀Fries,฀falls฀within฀that฀of฀the฀species฀as฀a฀whole฀ and฀is฀therefore฀synonymised฀here. 18. Cremastosperma napoense฀ Pirie฀-฀Fig.฀22;฀Map฀3 Cremastosperma฀ napoense฀ Pirie฀(2005)฀54,฀f.฀7.฀-฀Type:฀Alvarado฀267฀(holo฀U;฀iso฀ AAU,฀MO,฀QCNE),฀Ecuador,฀Napo:฀Cantón฀Archidonia,฀foothills฀south฀of฀Volcano฀ Sumaco,฀km฀50฀on฀Hollín฀-฀Loreto฀road,฀community฀Huahua฀Sumaco,฀1100฀m,฀3฀ May฀1989. Tree฀5-20฀m฀tall,฀10-15฀cm฀diam.;฀young฀twigs฀and฀petioles฀rather฀densely฀ covered฀with฀appressed฀whitish-golden฀hairs฀to฀0.2฀mm฀long.฀Leaves:฀petioles฀812(-18)฀by฀3-4฀mm;฀lamina฀narrowly฀elliptic,฀17-42฀by฀7-13฀cm฀(index฀1.8-3.9),฀ chartaceous,฀olive฀green฀or฀brown฀on฀both฀sides,฀venation฀darker฀below,฀glabrous฀ above,฀rather฀densely฀covered฀with฀appressed฀whitish-golden฀hairs฀to฀0.2฀mm฀long฀ on฀ veins฀ below,฀ base฀ acute,฀ apex฀ acute฀ to฀ acuminate฀ (acumen฀ 5-15฀ mm฀ long),฀ primary฀vein฀grooved฀over฀entire฀leaf฀length,฀verrucose฀at฀the฀base,฀3-4฀mm฀wide฀ at฀widest฀point,฀secondary฀veins฀(8-)10-15,฀occasionally฀1฀or฀2฀intersecondary฀veins,฀ distance฀between฀from฀4-9฀mm฀at฀the฀base,฀20-50(-60)฀mm฀in฀the฀centre฀to฀10-30฀ mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀from฀40-50°฀at฀the฀base฀to฀7080°฀closer฀to฀the฀apex,฀occasionally฀branching,฀forming฀more฀or฀less฀distinct฀loops฀ in฀the฀apical฀half,฀tertiary฀veins฀percurrent.฀Inflorescence฀of฀1-8฀flowers,฀branching,฀ solitary฀or฀clustered฀in฀groups฀of฀2,฀on฀leafless฀twigs฀and฀branches;฀peduncles฀722฀by฀1-1.5฀mm฀(in฀flower),฀10-22฀by฀2.5-3฀mm฀(in฀fruit);฀pedicels฀25-38฀by฀c.฀ 1฀mm฀diam.฀at฀the฀base,฀1.5-2฀mm฀diam.฀at฀the฀apex฀(in฀flower),฀25-38฀by฀2-3฀ mm฀diam.฀at฀the฀base,฀2-4฀mm฀diam.฀at฀the฀apex฀(in฀fruit),฀peduncles฀and฀pedicels฀ Revision and phylogeny of Cremastosperma • 145 F i g . 2 2 . Cremastosperma napoense Pirie. a. inflorescences; b. infrutescence and leaf (a: Cerón 2986; b: Alvarado 267) rather฀densely฀to฀densely฀covered฀with฀appressed฀whitish฀to฀golden฀hairs฀to฀0.2฀ mm฀long;฀single฀lower฀bract,฀caducous;฀upper฀bract฀attached฀around฀halfway฀along฀ the฀pedicel,฀broadly฀to฀depressed฀elliptic฀or฀broadly฀to฀depressed฀ovate,฀1-2฀by฀1.51.8฀mm,฀obtuse,฀outer฀side฀densely฀covered฀with฀appressed฀whitish-golden฀hairs฀to฀ 146 • Chapter 5 0.2฀mm฀long;฀flower฀buds฀depressed฀ovoid,฀opening฀early฀in฀development;฀flowers฀ green฀maturing฀to฀greenish฀yellow฀or฀cream฀in฀vivo,฀brown฀in฀sicco;฀sepals฀fused฀for฀ first฀0.5฀mm,฀deltate,฀appressed฀(basal฀1฀mm฀of฀sepals฀and฀petals฀reflexed),฀2.5-3฀by฀ c.฀3฀mm,฀acute฀or฀obtuse,฀caducous,฀inner฀side฀glabrous,฀outer฀side฀rather฀densely฀to฀ densely฀covered฀with฀appressed฀whitish-golden฀hairs฀to฀0.2฀mm฀long;฀outer฀petals฀ elliptic,฀8-15฀by฀5-9฀mm,฀inner฀petals฀elliptic,฀12-14฀by฀5-7฀mm,฀inner฀side฀of฀inner฀ and฀outer฀petals฀glabrous,฀most฀of฀outer฀side฀of฀outer฀petals฀rather฀densely฀to฀densely฀ covered฀with฀appressed฀whitish-golden฀hairs฀to฀0.2฀mm฀long฀(towards฀the฀margins฀ and฀apex฀glabrous),฀outer฀side฀of฀inner฀petals฀largely฀glabrous฀but฀with฀a฀narrow,฀ dense,฀sometimes฀branching฀band฀of฀appressed,฀whitish-golden,฀to฀0.2฀mm฀long฀ hairs฀extending฀from฀the฀base฀to฀halfway฀towards฀the฀apex;฀receptacle฀depressed฀ ovoid;฀androecium฀c.฀6฀mm฀diam.,฀stamens฀1-1.5฀mm฀long,฀connective฀appendage฀ roughly฀hexagonal,฀0.5-0.8฀mm฀wide,฀glabrous;฀gynoecium฀c.฀2฀mm฀diam.,฀carpels฀ up฀to฀c.฀40,฀c.฀1.5฀mm฀long,฀glabrous฀or฀sparsely฀covered฀with฀appressed,฀whitishgolden,฀to฀0.2฀mm฀long฀hairs.฀Monocarps฀16-37,฀ellipsoid,฀asymmetrical,฀12-20฀by฀ 10-13฀mm,฀green฀maturing฀to฀dark฀purple฀or฀black฀in฀vivo,฀black฀in฀sicco,฀glabrous;฀ stipes฀20-30฀by฀2-3฀mm,฀glabrous;฀fruiting฀receptacle฀depressed฀ovoid,฀7-12฀mm฀ diam.,฀glabrous.฀Seeds฀ellipsoid,฀brown,฀lightly฀furrowed฀(not฀pitted),฀c.฀13฀by฀9฀mm,฀ raphe฀neither฀sunken฀nor฀raised,฀encircling฀seed฀longitudinally. Distribution฀-฀Ecuador฀(Napo,฀one฀collection฀in฀Pastaza). Habitat฀and฀Ecology฀-฀Primary฀pluvial฀premontane฀forest,฀often฀on฀volcanic฀ soils฀ but฀ also฀ reported฀ growing฀ on฀ limestone.฀ At฀ elevations฀ of฀ 600-1300฀ m.฀ Flowering:฀September,฀November,฀December,฀and฀February;฀fruiting:฀August฀to฀ December,฀March฀to฀May. Vernacular฀name฀-฀Ecuador:฀Ayacara. Notes฀ -฀ The฀ characteristic฀ pattern฀ of฀ indument฀ on฀ the฀ inner฀ petals฀ of฀ Cremastosperma฀ napoense฀ appears฀ to฀ be฀ unique฀ for฀ the฀ genus.฀The฀ species฀ can฀ be฀ further฀ distinguished฀ by฀ the฀ combination฀ of฀ a฀ branching฀ inflorescence฀ and฀ glabrous฀fruits.฀The฀only฀other฀species฀in฀the฀genus฀with฀such฀an฀inflorescence฀ is฀C.฀cauliflorum,฀which฀differs฀both฀in฀the฀presence฀of฀brown฀indument฀on฀the฀ (characteristic฀globose฀to฀transversely฀broadly฀elliptic)฀monocarps฀and฀in฀the฀dense฀ covering฀of฀much฀longer฀hairs฀on฀the฀flowers. 19. Cremastosperma novogranatense฀R.E.Fr.฀-฀Fig.฀23;฀Map฀2 Cremastosperma฀novogranatense฀R.E.Fr.฀(1950)฀329.฀-฀Type:฀Cuatrecasas฀17573฀(holo฀S;฀ iso฀F,฀US),฀Colombia,฀El฀Valle:฀Costa฀del฀Pacifico,฀Río฀Cajambre,฀Silva,฀5-80฀m,฀5-15฀ May฀1944. Tree฀8-20฀m฀tall,฀5-13฀cm฀diam.;฀young฀twigs฀and฀petioles฀densely฀covered฀ with฀appressed฀or฀erect฀golden฀to฀whitish฀hairs฀to฀0.4฀mm฀long.฀Leaves:฀petioles฀(6-)฀ 10-20฀by฀3-5฀mm;฀axillary฀buds฀densely฀covered฀with฀appressed฀to฀erect฀golden฀ Revision and phylogeny of Cremastosperma • 147 a b F i g . 2 3 . Cremastosperma novogranatense R.E.Fr. a. fruiting specimen; b. flower (a: Cuatrecasas 17573; b: Cuatrecasas & Willard 26031) to฀whitish฀hairs฀to฀0.4฀mm฀long;฀lamina฀elliptic฀to฀obovate฀or฀narrowly฀so,฀(24-)฀ 32-50฀by฀11-21฀cm฀(index฀2.2-2.8),฀chartaceous,฀pale฀brownish฀green฀to฀greyish฀ green฀above,฀pale฀brownish฀green฀below,฀glabrous฀above,฀veins฀sparsely฀to฀rather฀ 148 • Chapter 5 densely฀covered฀with฀appressed฀golden฀to฀whitish฀hairs฀to฀0.5฀mm฀long฀below,฀ base฀obtuse฀to฀rounded,฀apex฀acuminate฀to฀cuspidate฀(acumen฀15-40฀mm฀long),฀ primary฀vein฀2-4฀mm฀wide฀at฀widest฀point,฀secondary฀veins฀10-13,฀intersecondary฀ veins฀1-3,฀distance฀between฀from฀4฀mm฀at฀the฀base฀to฀up฀to฀40฀mm฀closer฀to฀ the฀apex,฀angles฀with฀primary฀vein฀from฀50-70°฀at฀the฀base฀to฀45-55°฀closer฀to฀ the฀apex,฀not฀branching฀forming฀mostly฀distinct฀loops,฀smallest฀distance฀between฀ loops฀and฀margin฀2-7฀mm,฀tertiary฀veins฀percurrent.฀Inflorescence฀of฀single฀flowers,฀ solitary฀(or฀clustered฀in฀groups฀of฀2),฀on฀leafy฀or฀leafless฀twigs;฀peduncles฀2-3฀by฀ c.฀2฀mm฀(in฀flower),฀2-4฀by฀2.5-3฀mm฀(in฀fruit);฀pedicels฀10-20฀by฀c.฀2฀mm฀(in฀ flower),฀13-27฀by฀2.5-3฀mm฀(in฀fruit),฀peduncles฀and฀pedicels฀sparsely฀to฀densely฀ covered฀with฀appressed฀golden฀hairs฀to฀0.4฀mm฀long;฀single฀lower฀bract,฀ovate,฀c.฀ 2฀by฀1.5฀mm,฀acute,฀often฀persistent,฀outer฀sides฀of฀bracts,฀of฀sepals฀and฀of฀petals฀ densely฀covered฀with฀appressed฀golden฀hairs฀to฀0.6฀mm฀long;฀upper฀bract฀in฀the฀ lower฀half฀of฀the฀pedicel,฀broadly฀to฀depressed฀ovate,฀2-3.5฀by฀1.5-2฀mm,฀acute;฀ flower฀buds฀globose;฀flowers฀pale฀green,฀sepals฀light฀brownish฀green฀in฀vivo,฀outer฀ side฀of฀sepals฀and฀petals฀dark฀yellow฀or฀greyish฀green,฀inner฀side฀reddish฀brown฀or฀ dark฀brown฀in฀sicco;฀sepals฀free,฀ovate฀to฀broadly฀ovate,฀appressed฀or฀patent,฀7-10฀ by฀7-8฀mm,฀obtuse,฀often฀persistent,฀with฀prominent฀venation;฀outer฀petals฀elliptic,฀ 17-29฀by฀10-16฀mm,฀obtuse,฀inner฀petals฀narrowly฀elliptic,฀28-33฀by฀9-12฀mm,฀ obtuse;฀androecium฀not฀seen;฀stamens฀1.6-1.9฀mm฀long,฀connective฀appendage฀0.81฀mm฀by฀0.5-0.8฀mm฀wide;฀gynoecium฀not฀seen.฀Monocarps,฀stipes,฀and฀receptacle฀ sparsely฀to฀rather฀densely฀covered฀with฀appressed฀brown฀hairs฀to฀0.2฀mm฀long,฀ monocarps฀3-14,฀ellipsoid฀to฀broadly฀ellipsoid,฀asymmetrical,฀16-22฀by฀10-13฀mm,฀ yellow,฀orange,฀red฀or฀pale฀brown฀in฀vivo,฀blackish฀brown฀to฀black฀in฀sicco,฀with฀an฀ excentric฀apicule;฀stipes฀1.5-4฀by฀2฀mm;฀fruiting฀receptacle฀4-10฀mm฀diam.฀Seeds฀ ellipsoid,฀yellowish-brown,฀very฀shallowly฀pitted,฀c.฀16฀by฀c.฀10฀mm,฀raphe฀sunken,฀ irregular. Distribution฀-฀Pacific฀coast฀of฀Colombia฀(El฀Valle). Habitat฀ and฀ ecology฀ -฀ Rain฀ forest.฀At฀ elevations฀ of฀ 0-130฀ m.฀ Flowering:฀ September฀and฀October;฀fruiting:฀May,฀September. Notes฀-฀Cremastosperma฀novogranatense฀can฀be฀distinguished฀by฀its฀almost฀ sessile฀monocarps฀and฀by฀the฀large฀and฀densely฀hairy฀flowers฀with฀unusually฀large,฀ often฀persistent,฀sepals.฀It฀is฀most฀similar฀to฀C.฀westrae฀Pirie,฀the฀sepals฀of฀which฀are฀ much฀smaller,฀and฀indument฀in฀general฀of฀shorter,฀less฀dense฀hairs. 20. Cremastosperma oblongum฀ R.E.Fr.฀-฀Fig.฀24;฀Map฀6;฀App.฀2,฀Fig.฀9 Cremastosperma฀oblongum฀R.E.Fr.฀(1948)฀4.฀-฀Type:฀Killip฀&฀Smith฀23622฀(holo฀US;฀iso฀ NY,฀S),฀Peru,฀Junín:฀Río฀Pinedo,฀N฀of฀La฀Merced,฀30฀May฀1929. Tree฀or฀shrub฀2-12฀m฀tall,฀6-15฀cm฀diam.;฀young฀twigs฀and฀petioles฀sparsely฀ covered฀with฀appressed฀golden฀hairs฀to฀0.4฀mm฀long.฀Leaves:฀petioles฀5-16฀by฀2Revision and phylogeny of Cremastosperma • 149 a b F i g . 2 4 . Cremastosperma oblongum R.E.Fr. a. fruiting specimen; b. flower (a: Foster & d’Achille 10224; b: Maas et al. 4592) 4฀mm;฀lamina฀elliptic,฀obovate฀or฀narrowly฀so,฀13-45฀by฀5-12฀cm฀(index฀2.5-4),฀ coriaceous,฀shiny฀grey-green฀or฀brown฀above,฀dull฀brown฀below,฀glabrous฀above,฀ sparsely฀ covered฀ with฀ appressed฀ golden฀ hairs฀ to฀ 0.4฀ mm฀ long฀ on฀ veins฀ below,฀ base฀acute฀(rarely฀cordate),฀apex฀acuminate฀(acumen฀(5-)10-20฀mm฀long),฀primary฀ 150 • Chapter 5 vein฀grooved฀in฀basal฀half,฀2-4฀mm฀wide฀at฀widest฀point,฀secondary฀veins฀8-19,฀ intersecondary฀veins฀often฀1(-2),฀distance฀between฀from฀8-24฀mm฀at฀the฀base฀to฀ 12-27฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀from฀40-70°฀at฀the฀base฀ to฀20-60°฀closer฀to฀the฀apex,฀sometimes฀branching,฀forming฀distinct฀loops,฀smallest฀ distance฀ between฀ loops฀ and฀ margin฀ 2-5฀ mm,฀ tertiary฀ veins฀ mostly฀ percurrent.฀ Inflorescence฀of฀single฀flowers฀clustered฀in฀groups฀of฀up฀to฀three,฀on฀leafy฀or฀leafless฀ twigs฀(occasionally฀on฀main฀stem,฀then฀on฀brachyblasts);฀peduncles฀c.฀2฀by฀1฀mm฀(in฀ flower),฀2-5฀by฀1.5-3฀mm฀(in฀fruit);฀pedicels฀7-18฀by฀c.฀1฀mm฀at฀the฀base฀(in฀flower),฀ 12-35(-55)฀by฀1.5-3฀mm฀(in฀fruit),฀peduncles฀and฀pedicels฀sparsely฀covered฀with฀ appressed฀whitish฀hairs฀to฀0.1฀mm฀long;฀2฀or฀3฀lower฀bracts,฀deltate,฀0.5-1฀by฀0.5-1฀ mm,฀obtuse,฀caducous฀or฀persistent,฀sparsely฀covered฀with฀appressed฀whitish฀hairs฀ to฀0.1฀mm฀long;฀upper฀bract฀attached฀in฀basal฀half฀of฀pedicel,฀shallowly฀triangular,฀ c.฀1฀by฀c.฀2฀mm,฀obtuse,฀persistent,฀sparsely฀covered฀with฀appressed฀whitish฀hairs฀to฀ 0.1฀mm฀long;฀no฀closed฀flower฀buds฀seen;฀flowers฀green฀maturing฀to฀cream,฀yellow฀ or฀orange/yellow฀in฀vivo,฀black฀in฀sicco;฀sepals฀free,฀deltate,฀often฀recurved,฀c.฀2฀by฀2฀ mm,฀acute,฀caducous,฀sparsely฀covered฀with฀appressed฀whitish฀hairs฀<฀0.1฀mm฀long฀ or฀glabrous;฀petals฀glabrous,฀outer฀petals฀obovate฀to฀narrowly฀so,฀c.฀16฀by฀6-8฀mm,฀,฀ inner฀petals฀elliptic฀to฀narrowly฀so,฀15-16฀by฀4-6฀mm;฀androecium฀c.฀8฀mm฀diam.,฀ stamens฀1-1.5฀mm฀long,฀connective฀appendage฀0.5-0.8฀mm฀wide;฀gynoecium฀c.฀ 1.5฀mm฀diam.,฀stigmas฀sparsely฀covered฀with฀erect฀golden฀hairs฀<฀0.1฀mm฀long.฀ Monocarps,฀stipes,฀and฀receptacle฀glabrous,฀monocarps฀6-20,฀ellipsoid,฀asymmetrical,฀ 16-20฀by฀10-14฀mm,฀green฀maturing฀through฀orange฀or฀red฀to฀black฀in฀vivo,฀brown฀ or฀black฀in฀sicco,฀with฀an฀excentric฀apicule;฀stipes฀green฀maturing฀to฀red฀in฀vivo,฀ 9-17฀by฀2-3฀mm;฀fruiting฀receptacle฀7-12฀mm฀diam.฀Seeds฀ellipsoid,฀reddish฀brown,฀ pitted,฀c.฀13฀by฀8฀mm,฀raphe฀raised,฀regular. Distribution฀-฀Central฀and฀southern฀Peru฀(Cuzco,฀Huánuco,฀Junín,฀Loreto,฀ Madre฀de฀Dios,฀Pasco,฀San฀Martín,฀and฀Ucayali),฀and฀adjacent฀Brazil฀(Acre). Habitat฀and฀Ecology฀-฀Primary,฀often฀upland,฀rainforest,฀on฀white฀sands,฀brown฀ latosols฀and฀limestone฀soils.฀At฀elevations฀of฀100-1300฀m.฀Flowering:฀September,฀ December;฀fruiting:฀March-July,฀November-January. Vernacular฀names฀-฀Peru:฀Bara฀caspi,฀Carahuasca,฀Carahuasca฀amarilla,฀Hicoja,฀ Palo฀blanco,฀Tortuga฀blanca. Notes฀-฀Cremastosperma฀oblongum฀is฀best฀discerned฀from฀the฀most฀similar฀other฀ species฀on฀the฀basis฀of฀floral฀characters:฀the฀sepals฀are฀small฀and฀recurved฀(unlike฀C.฀ megalophyllum)฀and฀borne฀on฀short฀pedicels,฀whilst฀bud฀development฀is฀open฀(not฀ the฀case฀in฀C.฀yamayakatense).฀The฀leaves฀are฀also฀distinctive:฀rather฀leathery฀with฀a฀ greyish฀colour฀on฀the฀upper฀side,฀with฀secondary฀veins฀forming฀conspicuous฀loops฀ and฀often฀narrowly฀elliptic.฀Fruiting฀specimens฀display฀more฀variation฀-฀particularly฀ in฀the฀length฀of฀the฀pedicel.฀In฀particular,฀cauliflorous฀specimens฀from฀the฀Peruvian฀ department฀of฀Pasco฀have฀longer฀pedicels,฀as฀does฀the฀type฀specimen฀itself. Revision and phylogeny of Cremastosperma • 151 21. Cremastosperma pacificum฀ R.E.Fr.฀-฀Fig.฀25;฀Map฀3฀ Cremastosperma฀pacificum฀R.E.Fr.฀(1950)฀330.฀-฀Type:฀Cuatrecasas฀17463฀(holo฀S;฀iso฀F,฀ US),฀Colombia,฀El฀Valle:฀Costa฀del฀Pacífico,฀Río฀Cajambre,฀Silva,฀5-80฀m,฀5-15฀May฀ 1944. Tree฀3-15฀m฀tall,฀2.5-25฀cm฀diam.;฀young฀twigs฀and฀petioles฀sparsely฀to฀rather฀ densely฀covered฀with฀appressed฀golden฀hairs฀to฀0.4฀mm฀long฀or฀glabrous.฀Leaves:฀ petioles฀8-16฀by฀2-4฀mm;฀lamina฀elliptic฀to฀obovate,฀or฀narrowly฀so,฀19-41฀by฀9-16฀ cm฀(index฀2.1-3.1),฀chartaceus฀to฀slightly฀coriaceous,฀brown,฀brownish฀green,฀or฀ greyish฀green฀and฀shiny฀above,฀brown,฀pale฀brown฀or฀greenish฀brown฀below,฀glabrous฀ above,฀glabrous฀or฀sparsely฀to฀densely฀covered฀with฀appressed฀golden฀hairs฀to฀0.4฀ mm฀long฀particularly฀on฀veins฀below,฀base฀acute,฀apex฀acuminate฀(acumen฀10-20฀ mm฀long),฀primary฀vein฀shallowly฀grooved฀at฀base,฀2-3฀mm฀wide฀at฀widest฀point,฀ secondary฀veins฀7-12,฀intersecondary฀veins฀occasionally฀1-2,฀distance฀between฀from฀ 8฀mm฀at฀the฀base฀to฀up฀to฀55฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀ from฀30-50°฀at฀the฀base฀to฀50-70°฀closer฀to฀the฀apex,฀not฀branching,฀often฀forming฀ distinct฀loops฀for฀the฀apical฀third,฀smallest฀distance฀between฀loops฀and฀margin฀1.5-2฀ mm,฀tertiary฀veins฀percurrent.฀Inflorescence฀of฀single฀solitary฀flowers,฀on฀leafless฀twigs;฀ peduncles฀c.฀1฀by฀1฀mm฀(in฀flower),฀2-3฀by฀c.฀2฀mm฀(in฀fruit),฀rather฀densely฀to฀ densely฀covered฀with฀appressed฀golden฀or฀whitish฀hairs฀to฀0.2฀mm฀long;฀pedicels฀ 12-20฀by฀c.฀1฀mm฀(in฀flower),฀22-35฀by฀1.5฀mm฀(in฀fruit),฀longitudinally฀furrowed,฀ glabrous฀or฀sparsely฀to฀rather฀densely฀covered฀with฀appressed฀gold฀or฀whitish฀hairs฀ to฀0.2฀mm฀long;฀single฀lower฀bract,฀deltate,฀1-1.5฀by฀c.฀1฀mm,฀obtuse฀or฀acute,฀ occasionally฀persistent,฀densely฀covered฀with฀appressed฀gold฀or฀whitish฀hairs฀0.2฀ mm฀ long;฀ upper฀ bract฀ in฀ the฀ lower฀ half฀ of฀ the฀ pedicel,฀ deltate,฀ 1-1.5฀ by฀ 1-1.5฀ mm,฀rounded,฀outer฀side฀sparsely฀to฀rather฀densely฀covered฀with฀appressed฀gold฀ or฀whitish฀hairs฀to฀0.2฀mm฀long฀or฀glabrous;฀flowers฀pale฀greenish฀yellow,฀or฀pale฀ green฀in฀vivo,฀black฀or฀dark฀brown฀in฀sicco,฀sepals฀and฀petals฀glabrous;฀sepals฀free,฀ very฀broadly฀ovate,฀3฀by฀3-3.5฀mm,฀obtuse,฀caducous;฀outer฀petals฀elliptic฀to฀broadly฀ elliptic,฀16฀by฀11-12฀mm,฀inner฀petals฀obovate,฀15฀by฀7฀mm,฀obtuse;฀androecium฀ not฀seen;฀stamens฀1.4฀mm฀long,฀connective฀appendage฀0.6฀mm฀wide;฀gynoecium฀ not฀seen.฀Monocarps฀(2-)7-21,฀ellipsoid,฀slightly฀asymmetrical,฀15-18฀by฀10-12฀mm,฀ green฀(immature)฀in฀vivo,฀black฀or฀dark฀brown฀in฀sicco,฀with฀an฀excentric฀apicule,฀ monocarps,฀stipes฀and฀receptacle฀glabrous;฀stipes฀10-18฀by฀1฀mm;฀fruiting฀receptacle฀ depressed฀ovoid,฀3-9฀mm฀diam.฀Seeds฀ellipsoid,฀yellow,฀furrowed฀and฀lightly฀pitted,฀ c.฀10฀by฀8฀mm,฀raphe฀raised฀within฀sunken฀groove,฀regular. Distribution฀-฀Pacific฀coast฀of฀Colombia฀(El฀Valle฀and฀Chocó). Habitat฀and฀Ecology฀-฀Tropical฀wet฀and฀pluvial฀forest,฀reported฀as฀growing฀on฀ yellow฀clay฀with฀alluvial฀substrate.฀At฀elevations฀of฀5-100฀m.฀Flowering:฀December฀ and฀August;฀fruiting:฀April฀and฀May. Notes฀-฀Cremastosperma฀pacificum฀appears฀similar฀to฀a฀number฀of฀other฀species฀ characterised฀ by฀ the฀ absence฀ of฀ (visible)฀ indument฀ on฀ flowers฀ and฀ fruits.฀ Most฀ 152 • Chapter 5 a b F i g . 2 5 . Cremastosperma pacificum R.E.Fr. a. fruiting specimen; b. flower (a: Cuatrecasas 17463; b: Sánchez et al. 323) Revision and phylogeny of Cremastosperma • 153 similar฀are฀C.฀magdalenae฀and฀C.฀megalophyllum,฀the฀sepals฀of฀both฀of฀which฀are฀ much฀larger฀(4-7฀mm฀long,฀as฀opposed฀to฀up฀to฀3฀mm฀in฀C.฀pacificum).฀From฀the฀ limited฀floral฀material฀available,฀bud฀development฀in฀C.฀pacificum฀would฀not฀appear฀ to฀be฀open,฀with฀bud฀shape฀similar฀to฀that฀of฀C.฀yamayakatense.฀In฀contrast฀to฀both฀ C.฀yamayakatense฀and฀C.฀panamense฀(with฀open฀bud฀development)฀the฀monocarps฀ of฀C.฀pacificum฀are฀relatively฀large฀(>฀15฀mm฀long,฀as฀opposed฀to฀up฀to฀14฀mm)฀and฀ around฀the฀same฀length฀as,฀rather฀than฀longer฀than,฀the฀stipes. 22. Cremastosperma panamense฀Maas฀-฀Fig.฀26;฀Map฀2 Cremastosperma฀panamense฀Maas฀in฀Maas฀et฀al.฀(1986)฀254,฀f.฀5฀&฀6฀-฀Type:฀Johnston฀ 1812฀(holo฀MO;฀iso฀A,฀MICH),฀Panama,฀Canal฀Zone:฀NW฀part฀of฀Canal฀Zone,฀area฀ W฀of฀Limon฀Bay,฀Gatun฀Locks฀and฀Gatun฀Lake,฀Maru฀Towers,฀7฀Apr.฀1956. Tree฀ or฀ shrub฀ 1.5-7(-20)฀ m฀ tall,฀ 3-10฀ cm฀ diam.;฀ young฀ twigs฀ and฀ petioles฀ glabrous.฀ Leaves:฀ petioles฀ 2-10฀ by฀ 1-2.5฀ mm,฀ caniculate฀ above,฀ verrucose฀ or฀ furrowed;฀lamina฀narrowly฀elliptic,฀8-22฀by฀2-7฀cm฀(index฀2.7-4.7),฀chartaceous,฀ green,฀ brownish฀ green,฀ or฀ greyish฀ green฀ (or฀ brown)฀ above,฀ (pale)฀ green,฀ (pale)฀ brownish฀green฀(or฀brown)฀below,฀shiny฀on฀both฀sides,฀glabrous฀on฀both฀sides,฀ base฀acute฀to฀obtuse,฀decurrent,฀rarely฀narrowly฀cuneate,฀apex฀acuminate฀(acumen฀ 5-25฀mm฀long),฀primary฀vein฀occasionally฀shallowly฀grooved฀at฀the฀base,฀1-1.5฀ mm฀wide฀at฀widest฀point,฀verrucose฀below,฀secondary฀veins฀5-10,฀intersecondary฀ veins฀1-4,฀distance฀between฀from฀5-20฀mm฀at฀the฀base฀to฀15-25฀mm฀closer฀to฀the฀ apex,฀angles฀with฀primary฀vein฀from฀35-55(-70)°฀at฀the฀base฀to฀60-75°฀closer฀to฀ the฀apex,฀not฀branching,฀forming฀mostly฀distinct฀loops,฀smallest฀distance฀between฀ loops฀and฀margin฀2-5฀mm,฀tertiary฀veins฀largely฀reticulate฀with฀little฀distinction฀ between฀tertiary฀and฀quaternary฀veins.฀Inflorescence฀of฀single,฀solitary฀flowers,฀on฀ leafy฀or฀leafless฀twigs;฀peduncles฀1-3฀by฀0.5-1฀mm฀(in฀flower),฀1-3฀by฀1.5-2฀mm฀(in฀ fruit),฀sparsely฀covered฀with฀erect฀whitish฀to฀golden฀hairs฀to฀0.1฀mm฀long;฀pedicels฀ 12-20฀by฀0.5-0.8฀mm฀at฀the฀base฀(in฀flower),฀13-22฀by฀1-2฀mm฀(in฀fruit),฀glabrous;฀ (1-)2-3(-several)฀lower฀bract(s),฀deltate,฀c.฀0.5฀by฀0.5฀mm,฀obtuse,฀mostly฀caducous,฀ sparsely฀covered฀with฀erect฀whitish฀to฀golden฀hairs฀to฀0.1฀mm฀long;฀upper฀bract฀ halfway฀ along฀ the฀ pedicel,฀ ovate฀ to฀ depressed฀ ovate,฀ 1-1.5฀ by฀ 0.7-1mm,฀ acute,฀ obtuse฀or฀emarginate,฀glabrous;฀closed฀flower฀buds฀very฀broadly฀ovoid,฀opening฀in฀ development;฀flowers฀green฀when฀immature,฀maturing฀to฀white,฀cream,฀or฀(pale)฀ yellow฀in฀vivo,฀yellow฀brown,฀dark฀brown,฀or฀blackish฀brown฀in฀sicco,฀sepals฀and฀ petals฀glabrous;฀sepals฀free,฀broadly฀to฀very฀broadly฀ovate,฀appressed฀or฀patent,฀1.5-3฀ by฀1.5-3฀mm,฀obtuse,฀caducous฀or฀persistent;฀outer฀petals฀narrowly฀ovate฀to฀narrowly฀ elliptic,฀7-18฀by฀4-6฀mm,฀obtuse,฀inner฀petals฀narrowly฀ovate฀to฀narrowly฀elliptic,฀ 10-25฀by฀3-5฀mm,฀obtuse;฀androecium฀diam.฀unknown,฀stamens฀1.2-1.5฀mm฀long,฀ connective฀appendage฀0.7-0.9฀mm฀wide;฀gynoecium฀diam.฀unknown,฀carpels฀c.฀35,฀ 2฀mm฀long,฀glabrous.฀Monocarps,฀stipes,฀and฀receptacle฀glabrous,฀monocarps฀(2-)8154 • Chapter 5 a b F i g . 2 6 . Cremastosperma panamense Maas. a. fruiting specimen; b. flowers (a: Johnston 1812; b: Pérez 832) Revision and phylogeny of Cremastosperma • 155 30,฀more฀or฀less฀globose,฀slightly฀asymmetrical,฀8-13฀by฀7-11฀mm,฀green฀maturing฀ to฀yellow,฀orange,฀red,฀or฀black฀in฀vivo,฀pale฀brown,฀reddish฀brown,฀or฀brown฀in฀ sicco,฀with฀an฀excentral฀apicule;฀stipes฀7-21฀by฀1-1.5(-3)฀mm;฀fruiting฀receptacle฀ 3-8฀mm฀diam.฀Seeds฀broadly฀ellipsoid,฀globose฀or฀transversally฀ellipsoid,฀light฀brown,฀ pitted,฀7-12฀by฀6-10฀mm,฀raphe฀sunken,฀regular. Distribution฀-฀Panama฀(Coclé,฀Colón,฀Panamá,฀and฀San฀Blas). Habitat฀ and฀ Ecology฀ -฀ Evergreen฀ tropical฀ wet฀ forest,฀ cloud฀ forest฀ or฀ low฀ swampy฀places฀or฀in฀disturbed฀areas.฀At฀elevations฀of฀0-800฀m.฀Flowering:฀April,฀ May,฀July,฀and฀August;฀fruiting:฀throughout฀the฀year. Vernacular฀names฀-฀Panama:฀Palo฀santo,฀Sate฀wawa,฀Waras฀gid. Notes฀-฀Cremastosperma฀panamense฀appears฀similar฀to฀C.฀magdalenae,฀but฀can฀ be฀distinguished฀by฀the฀smaller฀size฀of฀the฀sepals,฀which฀additionally฀persist฀less฀ frequently฀into฀fruiting.฀C.฀pacificum฀and฀C.฀chococola฀are฀both฀geographically฀close฀ (Pacific฀coast฀of฀Colombia)฀and฀share฀the฀characters฀of฀glabrous฀pedicels,฀(flowers)฀ and฀fruits.฀However,฀the฀pedicels฀of฀C.฀chococola฀are฀considerably฀longer,฀and฀the฀ shape฀ of฀ the฀ larger฀ monocarps฀ of฀ C.฀ pacificum฀ (ellipsoid฀ as฀ opposed฀ to฀ roughly฀ globose),฀as฀well฀as฀the฀larger฀leaves,฀allow฀easy฀distinction฀in฀both฀cases.฀The฀colour฀ of฀the฀relatively฀small฀leaves฀(drying฀consistently฀green),฀the฀relatively฀large฀distance฀ between฀ the฀ loops฀ of฀ their฀ secondary฀ veins฀ and฀ the฀ margin,฀ and฀ the฀ reticulate฀ and฀indistinct฀nature฀of฀tertiary฀and฀quaternary฀venation฀of฀C.฀panamense฀are฀also฀ distinctive. 23. Cremastosperma pedunculatum฀ (Diels)฀R.E.Fr.฀-฀Fig.฀27;฀Map฀7 Cremastosperma฀pedunculatum฀(Diels)฀R.E.Fr.฀(1930)฀48.฀-฀Aberemoa฀pedunculata฀Diels฀ (1906)฀ 409.฀ -฀Type:฀ Weberbauer฀ 4558฀ (holo฀ B;฀ iso฀ F,฀ G,฀ MOL),฀ Peru,฀ San฀ Martín:฀ Moyobamba,฀1100-1200฀m,฀1906. Tree฀4-15฀m฀tall,฀8-15฀cm฀diam.;฀young฀twigs฀and฀petioles฀glabrous.฀Leaves:฀ petioles฀4-15฀by฀1-3฀mm;฀lamina฀elliptic฀to฀obovate,฀or฀narrowly฀so,฀12-27฀by฀4-10฀ cm฀(index฀2-3.6),฀chartaceous,฀(dark)฀greyish฀brown฀above,฀blackish฀brown฀with฀ darker฀veins฀below,฀glabrous฀above,฀glabrous฀to฀sparsely฀covered฀with฀appressed฀ golden฀hairs฀to฀0.6฀mm฀long,฀particularly฀on฀veins฀below,฀base฀acute฀to฀rounded,฀ decurrent,฀apex฀acuminate฀(acumen฀5-20฀mm฀long),฀primary฀vein฀1.5-2.5฀mm฀ wide฀at฀widest฀point,฀secondary฀veins฀7-13,฀intersecondary฀veins฀often฀1-3,฀distance฀ between฀ from฀ 5฀ mm฀ at฀ the฀ base฀ to฀ 15-20฀ mm฀ closer฀ to฀ the฀ apex,฀ angles฀ with฀ primary฀vein฀inconsistent,฀40-60°฀at฀the฀base฀and฀closer฀to฀the฀apex,฀not฀branching,฀ forming฀distinct฀loops,฀smallest฀distance฀between฀loops฀and฀margin฀1-4฀mm,฀tertiary฀ veins฀more฀or฀less฀percurrent.฀Inflorescence฀of฀single฀flowers฀clustered฀in฀groups฀of฀ up฀ to฀ 2,฀ on฀ leafy฀ or฀ leafless฀ twigs฀ or฀ main฀ trunk;฀ peduncles฀ 2-10฀ by฀ c.฀ 1฀ mm฀ (in฀flower),฀5-15฀by฀c.1.5฀mm฀(in฀fruit),฀sparsely฀to฀rather฀densely฀covered฀with฀ appressed฀golden฀or฀whitish฀hairs฀0.1-฀0.4฀mm฀long;฀pedicels฀(30-)35-75(-95)฀by฀c.฀ 156 • Chapter 5 F i g . 2 7 . Cremastosperma pedunculatum (Diels) R.E.Fr. a. flowering twig; b receptacle, illustrating insertion of carpels; c. carpel with single apical/lateral ovule (a-c: Weberbauer 4558, reproduced from Fries (1930)) 1฀mm฀at฀the฀base,฀to฀3฀mm฀diam.฀at฀the฀apex฀(in฀flower),฀(30-)55-85(-110)฀by฀11.5฀mm฀at฀the฀base,฀to฀3฀mm฀diam.฀at฀the฀apex฀(in฀fruit),฀sparsely฀to฀rather฀densely฀ (at฀the฀base)฀covered฀with฀appressed฀golden฀or฀whitish฀hairs฀0.1-฀0.4฀mm฀long฀or฀ glabrous;฀1-several฀lower฀bracts,฀elliptic,฀c.฀1.5฀by฀1฀mm,฀acute,฀caducous,฀sparsely฀to฀ Revision and phylogeny of Cremastosperma • 157 rather฀densely฀covered฀with฀appressed฀golden฀or฀whitish฀hairs฀0.1-฀0.4฀mm฀long;฀ upper฀bract฀attached฀in฀variable฀position฀on฀pedicel,฀ovate฀to฀very฀broadly฀ovate,฀ 1-2.5฀by฀1-2฀mm,฀acute,฀obtuse,฀or฀rounded,฀sparsely฀to฀rather฀densely฀covered฀ with฀appressed฀golden฀or฀whitish฀hairs฀0.1-฀0.4฀mm฀long฀or฀glabrous;฀closed฀flower฀ buds฀very฀broadly฀ovoid฀to฀globose,฀opening฀loosely฀in฀development;฀flowers฀green,฀ maturing฀to฀green-violet,฀yellow฀or฀pale฀cream-yellow,฀inner฀petals฀with฀purple฀ base฀in฀vivo,฀dark฀brown฀or฀reddish฀brown฀in฀sicco,฀sepals฀and฀petals฀glabrous;฀sepals฀ free฀or฀connate฀for฀1฀mm,฀broadly฀to฀very฀broadly฀ovate฀or฀broadly฀ovate-triangular,฀ appressed,฀patent฀or฀recurved,฀2.5-4฀by฀2.5-4฀mm,฀obtuse,฀mostly฀persistent;฀outer฀ petals฀elliptic฀to฀broadly฀elliptic,฀11-17฀by฀7-13฀mm,฀inner฀petals฀elliptic,฀obovate฀or฀ narrowly฀so,฀11-19฀by฀4-8฀mm,฀obtuse฀or฀rounded,฀petals฀with฀prominent฀venation;฀ androecium฀c.฀7฀mm฀diam.,฀stamens฀1.3-1.8฀mm฀long,฀connective฀appendage฀0.50.8฀ mm฀ wide;฀ gynoecium฀ c.฀ 1.5฀ mm฀ diam.,฀ glabrous,฀ carpels฀ 2-2.2฀ mm฀ long.฀ Monocarps฀3-27,฀ellipsoid฀to฀broadly฀ellipsoid,฀asymmetrical,฀12-17฀by฀10-12฀mm,฀ green฀(immature)฀in฀vivo,฀black,฀dark฀brown,฀or฀reddish฀brown฀in฀sicco,฀with฀an฀ excentric฀apicule,฀monocarps,฀stipes฀and฀receptacle฀glabrous;฀stipes฀11-21฀by฀1.5-2฀ mm;฀fruiting฀receptacle฀4-10฀mm฀diam.฀Seeds฀ellipsoid฀to฀broadly฀ellipsoid,฀reddish฀ brown,฀pitted,฀c.฀10฀by฀c.฀7฀mm,฀raphe฀sunken,฀regular. Distribution฀-฀Peru฀(San฀Martín฀and฀Cajamarca),฀Ecuador฀(Zamora-Chinchipe).฀ Two฀collections฀of฀less฀certain฀affinity฀have฀been฀made฀further฀north฀in฀Ecuador฀ (Pastaza,฀Morona-Santiago)฀and฀one฀in฀Colombia฀(Caquetá). Habitat฀ and฀ Ecology฀ -฀ Premontane฀ and฀ montane฀ primary฀ and฀ secondary฀ forest,฀sometimes฀inundated,฀mainly฀on฀soils฀with฀calcareous฀bedrock.฀At฀elevations฀ of฀850-1800฀m฀(except฀the฀single฀specimen฀collected฀in฀Pastaza฀(Ecuador)฀at฀360฀ m).฀ Flowering:฀ July,฀ October-December;฀ fruiting฀ February,฀ July,฀ October,฀ and฀ December. Notes฀-฀The฀length฀of฀the฀pedicels฀of฀C.฀pedunculatum฀is฀only฀matched฀or฀ exceeded฀by฀those฀of฀C.฀bullatum฀(distinguished฀by฀the฀bullate฀appearance฀of฀the฀ leaves฀and฀dense,฀long,฀indument฀on฀most฀parts)฀or฀C.฀longipes฀(from฀the฀western฀ side฀of฀the฀Andes,฀and฀with฀much฀longer฀pedicels฀and฀larger฀leaves).฀ The฀holotype฀of฀Guatteria฀socialis,฀C.฀Schunke฀395,฀was฀determined฀by฀ Diels฀as฀C.฀pedunculatum.฀Fries฀(1931)฀deliberately฀omitted฀placing฀G.฀socialis฀in฀ synonymy฀under฀C.฀pedunculatum,฀citing฀differences฀in฀the฀reported฀growth฀form.฀ The฀type฀is฀reported฀by฀the฀collector฀to฀be฀a฀liana.฀This฀has฀not฀been฀recorded฀ for฀any฀other฀collections฀of฀C.฀pedunculatum฀(although฀it฀is฀reported฀for฀the฀type฀ specimen฀ of฀ C.฀ oblongum).฀ In฀ addition,฀ the฀ collection฀ was฀ made฀ in฀ the฀ central฀ Peruvian฀department฀of฀Junin,฀much฀further฀south฀than฀the฀known฀distribution฀ of฀C.฀pedunculatum฀(in฀northern฀Peru฀and฀Ecudor).฀A฀photo฀of฀the฀holotype฀(not฀ the฀specimen฀itself)฀was฀available฀to฀the฀authors.฀The฀collection฀appears฀to฀be฀of฀ a฀Cremastosperma฀(the฀leaves฀with฀a฀raised฀midrib),฀but฀includes฀only฀immature฀ buds.฀Although฀ Maas฀ et฀ al.฀ (1994)฀ listed฀ G.฀ socialis฀ as฀ a฀ taxonomic฀ synonym฀ of฀ 158 • Chapter 5 C.฀ pedunculatum,฀ we฀ consider฀ the฀ available฀ evidence฀ insufficient฀ to฀ assign฀ this฀ specimen฀to฀a฀particular฀species฀in฀Cremastosperma. 24. Cremastosperma pendulum฀(Ruiz฀&฀Pav.)฀R.E.Fr.฀-฀Fig.฀28;฀Map฀8;฀App.฀2,฀฀ ฀ Fig.฀10 Cremastosperma฀pendulum฀(Ruiz฀&฀Pav.)฀R.E.Fr.฀(1931)฀325.฀-฀Guatteria฀pendula฀Ruiz฀ &฀Pav.฀(1798)฀146.฀-฀Typus:฀Pavón฀s.n.฀(holo฀G),฀Peru,฀without฀location. Tree฀or฀shrub฀4-10฀m฀tall;฀young฀twigs฀and฀petioles฀glabrous.฀Leaves:฀petioles฀ (3-)6-13฀by฀1-3฀mm;฀lamina฀narrowly฀elliptic,฀7-30฀by฀3.5-10฀cm฀(index฀2.9-4.2),฀ chartaceous,฀drying฀green฀(darker฀above),฀glabrous฀above,฀sparsely฀covered฀with฀ appressed฀whitish฀hairs฀to฀0.2฀mm฀long฀particularly฀on฀veins฀or฀glabrous฀below,฀ base฀acute,฀apex฀acuminate฀(acumen฀15-25฀mm฀long),฀primary฀vein฀1-2฀mm฀wide฀ at฀widest฀point,฀deeply฀grooved฀in฀basal฀half,฀secondary฀veins฀7-10,฀intersecondary฀ veins฀often฀1(-2),฀distance฀between฀from฀9-16฀mm฀at฀the฀base฀to฀11-25฀mm฀closer฀ to฀the฀apex,฀angles฀with฀primary฀vein฀from฀50-60°฀at฀the฀base฀to฀45-50°฀closer฀to฀ the฀apex,฀sometimes฀branching,฀forming฀distinct฀loops,฀smallest฀distance฀between฀ loops฀and฀margin฀2-4฀mm,฀tertiary฀veins฀more฀or฀less฀percurrent.฀Inflorescence฀of฀ solitary฀flowers฀on฀leafy฀twigs;฀peduncles฀0.5-1.5฀by฀c.฀0.5฀mm฀(in฀flower),฀3-5฀ by฀0.8-1฀mm฀(in฀fruit),฀sparsely฀covered฀with฀appressed฀whitish฀hairs฀<฀0.1฀mm฀ long฀or฀glabrous;฀pedicels฀28-70฀by฀0.3-0.5฀mm฀at฀the฀base,฀to฀1฀mm฀diam.฀at฀the฀ apex฀(in฀flower),฀50-70฀by฀0.5-1฀mm฀at฀the฀base,฀to฀1.5฀mm฀diam.฀at฀the฀apex฀(in฀ fruit),฀green฀or฀purple฀in฀vivo,฀glabrous;฀1฀or฀2฀lower฀bracts,฀deltate,฀c.฀0.5฀by฀0.5฀ mm,฀acute,฀caducous,฀rather฀densely฀covered฀with฀appressed฀whitish฀hairs฀<฀0.1฀ mm฀long;฀upper฀bract฀attached฀halfway฀along฀pedicel,฀elliptic฀or฀deltate,฀0.5-1฀by฀ 0.5-1฀mm,฀acute฀or฀obtuse,฀sparsely฀covered฀with฀appressed฀whitish฀hairs฀<฀0.1฀ mm฀long฀or฀glabrous;฀flower฀buds฀depressed฀ovoid,฀petals฀remaining฀closed฀(or฀ nearly฀so)฀throughout฀development;฀flowers฀green฀in฀vivo,฀black฀in฀sicco,฀sepals฀ and฀petals฀glabrous;฀sepals฀fused฀for฀basal฀0.5฀mm,฀elliptic,฀appressed฀(particularly฀ in฀bud)฀or฀recurved,฀c.฀2฀by฀1.5฀mm,฀obtuse,฀caducous;฀outer฀petals฀ovate,฀5-6฀by฀ 3-4฀mm,฀inner฀petals฀ovate,฀c.฀3.5฀by฀2฀mm;฀androecium฀c.฀4฀mm฀diam.,฀connective฀ appendage฀0.5-0.8฀mm฀wide;฀gynoecium฀c.฀1฀mm฀diam.,฀glabrous.฀Monocarps฀611,฀ellipsoid,฀slightly฀asymmetrical,฀10-13฀by฀7-9฀mm,฀monocarps฀and฀stipes฀green฀ maturing฀to฀dark฀reddish฀brown฀in฀vivo,฀medium฀to฀dark฀brown฀in฀sicco,฀with฀an฀ excentric฀apicule,฀monocarps,฀stipes,฀and฀receptacle฀glabrous;฀stipes฀9-15฀by฀1-2฀ mm;฀fruiting฀receptacle฀depressed฀ovoid,฀3-6฀mm฀diam.฀ Seeds฀ellipsoid,฀reddish฀ brown,฀pitted,฀c.฀13฀by฀7฀mm,฀raphe฀sunken,฀regular. Distribution฀-฀Peru฀(Huánuco,฀Pasco,฀Ucayali). Habitat฀and฀Ecology฀-฀Primary฀and฀secondary฀tropical฀lowland฀and฀upland฀ forest฀on฀brown฀and฀red฀latosols.฀At฀elevations฀of฀180-500฀m.฀Flowering:฀January,฀ August,฀September,฀and฀November;฀fruiting:฀May,฀July,฀August. Revision and phylogeny of Cremastosperma • 159 a b F i g . 2 8 . Cremastosperma pendulum (Ruiz & Pav.) R.E.Fr. a. fruiting specimen; b. flower bud (a: Foster & d’Achille 10008; b: Foster 9842) Notes฀ -฀ Cremastosperma฀ pendulum฀ can฀ be฀ distinguished฀ by฀ its฀ long฀ thin฀ pedicel฀in฀combination฀with฀the฀shape฀of฀the฀small฀flower฀buds฀which฀remain฀ closed฀throughout฀development:฀it฀differs฀from฀C.฀yamayakatense฀in฀the฀greater฀ 160 • Chapter 5 length฀of฀pedicel฀and฀from฀C.฀monospermum฀in฀the฀depressed฀ovoid฀rather฀than฀ triangular฀flower฀bud.฀ 25. Cremastosperma peruvianum฀ R.E.Fr.฀-฀Fig.฀29;฀Map฀7 Cremastosperma฀peruvianum฀R.E.Fr.฀(1934)฀204.฀-฀Type:฀Tessmann฀4176฀(holo฀B;฀iso฀S),฀ Peru,฀Loreto:฀upper฀Río฀Marañon,฀Pongo฀de฀Manseriche,฀160฀m,฀1฀Oct.฀1924. Tree฀2-7฀m฀tall;฀young฀twigs฀and฀petioles฀glabrous.฀Leaves:฀petioles฀3-15฀by฀ 4-8฀mm,฀verrucose฀or฀transversally฀furrowed;฀lamina฀narrowly฀elliptic฀to฀narrowly฀ obovate,฀38-64฀by฀11-17(-24)฀cm฀(index฀2.7-4.4),฀chartaceous฀or฀coriaceous,฀pale฀ olive฀or฀brownish฀green฀on฀both฀sides,฀glabrous,฀base฀obtuse,฀rounded,฀or฀cordate,฀ often฀asymetrical,฀apex฀acuminate฀(acumen฀15-35฀mm฀long),฀primary฀vein฀deeply฀ grooved฀in฀basal฀half,฀3-5฀mm฀wide฀at฀widest฀point,฀glabrous,฀secondary฀veins฀2026,฀intersecondary฀veins฀1-2,฀distance฀between฀from฀2-10฀mm฀at฀the฀base,฀20-40฀ mm฀in฀the฀centre,฀10-20฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀from฀ 55-65°฀at฀the฀base฀to฀70-80°฀closer฀to฀the฀apex,฀not฀branching,฀forming฀mostly฀ distinct฀loops,฀smallest฀distance฀between฀loops฀and฀margin฀2-4฀mm,฀tertiary฀veins฀ percurrent.฀Inflorescence฀of฀single฀solitary฀flowers,฀on฀leafy฀or฀leafless฀twigs;฀short฀ axillary฀shoot,฀1-1.5฀by฀1-1.5฀mm฀(in฀flower),฀c.฀3฀by฀c.฀2.5฀mm฀(in฀fruit),฀sparsely฀ covered฀with฀appressed฀or฀erect฀golden฀hairs฀<฀0.1฀mm฀long฀or฀glabrous;฀pedicels฀ 18-20฀by฀c.฀1฀mm฀at฀the฀base฀(in฀flower),฀20-25฀by฀c.฀2฀mm฀(in฀fruit),฀red฀in฀vivo,฀ glabrous;฀2฀lower฀bracts,฀elliptic,฀c.฀1฀by฀0.5฀mm,฀obtuse,฀caducous,฀sparsely฀covered฀ with฀appressed฀golden฀hairs฀<฀0.1฀mm฀long;฀upper฀bract฀attached฀near฀the฀base฀ of฀the฀pedicel,฀(broadly)฀ovate,฀1.5-3฀by฀c.฀1.5฀mm,฀rounded,฀glabrous;฀flower฀buds฀ broadly฀ovoid-triangular,฀remaining฀loosely฀closed฀in฀development;฀flowers฀green,฀ maturing฀to฀bright฀yellow฀(green฀at฀the฀base)฀outside,฀brown฀inside฀in฀vivo,฀orange฀ or฀yellowish฀brown฀(sepals฀and฀bracts฀lighter)฀in฀sicco,฀sepals฀and฀petals฀glabrous;฀ sepals฀free,฀broadly฀ovate,฀appressed,฀c.฀4฀by฀3.5฀mm,฀obtuse,฀caducous;฀outer฀petals฀ elliptic฀ to฀ narrowly฀ obovate,฀ 15-29฀ by฀ 8-13฀ mm,฀ obtuse,฀ inner฀ petals฀ narrowly฀ elliptic฀to฀narrowly฀obovate,฀(13-)21-25฀by฀4-8฀mm;฀androecium฀diam฀unknown,฀ stamens฀1.5-1.9฀mm฀long,฀connective฀appendage฀0.6-0.8฀mm฀wide;฀gynoecium฀ diam.฀unknown,฀carpels฀c.฀2฀mm฀long.฀Monocarps,฀stipes,฀and฀receptacle฀glabrous,฀ monocarps฀3-20,฀ellipsoid,฀asymmetrical,฀16-19฀by฀12-13฀mm,฀green฀maturing฀to฀ yellow,฀purple฀or฀black฀in฀vivo,฀reddish฀brown,฀dark฀brown฀or฀black฀in฀sicco,฀without฀ apparent฀apicule;฀stipes฀20-40฀by฀2฀mm;฀fruiting฀receptacle฀4-9฀mm฀diam.฀Seeds฀ broadly฀ellipsoid,฀reddish฀brown,฀pitted,฀11฀by฀10฀mm,฀raphe฀sunken,฀regular. Distribution฀-฀Peru฀(Amazonas฀and฀Loreto). Habitat฀and฀Ecology฀-฀Primary,฀non-inundated฀forest,฀on฀white฀sand฀or฀red฀ clay.฀At฀ elevations฀ of฀ 170-400฀ m.฀ Flowering:฀ September฀ and฀ October;฀ fruiting:฀ April,฀June฀-฀August,฀and฀October. Vernacular฀names฀-฀Peru:฀Achuana,฀Chiwanim. Revision and phylogeny of Cremastosperma • 161 b a F i g . 2 9 . Cremastosperma peruvianum R.E.Fr. a. fruiting specimen; b. flower bud (a: Knapp et al. 7645; b: Díaz et al. 8225) 162 • Chapter 5 Notes฀-฀Cremastosperma฀peruvianum฀can฀be฀distinguished฀from฀other฀species฀ of฀ the฀ genus฀ by฀ its฀ long,฀ relatively฀ narrow฀ leaves,฀ and฀ long฀ stipes.฀The฀ flower฀ resembles฀that฀of฀C.฀monospermum฀in฀shape,฀but฀is฀larger,฀with฀relatively฀larger฀sepals,฀ and฀borne฀on฀a฀shorter,฀thicker฀pedicel. 26. Cremastosperma stenophyllum฀ Pirie฀-฀Fig.฀30;฀Map฀2 Cremastosperma฀stenophyllum฀Pirie฀(2005)฀56,฀f.฀8.฀-฀Type:฀Knapp฀&฀Mallet฀6159฀(holo฀ QCNE),฀Ecuador,฀Pichincha:฀‘Tinalandia’,฀km฀112฀on฀the฀road฀to฀Santo฀Domingo฀de฀ los฀Colorados฀from฀Quito,฀500-1000฀m,฀15฀Jan.฀1984. Tree฀ c.฀ 10฀ m฀ tall,฀ c.฀ 20฀ cm฀ diam.;฀ young฀ twigs฀ and฀ petioles฀ rather฀ densely฀ covered฀with฀appressed฀golden฀hairs฀to฀0.2฀mm฀long.฀Leaves:฀petioles฀5-9฀by฀2-2.5฀ mm;฀ lamina฀ narrowly฀ elliptic,฀ 25-30฀ by฀ 6.5-8฀ cm฀ (index฀ 3.8-4.3),฀ chartaceous,฀ minutely฀verrucose,฀greyish฀green฀above,฀green฀below,฀very฀sparsely฀covered฀with฀ appressed฀yellowish-white฀hairs฀to฀0.2฀mm฀long฀below฀and฀on฀veins฀above,฀base฀ acute,฀apex฀acuminate฀(acumen฀20-25฀mm฀long),฀primary฀vein฀1-1.5฀mm฀wide฀ at฀widest฀point,฀verrucose,฀secondary฀veins฀8-10,฀intersecondary฀veins฀occasional,฀ distance฀between฀from฀5฀mm฀at฀the฀base฀to฀50฀mm฀closer฀to฀the฀apex,฀angles฀with฀ primary฀vein฀from฀45-55°฀at฀the฀base฀to฀70-80°฀closer฀to฀the฀apex,฀not฀branching,฀ not฀ forming฀ loops,฀ tertiary฀ veins฀ with฀ some฀ reticulation.฀ Inflorescence฀ of฀ single฀ flowers฀(1฀flower฀observed)฀on฀brachyblasts฀on฀thicker฀twigs฀or฀branches;฀peduncle฀ c.฀1.5฀by฀c.฀1฀mm฀(in฀flower);฀pedicels฀c.฀45฀by฀c.฀1฀mm฀(in฀flower),฀peduncles฀and฀ pedicels฀and฀outer฀side฀of฀bracts฀(densely),฀sepals฀(densely),฀and฀petals฀(sparsely฀to฀ rather฀densely)฀covered฀with฀appressed฀yellowish-white฀hairs฀to฀0.2฀mm฀long;฀2฀ lower฀bracts,฀deltate,฀c.฀1฀mm฀long,฀obtuse;฀upper฀bract฀attached฀on฀basal฀half฀of฀ pedicel,฀ovate,฀c.฀1.5฀by฀0.8฀mm,฀acute;฀flowers฀green,฀maturing฀to฀yellow฀in฀vivo,฀ light฀brown฀with฀dark฀brown฀patches฀at฀the฀base฀of฀the฀petals฀in฀sicco;฀sepals฀deltate,฀ 2฀mm฀long,฀obtuse;฀outer฀petals฀elliptic,฀c.฀18฀by฀8฀mm,฀inner฀petals฀elliptic,฀c.฀18฀ mm฀long฀(diam.฀unknown).฀Fruit฀not฀seen. Distribution฀ -฀ Ecuador,฀ (Pichincha฀ and฀ Bolívar).฀At฀ elevations฀ of฀ 500-฀ 1200฀m. Habitat฀ and฀ Ecology฀ -฀ Secondary฀ vegetation฀ with฀ primary฀ elements.฀ Flowering:฀January. Notes฀-฀Only฀two฀collections฀of฀Cremastosperma฀stenophyllum฀one฀of฀which฀ sterile,฀have฀been฀observed฀by฀the฀author.฀However,฀these฀are฀consistently฀distinct฀ from฀all฀other฀species฀of฀the฀genus.฀C.฀stenophyllum฀can฀be฀distinguished฀even฀when฀ sterile฀ by฀ the฀ conspicuously฀ green-drying,฀ narrowly฀ elliptic฀ leaves.฀The฀ flower฀ somewhat฀resembles฀those฀of฀C.฀awaense฀but฀both฀the฀pedicel฀and฀leaf฀acumen฀are฀ longer฀and฀C.฀stenophyllum฀also฀lacks฀the฀distinctive฀pattern฀of฀indument฀on฀the฀ petals฀of฀C.฀awaense:฀the฀hairs฀are฀instead฀evenly฀distributed฀on฀the฀outer฀surfaces. Revision and phylogeny of Cremastosperma • 163 F i g . 3 0 . Cremastosperma stenophyllum Pirie. a. flower and leaf (Knapp & Mallet 6159) 164 • Chapter 5 27. Cremastosperma venezuelanum฀ Pirie฀-฀Fig.฀31;฀Map฀9 Cremastosperma฀venezuelanum฀Pirie฀in฀Chatrou฀&฀Pirie฀(2005)฀34,฀f.฀1.฀-฀Type:฀Diaz฀&฀ Niño฀231฀(holo฀U;฀iso฀NY),฀Venezuela,฀Carababo:฀Autonomo฀Mora,฀Watershed฀of฀Río฀ Morón,฀700-1100฀m,฀3฀May฀1991. Tree฀7-10฀m฀tall;฀young฀twigs฀and฀petioles฀glabrous.฀Leaves:฀petioles฀10-20฀mm฀ long,฀3-4฀mm฀diam.;฀lamina฀narrowly฀elliptic,฀30-53฀by฀9-15฀cm฀(index฀3฀-฀3.6),฀ chartacous฀to฀subcoriacous,฀shiny,฀dark฀brown฀above,฀olive฀green/greyish/light฀to฀ dark฀ brown฀ below,฀ glabrous฀ on฀ both฀ sides,฀ base฀ cordate฀ (rarely฀ rounded),฀ apex฀ obtuse฀to฀acute,฀primary฀vein฀2.5-5฀mm฀wide฀at฀the฀widest฀point,฀secondary฀veins฀ 15-30,฀intersecondary฀veins฀occasional,฀distance฀between฀from฀12฀mm฀at฀the฀base฀ to฀40฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀from฀90°฀at฀the฀base฀to฀ 60-50°฀closer฀to฀the฀apex,฀rarely฀branching,฀forming฀more฀or฀less฀indistinct฀loops,฀ smallest฀distance฀between฀loops฀and฀margin฀2-3฀mm;฀tertiary฀veins฀showing฀some฀ reticulation.฀Inflorescence฀of฀single฀flowers฀clustered฀in฀groups฀of฀up฀to฀6,฀produced฀ from฀leafless฀branches฀or฀from฀the฀main฀trunk;฀peduncles฀c.฀2฀by฀1฀mm฀(in฀flower),฀ 3-5฀by฀1.5-3฀mm฀(in฀fruit);฀pedicels฀c.฀11฀by฀1฀mm฀at฀the฀base฀(in฀flower),฀16-22฀by฀ 1.5-2฀mm฀(in฀fruit),฀glabrous;฀single฀lower฀bract,฀very฀broadly฀triangular,฀0.5-1฀mm฀ long,฀obtuse,฀persistent,฀glabrous;฀upper฀bract฀attachment฀variable฀within฀central฀ 80%฀of฀length,฀c.฀1฀by฀1.5฀mm,฀depressed฀triangular,฀obtuse,฀glabrous;฀closed฀flower฀ buds฀not฀seen;฀flowers฀dark฀brown฀to฀black฀in฀sicco,฀sepals฀and฀petals฀glabrous;฀sepals฀ free,฀depressed฀triangular,฀reflexed,฀c.฀1฀by฀1.5฀mm,฀acute฀to฀obtuse,฀persistent฀on฀less฀ mature฀fruits;฀outer฀petals฀elliptic,฀c.฀18฀by฀10฀mm,฀inner฀petals฀narrowly฀elliptic,฀c.฀ 21฀by฀6฀mm;฀androecium฀c.฀4฀mm฀diam.,฀stamens฀c.฀1฀mm฀long,฀connective฀rhombic,฀ c.฀0.4฀mm฀wide;฀gynoecium฀c.฀1.8฀mm฀diam.,฀carpels฀0.6-0.7฀mm฀long,฀glabrous.฀ Monocarps฀20-35,฀ellipsoid,฀asymmetrical,฀17-20฀by฀12-13฀mm,฀black฀in฀sicco,฀with฀ a฀strongly฀excentric฀apicule,฀monocarps,฀stipes,฀and฀receptacle฀glabrous;฀stipes฀1522฀by฀1.5-2฀mm;฀fruiting฀receptacle฀5-12฀mm฀diam.฀Seeds฀ellipsoid,฀orange-brown,฀ shallowly฀pitted,฀15-17฀by฀13-16฀mm,฀raphe฀raised,฀regular. Distribution฀-฀Venezuela฀(Aragua฀and฀Carabobo). Habitat฀ &฀ Ecology฀ -฀ Understorey฀ of฀ primary,฀ moist,฀ evergreen฀ forest.฀At฀ elevations฀of฀350-1100฀m.฀Fruiting:฀April฀and฀May;฀flowering:฀August. Note฀-฀Cremastosperma฀venezuelanum฀is฀best฀distinguished฀from฀other฀species฀of฀ Cremastosperma฀by฀its฀distinctive฀acute฀to฀obtuse฀leaf฀apex฀(as฀opposed฀to฀acuminate฀ or฀cuspidate฀in฀other฀species).฀The฀combination฀of฀cordate฀(rarely฀rounded)฀leaf฀ base,฀the฀large฀angles฀of฀the฀secondary฀with฀the฀primary฀veins฀near฀the฀base฀of฀the฀ leaves฀and฀the฀lack฀of฀indument฀on฀any฀parts฀is฀also฀unique.฀Only฀one฀other฀species฀ of฀Cremastosperma฀has฀been฀collected฀in฀Venezuela:฀C฀macrocarpum฀Maas,฀which฀has฀ longer฀pedicels฀and฀larger฀monocarps฀with฀shorter,฀thicker฀stipes.฀ Revision and phylogeny of Cremastosperma • 165 F i g . 3 1 . Cremastosperma venezuelanum Pirie. A. leaves; B. flower; C. immature fruit (A, C: Edwards 397; B: Steyermark 94314) 166 • Chapter 5 28. Cremastosperma westrae฀ Pirie฀-฀Fig.฀32;฀Map฀3 Cremastosperma฀westrae฀Pirie฀(2005)฀58,฀f.฀9.฀-฀Type:฀Liesner฀682฀(holo฀MO),฀Panama,฀ Panamá:฀road฀from฀El฀Llano฀to฀Carti-Tupile,฀12฀miles฀above฀Pan-American฀Highway,฀ 200-500฀m,฀13฀Mar.฀1973. Tree฀or฀shrub฀4-8฀m฀tall;฀young฀twigs฀and฀petioles฀sparsely฀to฀densely฀covered฀ with฀ appressed฀ brown฀ hairs฀ c.฀ 0.1฀ mm฀ long.฀ Leaves:฀ petioles฀ 6-20฀ by฀ 3-6฀ mm;฀ lamina฀narrowly฀elliptic฀to฀slightly฀obovate฀or฀narrowly฀so,฀(15-)30-50฀by฀(8-)1220฀cm฀(index฀1.6-3),฀chartaceus฀to฀coriaceus,฀dark฀to฀olive฀green฀or฀brown฀above,฀ lighter฀below,฀glabrous฀above,฀sparsely฀covered฀with฀appressed฀whitish฀hairs฀0.1฀ mm฀long฀on฀veins฀below,฀base฀acute฀to฀rounded,฀apex฀acuminate฀(acumen฀1020฀mm฀long),฀primary฀vein฀2-5฀mm฀wide฀at฀widest฀point,฀secondary฀veins฀8-12,฀ no฀intersecondary฀veins,฀distance฀between฀from฀5-10฀mm฀at฀the฀base฀to฀40-60฀ mm฀ closer฀ to฀ the฀ apex,฀ angles฀ with฀ primary฀ vein฀ from฀ 50-80°฀ at฀ the฀ base฀ to฀ 45-60°฀closer฀to฀the฀apex,฀not฀branching,฀forming฀more฀or฀less฀distinct฀loops฀in฀ apical฀ half,฀ smallest฀ distance฀ between฀ loops฀ and฀ margin฀ 2-5฀ mm,฀ tertiary฀ veins฀ mostly฀percurrent.฀Inflorescence฀of฀single,฀solitary฀flowers,฀on฀leafy฀or฀leafless฀twigs;฀ peduncles,฀2-3฀by฀2-3฀mm฀(in฀fruit);฀pedicels฀6-17฀by฀2-3฀mm฀(in฀fruit),฀peduncles฀ and฀pedicels฀rather฀densely฀to฀densely฀covered฀with฀erect฀whitish฀hairs฀0.1฀mm฀ long;฀single฀lower฀bract,฀caducous;฀upper฀bract฀attached฀in฀the฀basal฀half฀of฀the฀ pedicel,฀depressed฀ovate,฀c.฀1.5฀by฀2฀mm,฀rounded,฀outer฀side฀densely฀covered฀with฀ appressed฀brown฀hairs฀0.2฀mm฀long;฀flower฀buds฀depressed฀ovoid,฀flowers฀yellow฀ in฀vivo,฀black฀with฀yellow฀indument฀in฀sicco;฀sepals฀free฀or฀fused฀for฀basal฀0.5฀mm,฀ deltate,฀appressed,฀2.5-3฀by฀2.5-3฀mm,฀rounded,฀often฀persistent,฀densely฀covered฀ (outside,฀sparsely฀inside)฀with฀appressed฀brown฀hairs฀0.2฀mm฀long;฀outer฀petals฀ elliptic฀to฀slightly฀ovate,฀c.฀9฀by฀5฀mm,฀rather฀densely฀to฀densely฀covered฀(outside,฀ sparsely฀inside)฀with฀appressed฀brown฀hairs฀c.฀0.2฀mm฀long,฀inner฀petals฀elliptic,฀c.8฀ by฀4฀mm,฀rather฀densely฀to฀densely฀(towards฀the฀apex฀outside,฀sparsely฀inside฀and฀ at฀base)฀covered฀with฀appressed฀brown฀hairs฀0.2฀mm฀long;฀androecium฀c.฀6฀mm฀ diam.,฀stamens฀c.฀1.3฀mm฀long,฀connective฀appendage฀c.฀0.7฀mm฀wide,฀glabrous;฀ gynoecium฀c.฀1฀mm฀diam.,฀carpels฀sparsely฀covered฀with฀erect฀brown฀hairs฀<0.1฀ mm฀long.฀Monocarps฀6-10,฀ellipsoid,฀slightly฀asymmetrical,฀18-22฀by฀10-12฀mm,฀ green,฀maturing฀to฀yellow,฀orange,฀red฀or฀black฀in฀vivo,฀reddish฀to฀blackish฀brown฀ in฀sicco,฀with฀a฀small฀excentric฀apicule,฀monocarps฀and฀stipes฀sparsely฀to฀rather฀ densely฀covered฀with฀erect฀whitish฀hairs฀<0.1฀mm฀long฀or฀glabrous;฀stipes฀4-14฀ by฀1-2฀mm;฀fruiting฀receptacle฀depressed฀ovoid,฀3-6฀mm฀diam.,฀densely฀covered฀ with฀ erect฀ whitish฀ hairs฀ <0.1฀ mm฀ long.฀ Seeds฀ ellipsoid,฀ reddish฀ brown,฀ surface฀ wrinkled฀and฀slightly฀pitted,฀c.฀16฀by฀10฀mm,฀raphe฀slightly฀sunken,฀encircling฀seed฀ longitudinally. Revision and phylogeny of Cremastosperma • 167 F i g . 3 2 . Cremastosperma westrae Pirie. a. fruit and leaf (acuminate apex); b. flower and leaf (apex broken off) (a: de Nevers 4475; b: Liesner 682) Distribution฀-฀Panama฀(Darién,฀San฀Blas,฀and฀Panamá). Habitat฀ and฀ Ecology฀ -฀ Primary฀ seasonal฀ evergreen฀ forest฀ on฀ red฀ clay.฀At฀ elevations฀of฀50-600฀m.฀Flowering:฀March;฀fruiting:฀January,฀March,฀July,฀November,฀ and฀December. Notes฀-฀Cremastosperma฀westrae฀is฀most฀similar฀to฀C.฀novogranatense:฀it฀differs฀in฀ having฀longer฀stipes,฀smaller฀sepals,฀and฀less฀dense,฀shorter,฀hairs฀on฀the฀petals.฀The฀ shape฀of฀the฀fruits฀of฀C.฀pacificum฀bear฀a฀resemblance฀to฀those฀of฀C.฀westrae.฀A฀clear฀ distinction฀can฀be฀made฀due฀to฀the฀presence฀of฀indument฀on฀flowers฀and฀fruits฀in฀ C.฀westrae:฀those฀of฀C.฀pacificum฀are฀glabrous. 168 • Chapter 5 C.฀ westrae฀ is฀ reported฀ to฀ contain฀ cardiac฀ glycosides฀ (W.฀Wint,฀ on฀ Sudgen฀ 613). Cremastosperma฀westrae฀is฀named฀after฀the฀Dutch฀botanist฀Lubbert฀Y.฀Th.฀Westra,฀ whose฀lifelong฀dedication฀to฀plant฀systematics฀includes,฀in฀the฀last฀two฀decades,฀a฀ great฀contribution฀to฀the฀taxonomy฀of฀Neotropical฀Annonaceae.฀His฀encyclopaedic฀ knowledge฀remains฀an฀invaluable฀and฀much฀appreciated฀resource฀to฀colleagues฀at฀ the฀Utrecht฀branch฀of฀the฀National฀Herbarium฀of฀the฀Netherlands฀and฀further฀ afield. 29. Cremastosperma yamayakatense฀Pirie฀-฀Fig.฀33;฀Map฀8;฀App.฀2,฀Fig.฀11 Cremastosperma฀yamayakatense฀Pirie฀in฀Pirie฀&฀Zapata฀(2004)฀10,฀f.฀2,฀6-8.฀-฀Type:฀Pirie฀ et฀al.฀57฀(holo฀U;฀iso฀CUZ,฀HAO,฀HUT,฀K,฀MO,฀NY,฀USM),฀Peru,฀Amazonas:฀Bagua,฀ District฀Imaza,฀community฀Yamayakat,฀trail฀to฀Putuim,฀340฀m,฀22฀Nov.฀2003. Tree฀1.5-8฀(-20)฀m฀tall;฀young฀twigs฀and฀petioles฀shallowly฀grooved,฀glabrous.฀ Leaves:฀ petioles฀ 5-10฀ mm฀ long,฀ 1-5฀ mm฀ diam.;฀ lamina฀ narrowly฀ elliptic,฀ 11-24฀ (-38)฀ by฀ 3.5-8฀ (-13)฀ cm฀ (index฀ 2.4-3.4),฀ chartaceous,฀ olive-grey฀ green฀ above,฀ light฀brown฀below,฀glabrous฀on฀both฀sides,฀base฀acute,฀apex฀shortly฀acuminate฀to฀ acuminate฀(acumen฀10-25฀mm฀long),฀primary฀vein฀grooved฀in฀basal฀quarter฀to฀ third,฀1-4฀mm฀wide฀at฀widest฀point,฀secondary฀veins฀8-10฀(-14),฀intersecondary฀ veins฀occasional,฀distance฀between฀from฀5-10฀mm฀at฀the฀base฀to฀10-30฀mm฀closer฀ to฀the฀apex,฀angles฀with฀primary฀vein฀from฀70-80°,฀the฀angle฀thereafter฀decreasing฀ and฀subsequently฀increasing฀again฀towards฀the฀leaf฀margin,฀not฀branching,฀forming฀ distinct฀loops,฀smallest฀distance฀between฀loops฀and฀margin฀2-6฀mm,฀tertiary฀veins฀ largely฀ percurrent฀ with฀ some฀ reticulation.฀ Inflorescences฀ of฀ single,฀ successively฀ produced,฀flowers,฀axillary฀on฀leafy฀branches฀and฀on฀older฀(leafless)฀branches฀(then฀ on฀brachyblasts);฀peduncles฀c.฀1฀by฀1฀mm฀(in฀flower),฀1-3฀by฀2-2.5฀mm฀(in฀fruit),฀ sparsely฀covered฀with฀golden฀hairs฀c.฀0.1฀mm฀long;฀pedicels฀5-7฀by฀c.฀1.5฀mm฀at฀ the฀base฀(in฀flower),฀8-15(-20)฀by฀2-2.5฀mm฀at฀the฀base฀to฀4฀mm฀diam.฀at฀the฀ apex฀(in฀fruit),฀glabrous;฀single฀lower฀bract,฀deltate,฀1-2฀by฀1-2฀mm,฀acute,฀mostly฀ caducous฀in฀fruit,฀rather฀densely฀covered฀with฀golden฀hairs฀0.1฀mm฀long;฀upper฀ bract฀inserted฀within฀basal฀half฀of฀pedicel,฀deltate,฀1-2฀by฀1-2฀mm,฀acute,฀glabrous;฀ flower฀ buds฀ depressed฀ ovoid,฀ remaining฀ closed฀ in฀ development;฀ flowers฀ green฀ maturing฀to฀yellow฀in฀vivo,฀black฀in฀sicco,฀sepals฀and฀petals฀glabrous;฀sepals฀basally฀ connate,฀deltate,฀appressed,฀c.฀3฀by฀3฀mm,฀rounded,฀caducous,฀rarely฀persistent;฀outer฀ petals฀ovate,฀10-15฀by฀8-12฀mm,฀inner฀petals฀elliptic,฀c.฀12฀by฀6฀mm;฀androecium฀ 6-7฀ mm฀ diam.,฀ stamens฀ c.฀ 1฀ mm฀ long,฀ connective฀ appendage฀ c.฀ 0.5฀ mm฀ wide;฀ gynoecium฀c.฀3฀mm฀diam.,฀carpels฀length฀and฀indument฀unknown.฀Monocarps฀1022,฀green฀maturing฀through฀red฀to฀black฀in฀vivo,฀black฀in฀sicco,฀ellipsoid,฀slightly฀ asymmetrical,฀12-14฀by฀7-8฀mm,฀with฀an฀excentric฀apicule,฀glabrous฀or฀sparsely฀ Revision and phylogeny of Cremastosperma • 169 a b F i g . 3 3 . Cremastosperma yamayakatense Pirie. a. fruiting specimen with old flower (and loose petals); b. flower (a: Jaramillo & Apanu 584; b: Barbour 4432) 170 • Chapter 5 covered฀with฀golden฀hairs฀<0.1฀mm฀long;฀stipes฀green฀maturing฀to฀red฀in฀vivo,฀1112฀by฀c.฀1.5฀mm฀increasing฀to฀3฀mm฀when฀ripe,฀glabrous฀or฀sparsely฀covered฀with฀ golden฀hairs฀<0.1฀mm฀long;฀fruiting฀receptacle฀5-10฀mm฀in฀diam,฀glabrous.฀Seeds฀ ellipsoid,฀reddish-brown฀with฀small฀black฀pits฀surrounded฀by฀a฀slightly฀raised฀rim,฀ 9-13฀by฀6-7฀mm,฀raphe฀sunken,฀regular. Distribution฀-฀Peru฀(Amazonas),฀watershed฀of฀the฀upper฀Río฀Marañon. Habitat฀and฀Ecology฀-฀Primary฀and฀secondary฀forest.฀At฀elevations฀of฀2001000฀m.฀Flowering:฀November,฀January-March;฀fruiting:฀throughout฀the฀year฀except฀ December฀and฀April. Notes฀ -฀ Cremastosperma฀ yamayakatense฀ resembles฀ two฀ other฀ species฀ of฀ Cremastosperma;฀C.฀gracilipes,฀which฀has฀been฀collected฀in฀the฀departments฀of฀ Napo฀and฀Pastaza฀in฀Ecuador,฀Loreto฀in฀Peru฀and฀in฀adjacent฀Colombia,฀and฀C.฀ cenepense,฀from฀the฀Cenepa฀region฀of฀Amazonas,฀Peru,฀with฀which฀its฀distribution฀ therefore฀overlaps.฀The฀most฀important฀differences฀between฀C.฀yamayakatense฀and฀ C.฀gracilipes฀are฀in฀the฀flowers.฀C.฀gracilipes฀is฀characterised฀by฀flower฀buds฀which฀ open฀ during฀ development฀ and฀ which฀ bear฀ indument฀ on฀ all฀ parts.฀ In฀ contrast,฀ the฀flower฀buds฀of฀C.฀yamayakatense฀bear฀virtually฀no฀indument฀and฀appear฀to฀ remain฀ closed฀ throughout฀ development,฀ the฀ petals฀ only฀ opening฀ slightly฀ when฀ the฀ flowers฀ are฀ mature.฀Additionally,฀ the฀ flowers฀ of฀ C.฀ gracilipes฀ are฀ borne฀ on฀ longer฀more฀slender฀pedicels฀than฀those฀of฀C.฀yamayakatense.฀C.฀yamayakatense฀ differs฀from฀C.฀cenepense฀in฀the฀shape฀of฀the฀leaf฀base฀(acute฀in฀C.฀yamayakatense,฀ cordate฀to฀subcordate฀in฀C.฀cenepense)฀and฀the฀length฀of฀the฀stipes฀(longer฀than฀ the฀monocarps฀in฀C.฀yamayakatense,฀shorter฀than฀the฀monocarps฀in฀C.฀cenepense).฀ The฀lack฀of฀flowering฀material฀of฀C.฀cenepense฀makes฀further฀distinction฀currently฀ impossible. Flowering฀and฀fruiting฀specimens฀of฀C.฀yamayakatense฀of฀around฀1.5฀m฀tall฀were฀ observed฀in฀the฀province฀of฀Bagua,฀though฀specimens฀collected฀both฀in฀this฀area฀ and฀particularly฀those฀collected฀further฀north฀into฀the฀province฀of฀Condorcanqui,฀ in฀the฀area฀of฀the฀Río฀Cenepa,฀have฀been฀recorded฀as฀reaching฀heights฀of฀6-8฀m฀ and฀in฀one฀case฀20฀m฀tall.฀Differences฀between฀collections฀from฀these฀two฀regions฀ have฀been฀observed:฀The฀leaves฀of฀Condorcanqui฀specimens฀are฀generally฀larger฀ and฀the฀fruits฀have฀a฀slight฀indument฀whereas฀those฀of฀the฀Bagua฀collections฀are฀ glabrous.฀In฀the฀absence฀of฀floral฀material฀from฀the฀Cenepa฀region฀it฀is฀assumed฀ that฀these฀specimens฀do฀represent฀the฀same฀species฀due฀to฀the฀short฀pedicel,฀leaf฀ base฀shape฀(which฀excludes฀the฀possibility฀of฀their฀representing฀specimens฀of฀C.฀ cenepense)฀and฀leaf฀venation. Revision and phylogeny of Cremastosperma • 171 Dubious species Guatteria฀socialis฀J.F.Macbr.฀(1929)฀171.฀-฀Type:฀C.฀Schunke฀395฀(holo฀F),฀Peru,฀ Junín:฀Chanchamayo฀Valley,฀1500฀m,฀Oct.฀1924-1927. The฀type฀specimen฀of฀G.฀socialis฀represents฀a฀species฀of฀Cremastosperma,฀but฀ to฀which฀species฀it฀might฀belong฀is฀not฀clear฀due฀to฀the฀immature฀state฀of฀the฀ flower.฀Only฀a฀photo฀has฀been฀seen฀by฀the฀authors. Insufficiently known species 30. Cremastosperma spec. A฀ -฀Fig.฀34;฀Map฀2 Tree฀3-7฀m฀tall,฀5-7฀cm฀diam.;฀young฀twigs฀and฀petioles฀very฀sparsely฀covered฀ with฀appressed฀whitish฀hairs฀to฀0.2฀mm฀long฀or฀glabrous.฀Leaves:฀petioles฀7-15(-20)฀ by฀2-4฀mm;฀lamina฀narrowly฀obovate฀or฀elliptic,฀21-44฀by฀5-12฀cm฀(index฀2.6-3.4),฀ chartaceous,฀green฀or฀brown฀above฀and฀below,฀veins฀darker฀below,฀glabrous฀above,฀ very฀sparsely฀covered฀with฀appressed฀whitish฀hairs฀to฀0.2฀mm฀long,฀especially฀on฀ veins,฀or฀glabrous฀below,฀base฀acute฀or฀obtuse,฀apex฀acuminate฀(acumen฀11-20฀mm฀ long),฀primary฀vein฀shallowly฀grooved฀near฀base,฀1.5-2฀mm฀wide฀at฀widest฀point,฀ secondary฀veins฀7-10,฀intersecondary฀veins฀0-1,฀distance฀between฀from฀10-13฀mm฀ at฀the฀base฀to฀25-30฀mm฀closer฀to฀the฀apex,฀angles฀with฀primary฀vein฀from฀70-80°฀ at฀the฀base฀to฀45-60°฀closer฀to฀the฀apex,฀sometimes฀branching,฀forming฀distinct฀ loops฀in฀apical฀half฀of฀leaf,฀smallest฀distance฀between฀loops฀and฀margin฀2-3฀mm,฀ tertiary฀veins฀mostly฀percurrent.฀Inflorescence฀of฀single฀flowers฀solitary฀on฀leafless฀ twigs฀or฀clustered฀in฀groups฀on฀brachyblasts฀on฀the฀main฀stem;฀peduncles฀4฀by฀ 4฀mm฀(in฀fruit);฀pedicels฀18-23฀by฀3฀mm฀diam.฀at฀the฀base,฀3-4฀mm฀diam.฀at฀the฀ apex฀(in฀fruit),฀peduncles฀and฀pedicels฀glabrous;฀lower฀bract(s),฀upper฀bract฀and฀ flowers฀not฀seen.฀Monocarps฀20-32,฀ellipsoid฀(broadly฀so฀in฀immature฀specimens),฀ slightly฀asymmetrical,฀16-17฀by฀11-12฀mm,฀green฀maturing฀to฀yellowish,฀orange฀ and฀purple฀in฀vivo,฀black฀in฀sicco,฀with฀a฀small฀excentric฀apicule,฀monocarps,฀stipes฀ and฀receptacle฀glabrous;฀stipes฀17-18฀by฀2฀mm;฀fruiting฀receptacle฀14฀mm฀diam.฀ Seeds฀ellipsoid,฀reddish฀brown,฀shallowly฀pitted,฀c.฀12฀by฀10฀mm,฀raphe฀sunken,฀ regular. Distribution฀-฀Costa฀Rica:฀Osa฀peninsula. Habitat฀and฀Ecology฀-฀Tropical฀wet฀forest.฀At฀elevations฀of฀40-300฀m.฀Fruiting:฀ July,฀September. Notes฀-฀On฀the฀basis฀of฀the฀fruits฀Cremastosperma฀spec.฀A฀appears฀distinct,฀and฀ its฀distribution฀in฀Costa฀Rica฀is฀the฀furthest฀north฀into฀Central฀America฀of฀any฀ species฀of฀the฀genus.฀However,฀it฀is฀the฀authors฀opinion฀that฀formal฀description฀ should฀await฀the฀availability฀of฀flowering฀material.฀ The฀ fruits฀ are฀ reported฀ to฀ occasionally฀ be฀ inhabited฀ with฀ fiercely฀ biting฀ dolichotenne฀ants. 172 • Chapter 5 F i g . 3 4 . Cremastosperma spec. A. fruiting specimen (Aguilar 467) Revision and phylogeny of Cremastosperma • 173 Specimens฀examined: Costa฀rica:฀Puntarenas:฀Aguilar฀467฀(INB,฀MO),฀Reserva฀Forestal฀Golfo฀Dulce฀Rincón฀ de฀Osa,฀Quebrada฀Banegas;฀Aguilar฀4371฀(INB),฀Canton฀de฀Osa;฀Chatrou฀et฀al.฀103฀(U),฀ Cantón฀Osa,฀Bahia฀Chal,฀road฀to฀La฀Palma;฀Gentry฀et฀al.฀78657฀(F,฀INB,฀MO),฀Fila฀before฀ Rancho฀Quemado,฀near฀Rincón,฀Osa฀Peninsula.฀Transect฀1;฀Herrera฀3972฀(INB),฀Parque฀ Nacional฀Corcovado,฀Cerro฀Brujo,฀cabeceras฀de฀Quebrada฀Vaquedano;฀Kernan฀1224฀(U),฀ Parque฀Nacional฀Corcovado,฀Sirena,฀Los฀Patos฀Forest;฀Schatz฀et฀al.฀1002฀(MO,฀U,฀WIS);฀S฀of฀ the฀new฀road฀from฀the฀Interamerican฀Highway฀at฀Chacarita฀to฀Rincón฀de฀Osa,฀between฀ the฀road฀and฀the฀Golfo฀Dulce;฀Zamora฀2312฀(INB,฀U),฀Osa.฀ 31. Cremastosperma spec. B -฀Fig.฀35;฀Map฀6 Tree฀or฀shrub฀3-8฀m฀tall,฀3-10฀cm฀diam.;฀young฀twigs฀and฀petioles฀sparsely฀ covered฀ with฀ appressed฀ whitish฀ hairs฀ to฀ 0.3฀ mm฀ long.฀ Leaves:฀ petioles฀ 7-10฀ by฀ 1.5-3฀mm;฀lamina฀elliptic฀to฀obovate,฀or฀narrowly฀so,฀10-28(-34)฀by฀7-10(-12)฀cm฀ (index฀1.9-3.5),฀chartaceous฀to฀coriaceous,฀green฀(or฀greenish฀brown),฀darker฀above,฀ lighter฀or฀more฀brown฀with฀darker฀or฀reddish฀veins฀below,฀glabrous฀above,฀sparsely฀ covered฀with฀appressed฀whitish฀hairs฀to฀0.3฀mm฀long฀particularly฀on฀veins฀below,฀ base฀acute฀to฀obtuse฀(rarely฀rounded),฀apex฀acuminate฀(acumen฀6-20฀mm฀long),฀ primary฀vein฀1.5-3฀mm฀wide฀at฀widest฀point,฀secondary฀veins฀7-10,฀intersecondary฀ veins฀0-1,฀distance฀between฀from฀11-20฀mm฀at฀the฀base฀to฀12-32฀mm฀closer฀to฀ the฀apex,฀angles฀with฀primary฀vein฀from฀70-80°฀at฀the฀base฀to฀40-50°closer฀to฀the฀ apex,฀rarely฀branching,฀mostly฀forming฀distinct฀loops,฀smallest฀distance฀between฀ loops฀and฀margin฀2-5฀mm,฀tertiary฀veins฀mostly฀percurrent.฀Inflorescence฀of฀single฀ (very฀rarely฀branching)฀flowers,฀solitary,฀axillary฀on฀leafy฀or฀leafless฀twigs฀or฀thicker฀ branches;฀peduncles฀(1-)3-5฀by฀1-2฀mm฀(in฀flower),฀3-9฀by฀c.฀2฀mm฀(in฀fruit),฀rather฀ densely฀covered฀with฀appressed฀to฀erect฀whitish฀hairs฀to฀0.2฀mm฀long;฀pedicels฀ 20-45(-68)฀by฀c.฀1฀mm฀at฀the฀base฀(in฀flower),฀22-80฀by฀1-2.5฀mm฀(in฀fruit),฀pink,฀ purple฀or฀reddish฀in฀vivo,฀glabrous;฀2฀lower฀bracts฀of฀unequal฀dimensions,฀basal฀ lower฀bract฀depressed฀ovate,฀c.฀0.5฀by฀1฀mm,฀obtuse,฀sometimes฀persistent,฀densely฀ covered฀with฀appressed฀to฀erect฀whitish฀hairs฀to฀0.2฀mm฀long,฀apical฀lower฀bract฀ elliptic,฀c.฀1.5฀by฀1฀mm,฀obtuse,฀sometimes฀persistent,฀sparsely฀to฀rather฀densely฀ covered฀with฀appressed฀to฀erect฀whitish฀hairs฀to฀0.2฀mm฀long;฀upper฀bract฀attached฀ around฀midway฀along฀pedicel,฀(broadly)฀ovate,฀1.5-2฀by฀c.฀1-1.5฀mm,฀obtuse฀or฀ rounded,฀persistent,฀sparsely฀covered฀with฀appressed฀to฀erect฀whitish฀hairs฀to฀0.2฀mm฀ long;฀flower฀buds฀broadly฀ovoid,฀opening฀in฀development;฀flowers฀green฀maturing฀ to฀yellow฀or฀white฀and฀yellowish฀at฀base฀in฀vivo,฀dark฀to฀yellowish฀light฀brown฀ sometimes฀tinged฀with฀red,฀darker฀at฀the฀base฀in฀sicco,฀sepals฀and฀petals฀glabrous;฀ sepals฀free,฀deltate฀or฀triangular,฀appressed฀to฀recurved,฀c.฀3฀by฀2-3฀mm,฀acute฀or฀ obtuse,฀caducous฀or฀sometimes฀briefly฀persistent;฀outer฀petals฀elliptic,฀14-17฀by฀6-8฀ mm,฀inner฀petals฀elliptic฀to฀narrowly฀so,฀16-17฀by฀5-7฀mm;฀androecium฀c.฀6฀mm฀ diam.,฀connective฀appendage฀to฀0.8฀mm฀wide;฀gynoecium฀c.฀1.5฀mm฀diam.,฀carpels฀ 174 • Chapter 5 a b F i g . 3 5 . Cremastosperma spec. B. a. fruiting and flowering specimen; b. flower (a: Smith, S. et al. 1578; b: Smith, S. et al. 794) Revision and phylogeny of Cremastosperma • 175 glabrous.฀Monocarps฀10-25(-32),฀ellipsoid฀(broadly฀so฀when฀immature),฀slightly฀ asymmetrical,฀8-12฀by฀6-8฀mm,฀green฀maturing฀to฀greenish฀purple฀and฀brown฀in฀ vivo,฀light฀to฀dark฀brown฀or฀blackish฀in฀sicco,฀with฀an฀excentric฀apicule,฀monocarps,฀ stipes,฀and฀receptacle฀glabrous;฀stipes฀8-13฀by฀c.฀2฀mm;฀fruiting฀receptacle฀4-10฀mm฀ diam.฀Seeds฀ellipsoid,฀light฀brown,฀shallowly฀pitted,฀c.฀11฀by฀6฀mm,฀raphe฀sunken,฀ regular. Distribution฀-฀Peru฀(Madre฀de฀Dios฀and฀Cuzco). Habitat฀and฀Ecology฀-฀Primary฀and฀secondary฀moist฀and฀wet฀forest,฀occasionally฀ on฀floodplains.฀At฀elevations฀of฀210-670฀m.฀Flowering:฀August฀-฀December;฀fruiting:฀ January฀-฀May,฀August. Notes฀-฀A฀number฀of฀the฀specimens฀grouped฀under฀Cremastosperma฀spec.฀B฀ have฀in฀the฀past฀been฀identified฀as฀C.฀leiophyllum฀or฀C.฀monospermum,฀two฀well฀ known฀species฀found฀relatively฀nearby฀in฀Bolivia฀and฀widespread฀across฀Bolivia,฀ Brazil,฀and฀Peru฀respectively.฀C.฀spec.฀B฀is฀distinguished฀here฀from฀C.฀leiophyllum฀ (with฀which฀distributions฀do฀not฀overlap)฀on฀the฀basis฀of฀the฀shape฀and฀colour฀ of฀ the฀ monocarps:฀ those฀ of฀ C.฀ leiophyllum฀ dry฀ conspicuously฀ black฀ with฀ a฀ characteristic฀asymmetrical฀shape฀not฀observed฀in฀fruits฀of฀C.฀spec.฀B฀Flower฀buds฀ of฀C.฀monospermum฀remain฀closed,฀with฀a฀distinctive฀triangular฀shape,฀throughout฀ development,฀as฀opposed฀to฀the฀open฀flowers฀of฀C.฀spec.฀B฀In฀the฀area฀of฀Madre฀de฀ Dios฀where฀the฀distributions฀of฀C.฀monospermum฀and฀C.฀spec.฀B฀overlap,฀variation฀in฀ the฀otherwise฀relatively฀consistent฀leaf฀shape฀of฀C.฀monospermum฀appears฀to฀be฀greater,฀ and฀non-flowering฀specimens฀of฀the฀two฀are฀often฀not฀discernable.฀Morphological฀ variation฀in฀specimens฀here฀grouped฀under฀C.฀spec.฀B฀is฀relatively฀wide฀-฀particularly฀ in฀the฀size,฀shape฀and฀texture฀of฀leaves฀and฀length฀of฀pedicel.฀Including฀part,฀or฀the฀ total฀variation,฀of฀C.฀spec.฀B฀within฀that฀of฀either฀C.฀monospermum฀or฀C.฀leiophyllum฀ would,฀in฀the฀opinion฀of฀the฀first฀author,฀neither฀be฀justified฀nor฀helpful฀for฀their฀ recognition.฀However,฀the฀specimens฀currently฀defining฀C.฀spec.฀B฀are฀grouped฀ more฀on฀the฀basis฀of฀the฀absence฀of฀characters฀present฀in฀those฀species฀than฀on฀ characters฀that฀unambiguously฀diagnose฀it฀in฀its฀own฀right.฀Limited฀availability฀of฀ collections,฀particularly฀from฀the฀department฀of฀Cuzco,฀could฀also฀mean฀that฀this฀ grouping฀ represents฀ more฀ than฀ one฀ species.฀ For฀ this฀ reason,฀ this฀ it฀ is฀ described฀ here฀under฀an฀unofficial฀name,฀awaiting฀further฀material฀on฀which฀to฀base฀a฀more฀ satisfactory฀solution. Selection฀of฀other฀specimens฀(30)฀examined: Peru:฀Cuzco:฀Núñez฀12951฀(U,฀USM),฀Quispicanchi;฀Núñez฀et฀al.฀10146฀(MO,฀U),฀ La฀Convención,฀below฀Echarate,฀upper฀Río฀Urubamba,฀Manguyari.฀Madre฀de฀Dios:฀Foster฀ 11495฀(U,฀USM),฀Manu,฀Parque฀Nacional฀Manu.฀Río฀Manu:฀Tayakome;฀Foster฀et฀al.฀3105฀ (F),฀Manu,฀Shintuya,฀1฀km฀up฀small฀stream฀from฀Río฀Alto฀Madre฀de฀Dios;฀Smith,฀S.F.฀et฀ al.1578฀(NY,฀U,฀USM),฀Tambopata,฀Explorer’s฀Inn,฀near฀the฀confluence฀of฀Río฀Tambopata฀ and฀Río฀La฀Torre,฀39฀km฀SW฀of฀Puerto฀Maldonado;฀along฀the฀Big฀Tree฀Trail. 176 • Chapter 5 Excluded species Cremastosperma฀anomalum฀R.E.Fr.฀(1948)฀4,฀pl.฀1c-d.฀-฀Type:฀Killip฀&฀Garcia฀ 33600฀(holo฀S;฀iso฀COL,฀UC,฀US),฀Colombia,฀Chocó:฀Bahia฀Solano,฀near฀Ciudad฀ Mutis,฀along฀Quebrada฀Jella,฀0-75฀m,฀21-23฀Feb.฀1939. ≡ Klarobelia฀anomala฀(R.E.Fr.)฀Chatrou฀(1998)฀123,฀f.฀2. Cremastosperma฀ guianense฀ R.E.Fr.฀ (1934)฀ 205.฀ -฀Type:฀ Davis฀ 102฀ (holo฀ K),฀ Guyana,฀Apoteri:฀Rupununi฀River,฀21฀July฀1931. ≡ Pseudoxandra฀lucida฀R.E.Fr.฀(1937)฀230,฀f.฀3a-e. Cremastosperma฀polyphlebum฀(Diels)฀R.E.Fr.฀(1931)฀331.฀-฀Type:฀Ule฀5628฀(holo฀ B;฀iso฀F,฀G,฀K,฀MG,฀S),฀Brazil,฀Acre:฀Rio฀Juruá-Mirim,฀Aug.฀1901. ≡ Pseudoxandra฀polyphleba฀(Diels)฀R.E.Fr.฀(1937)฀230. Cremastosperma฀williamsii฀R.E.Fr.฀(1934)฀206.฀-฀Type:฀Ll.฀Williams฀3960฀(holo฀ F;฀iso฀S),฀Peru,฀Loreto:฀Yurimaguas,฀Recreo,฀23฀Oct.฀1929. ≡ Pseudoxandra฀williamsii฀(R.E.Fr.)฀R.E.Fr.฀(1937)฀227,฀f.฀2b,c. M a p 2 . Distribution of Cremastosperma antioquense Pirie (●), C. chococola Pirie ( ), C. longipes Pirie (▲), C. novogranatense R.E.Fr. (inverted ▼), C. panamense Maas ( ◆ ), and C. stenophyllum Pirie (■) Revision and phylogeny of Cremastosperma • 177 M a p 3 . Distribution of Cremastosperma awaense Pirie ( ), C. dolichocarpum Pirie (●), C. magdalenae Pirie (■), C. napoense Pirie (▲), C. pacificum R.E.Fr. ( ), and C. westrae Pirie (◆) M a p 4 . Distribution of Cremastosperma brevipes (DC.) R.E.Fr. (●); C. gracilipes R.E.Fr. ( ); C. macrocarpum Maas (■); C. monospermum (Rusby) R.E.Fr. (▲) 178 • Chapter 5 M a p 5 . Distribution of Cremastosperma bullatum Pirie (●); C. cauliflorum R.E.Fr. (▲; C. cenepense Pirie & Zapata (✚); C. longicuspe R.E.Fr. ( ) M a p 6 . Distribution of Cremastosperma leiophyllum R.E.Fr. (▲); C.oblongum R.E.Fr. (●); C. spec. B (◆) Revision and phylogeny of Cremastosperma • 179 M a p 7 . Distribution of Cremastosperma megalophyllum R.E.Fr. (▲); C. microcarpum R.E.Fr. ( ); C. pedunculatum (Diels) R.E.Fr. (●); C. peruvianum R.E.Fr. ( ) M a p 8 . Distribution of Cremastosperma pendulum (Ruiz & Pav.) R.E.Fr. (■); C. yamayakatense Pirie (●) 180 • Chapter 5 M a p 9 . Distribution of C. macrocarpum Maas ( ); C. venezuelanum Pirie (▲) Acknowledgements The฀ herbarium฀ curators฀ of฀ AAU,฀ B,฀ COL,฀ F,฀ G,฀ GH,฀ HOXA,฀ HUA,฀ K,฀ QCNE,฀LPB,฀MICH,฀MO,฀MOL,฀NY,฀S,฀U,฀US,฀USM,฀USZ,฀VEN,฀and฀WIS฀are฀ acknowledged฀for฀the฀loan฀of฀or฀access฀to฀collections.฀We฀are฀especially฀grateful฀ to฀the฀curators฀of฀MO,฀NY,฀US,฀and฀WAG฀for฀allowing฀destructive฀sampling฀of฀ loaned฀specimens.฀Two฀field฀expeditions฀were฀made฀possible฀by฀grants฀from฀NWO฀ (grant฀no.฀R฀85-351,฀Bolivia฀2001)฀and฀the฀Miquel฀Fonds฀(MDP,฀Peru฀2003)฀and฀ the฀Miquel฀Fonds฀and฀Alberta฀Mennega฀Stichting฀(Marleen฀Botermans,฀Robin฀ van฀Velzen)฀(Peru฀2003).฀Study฀of฀specimens฀at฀the฀Swedish฀Museum฀of฀Natural฀ History฀was฀supported฀by฀grant฀from฀the฀High฀Lat฀programme,฀which฀was฀made฀ available฀by฀the฀European฀Community฀-฀Access฀to฀Research฀Infrastructure฀action฀ of฀the฀Improving฀Human฀Potential฀Programme฀(MDP).฀Maps฀were฀produced฀using฀ ESRI฀data฀made฀available฀by฀the฀New฀York฀Botanical฀Garden’s฀‘Digital฀Basemap฀of฀ the฀Americas’.฀The฀majority฀of฀the฀line฀drawings฀were฀made฀by฀Hendrik฀Rypkema,฀ with฀the฀exceptions฀of฀Fig.฀1฀(P.฀Pardoen),฀Fig.฀2฀(T.฀Schipper),฀Fig.฀10฀(Mario฀ Zapata฀Cruz),฀and฀Fig.฀27,฀originally฀published฀in฀Fries฀(1930),฀and฀reproduced฀ here฀with฀the฀kind฀permission฀of฀the฀Bergius฀Foundation฀at฀the฀Royal฀Swedish฀ Academy฀of฀Sciences.฀Thanks฀to฀Paddy฀Haripersaud,฀Lubbert฀Westra,฀and฀Hans฀ter฀ Steege฀for฀help฀with฀photography฀of฀herbarium฀specimens.฀Lars฀Chatrou’s฀criticism฀ of฀the฀manuscript฀is฀also฀gratefully฀acknowledged.฀ Revision and phylogeny of Cremastosperma • 181 Identification list Collections฀are฀identified฀by฀the฀first฀collector฀and฀collector฀number฀only.฀The฀ abbreviations฀behind฀the฀collector฀numbers฀refer฀to฀the฀following฀taxa: Cremastosperma฀ ant฀ =฀ awa฀ =฀ bre฀ =฀ bul฀ =฀ cau฀ =฀ cen฀ =฀ cho฀ =฀ dol฀ =฀ gra฀ =฀ lei฀ =฀ lon฀ =฀ lonp฀ =฀ mac฀ =฀ mag฀ =฀ meg฀ =฀ ฀ antioquense฀฀ awaense฀฀ brevipes฀฀ bullatum฀฀ cauliflorum฀฀ cenepense฀฀ chococola฀฀ dolichocarpum฀฀ gracilipes฀฀ leiophyllum฀฀ longicuspe฀฀ longipes฀฀ macrocarpum฀฀ magdalenae฀฀ megalophyllum฀฀ mic฀ mon฀ nap฀ nov฀ obl฀ pac฀ pan฀ ped฀ pen฀ per฀ ste฀ ven฀ wes฀ yam฀ spA฀ spB฀ =฀ =฀ =฀ =฀ =฀ =฀ =฀ =฀ =฀ =฀ =฀ =฀ =฀ =฀ =฀ =฀ microcarpum฀ monospermum฀ napoense฀ novogranatense฀ oblongum฀ pacificum฀ panamense฀ pedunculatum฀ pendulum฀ peruvianum฀ stenophyllum฀ venezuelanum฀ westrae฀ yamayakatense฀ spec.฀A฀ spec.฀B฀ Acevedo-Rodrigues฀7512:฀meg;฀7523:฀meg;฀7543:฀gra;฀8635:฀spB;฀9132฀spB฀ —฀Acosta฀ Solís฀ 6429:฀ ste฀ —฀Aguilar฀ 467:฀ spA;฀ 691:฀ mon;฀ 764:฀ mon;฀ 905:฀ spB?;฀ 4371:฀spA฀—฀Alcorn฀11:฀lon?฀—฀Alvarado฀228:฀nap;฀267:฀nap;฀298:฀meg;฀427:฀nap฀ —฀Alvarez฀2405:฀gra฀—฀Ancuash฀262:฀per;฀1324:฀yam;฀1517:฀per฀—฀Arrazola฀134:฀lei฀ —฀Aulestia,฀M.฀637:฀awa;฀842:฀awa;฀1184:฀awa;฀1513:฀gra;฀1726:฀gra;฀2127:฀gra;฀3238:฀ cau;฀3395:฀cau:฀3520:฀gra฀—฀Ayala฀15:฀cau;฀2432:฀meg;฀3485:฀mic. Balslev฀2418:฀gra;฀4634:฀gra;฀62310:฀gra฀—฀Barbour฀4432:฀yam;฀4851:฀spB;฀ 4856:฀spB;฀5111:฀spB฀—฀Barclay฀4755:฀gra฀—฀Barrier฀5166:฀bre฀—฀Beck฀1675:฀lei;฀ 8234:฀mon;฀10092:฀mon;฀12221:฀mon;฀16540:฀mon;฀19323:฀mon฀—฀Beltran฀3245:฀ spB;฀3580:฀obl;฀5488:฀meg;฀5782:฀mic฀—฀Berlin฀218:฀per;฀1588:฀yam;฀1950:฀per;฀2079:฀ per฀—-฀Betancur฀3309฀(changed฀on฀label฀to฀33011):฀dol?;฀6043:฀awa฀—฀Boom฀ 5039:฀mon;฀10812:฀bre฀—฀Bourdy฀GB1743:฀mon;฀GB1828:฀mon฀—฀Brandbyge฀ 30017:฀meg;฀30472:฀gra;฀31854:฀cau;฀32225:฀cau;฀32566:฀meg;฀32589:฀gra;฀32681:฀ gra;฀32685:฀gra;฀33337:฀gra;฀33434:฀gra;฀33503:฀cau฀—฀Buchtien฀705:฀lei;฀706:฀lei฀ —฀Bulnes฀502:฀obl. Callejas฀3110:฀dol฀—฀Camp฀E1311:฀ped฀—฀Campbell฀9026:฀mon฀—฀Campos฀ 4160:฀ ped;฀ 4213:฀ ped;฀ 4268:฀ ped฀ —฀ Cárdenas฀ 2899:฀ mag฀ —฀ Cazalet฀ 7528:฀ cau฀ —฀Cerón฀404:฀meg;฀612:฀meg;฀702:฀gra;฀1268:฀meg;฀1353:฀gra;฀2321:฀gra;฀2361:฀gra;฀ 2378:฀meg;฀2428:฀gra;฀2483:฀meg;฀2500:฀meg;฀2675:฀gra;฀2986:฀nap;฀3009:฀gra;฀3031:฀ meg;฀3036:฀meg;฀3037:฀gra;฀3125:฀meg;฀3939:฀gra;฀3949:฀gra;฀4159:฀meg;฀5358:฀gra;฀ 182 • Chapter 5 5436:฀gra;฀5983:฀gra;฀6663:฀nap;฀9324:฀meg฀—-฀Chatrou฀103:฀spA;฀208:฀mic;฀224:฀ cau;฀227:฀mic;฀233:฀cau;฀238:฀cau;฀259:฀meg;฀265:฀meg;฀267:฀gra;฀268:฀meg฀—฀Chávez฀ 70:฀yam฀—฀Cheta฀6/173:฀cau฀—฀Cid฀Ferreira฀2724:฀mon;฀2868:฀mon;฀4507:฀mon;฀ 4556:฀mon;฀4590:฀mon;฀4673:฀mon;฀5262:฀cau;฀6301:฀mon;฀9959:฀meg;฀10019:฀cau;฀ 10120A:฀mon;฀10286:฀mon฀—฀CID-JH฀6-235:฀cau฀—฀Clark฀3158:฀meg฀—฀Coello฀ 60:฀ gra฀ —฀ Cogollo฀ 6039:฀ dol;฀ 6040:฀ dol;฀ 6044:฀ dol;฀ 6052:฀ dol;฀ 6056:฀ dol;฀ 6345:฀ dol;฀7869:฀dol฀—฀Cornejo,฀F.฀2276:฀mon;฀2546฀spB?฀—฀Cornejo,฀S.X.฀4493:฀gra฀ —฀Cornejo,฀X.฀6451:฀awa;฀6702:฀ped฀—฀Cremers฀15331:฀bre฀—฀Croat฀18494:฀cau฀ —฀Cuatrecasas฀10575:฀meg;฀10726:฀cau;฀11125:฀cau;฀11331:฀meg;฀17463:฀pac;฀17573:฀ nov;฀21288:฀nov;฀26031:฀nov. Dahlgren฀162:฀mon฀—฀Daly฀5764:฀mic;฀6186:฀meg;฀6498:฀mon;฀6573:฀mon;฀ 6630:฀mon;฀6686:฀mon;฀6909:฀mon;฀7347:฀mon;฀7450:฀mon;฀8043:฀mon;฀11719:฀mon฀ —฀Davidse฀23634:฀pan฀—฀Davidson฀9869:฀mic฀—฀Del฀Carpio฀1833:฀mic;฀1923:฀ mic;฀1942:฀mic;฀2213:฀mic;฀8009:฀mon฀—฀Devia฀2875:฀nov?;฀3762:฀pac;฀4065:฀pac;฀ 5335:฀nov฀—฀De฀Nevers฀3601:฀pan;฀3906:฀pan;฀4129:฀pan;฀4475:฀wes;฀4571:฀pan;฀ 4898:฀pan;฀5189:฀pan;฀5221:฀pan;฀5260:฀pan;฀6171:฀pan;฀6513:฀pan฀—฀DeWalt฀662:฀ mon฀—฀Díaz,฀S.฀825:฀mon;฀1229:฀cau;฀1411:฀gra;฀6851:฀yam;฀8225:฀per;฀8375:฀yam;฀ 8899:฀mon;฀9563:฀mon฀—฀Díaz,฀W.฀231:฀ven;฀274:฀mac฀—฀Dik฀323:฀meg;฀1732:฀gra฀ —฀Ducke฀RB19620:฀meg฀—฀Dwyer฀3314:฀pan;฀10248:฀pan. Edwards฀397:฀ven฀—฀Escobar฀3309:฀mag฀—฀Espina฀3188:฀cho฀—฀Espinoza฀ 129:฀meg;฀399:฀cau;฀578:฀gra. Faber-Langendoen฀476:฀nov;฀1208:฀pac฀—฀Fernández฀8872:฀lonp฀—฀Feuillet฀ 10239:฀bre;฀10337:฀bre฀—฀Figuieredo฀853:฀mon;฀906:฀mon฀—฀Flora฀Falcón฀937฀ (=฀Van฀der฀Werff฀937):฀mac฀—฀Flores฀165:฀obl:฀1241:฀obl฀—฀Folsom฀6155:฀pan฀ —฀Forero฀6576:฀lonp฀—฀Forget฀305:฀bre฀—฀Foster฀2477:฀spB?;฀2512:฀spB;฀3105:฀ spB;฀3128:฀spB;฀3157:฀spB;฀3418:฀spB?;฀3502:฀spB?;฀5004:฀spB?;฀5083:฀spB;฀5292:฀spB?;฀ 6026:฀spB?;฀6589:฀spB;฀8235:฀spB?;฀9482:฀pen;฀10008:฀pen;฀10224:฀obl;฀11495:฀spB;฀ 12549:฀mon;฀13422:฀mon฀—฀Freire฀407:฀meg;฀468:฀nap฀—฀Freitas฀8:฀mic฀—฀Fróes฀ 23962:฀cau. Galdames฀1162A:฀pan;฀1585:฀pan฀—฀García฀390:฀cho฀—฀Gentry฀5070:฀pan;฀ 7286:฀cho;฀8828:฀pan;฀8910:฀mon;฀21928:฀gra;฀25111:฀mic;฀29126:฀cau;฀32009:฀spB;฀ 32153:฀mic;฀35298:฀pac;฀35562:฀pac;฀37169:฀cau;฀38085:฀mic;฀38127:฀mic;฀41879:฀obl;฀ 42985:฀mic;฀43174:฀mic;฀43776:฀mic;฀45355:฀ped;฀45371:฀ped;฀45481:฀ped;฀45510:฀ ped;฀52233:฀mon;฀56210:฀mic;฀57078:฀nov;฀58434:฀mon;฀60032:฀gra;฀60865:฀cau;฀ 65601:฀cau;฀68663:฀mon;฀70042:฀awa;฀72190:฀mic;฀72235:฀cau;฀72995:฀lonp;฀76343:฀ cau;฀76357:฀mic;฀78657:฀spA;฀79039:฀dol;฀80543:฀ped;฀80904:฀ped฀—฀GHS฀703:฀obl฀ —฀Gómez-Pompa฀3067:฀pan฀—฀Gonzales฀53:฀mon฀—฀Graham฀635:฀pen;฀1713:฀ mon฀—฀Grández฀222:฀mic;฀491:฀mic;฀1591:฀cau;฀2032:฀mic;฀2488:฀gra฀-฀de฀Granville฀ 390:฀bre;฀B5186:฀bre;฀5670:฀bre;฀5672:฀bre;฀7555:฀bre;฀7672:฀bre;฀7799:฀bre;฀8714:฀bre;฀ 8876:฀bre฀—฀Grenand฀1509:฀bre฀—฀Grijalva฀305:฀meg฀—฀Gudiño฀310:฀meg;฀1123:฀ meg;฀1787:฀gra. Revision and phylogeny of Cremastosperma • 183 Harling฀17581:฀gra฀—฀Hedin฀57:฀spB?;฀74:฀spB?;฀94:฀mic;฀96:฀mic฀—฀Helme฀ 742:฀mon฀—฀Hequet฀432:฀bre฀—฀Hernández฀251:฀mag฀-฀Herrera,฀G.฀3972:฀spA;฀ Herrera,฀H.฀1122:฀wes;฀1231:฀pan฀—฀Hinojosa฀1313:฀lei฀—฀Hodges฀111:฀yam;฀112:฀ yam฀—฀Holm-Nielsen฀903:฀meg;฀21501:฀cau;฀22082:฀meg฀—฀Hurtado฀160:฀nap;฀ 186:฀nap;฀322:฀nap;฀896:฀nap;฀941:฀nap;฀1385:฀meg;฀2096:฀nap;฀2430:฀meg;฀2445:฀meg;฀ 2448:฀meg;฀3005:฀meg;฀3019:฀gra;฀3156:฀meg. Irvine฀374:฀meg;฀734:฀meg฀—฀Irwin฀47562:฀bre. Jacquemin฀2349:฀bre฀—฀Jaramillo,฀J.฀1420:฀ped;฀3533:฀meg;฀3577:฀gra;฀8352:฀ meg;฀8364:฀meg;฀12840:฀nap;฀13489:฀ped;฀30773:฀gra฀—฀Jaramillo,฀N.฀403:฀yam;฀ 584:฀yam;฀686:฀cau;฀918:฀yam;฀942:฀bul;฀972:฀bul;฀1110:฀yam;฀1248:฀yam;฀1409:฀yam฀ —฀ Jardim฀ 597:฀ mon;฀ 759:฀ mon;฀ 2408:฀ mon฀ —฀ Johnston฀ 1555:฀ pan;฀ 1812:฀ pan฀ —฀Juncosa฀736:฀mag. Kanehira฀215:฀ped?฀—฀Kayap฀33:฀yam;฀631:฀per;฀1078:฀cen฀—฀Kernan฀1224:฀ spA฀—฀Killeen฀3080:฀lei;฀3085:฀lei;฀3110:฀lei;฀3238:฀lei฀—฀Killip฀23622:฀obl;฀27902:฀ mon;฀28537:฀lon;฀28961:฀mon;฀29020:฀lon;฀37752:฀ven฀—฀Klug฀902:฀cau;฀3069:฀meg;฀ 3726:฀ped฀—฀Knapp฀6159:฀ste;฀6354:฀mon?;฀6450:฀spB;฀6588:฀ped;฀7178:฀lon?;฀7183:฀ lon?;฀7215:฀lon;฀7645:฀per;฀7864:฀lon?;฀8203:฀lon฀—฀Krukoff฀4697:฀mon;฀4748:฀mic;฀ 4975:฀mic;฀6151:฀mic;฀10556:฀lei฀—฀Kuhlmann฀1426:฀lon฀—฀Kvist฀892:฀mic;฀1127:฀ mic;฀1265:฀mic;฀1355:฀mic. Lawesson฀39560:฀gra;฀39598:฀meg฀—฀Lescure฀549:฀bre;฀2200:฀meg฀—฀Lewis,฀ M.฀37979:฀lei฀—฀Lewis,฀W.H.฀11166:฀lon;฀11831:฀lon;฀12186:฀lon฀—฀Liesner฀682:฀ wes;฀1362:฀pan;฀9763:฀mac฀—฀Lima฀582:฀mon฀—฀Lleras฀P16879:฀cau฀—฀Lowrie฀463:฀ mon฀—฀Luteyn฀4890:฀gra;฀8630:฀gra;฀9057:฀gra. Maas฀4592:฀obl;฀6271:฀cau;฀6281:฀mic;฀8064:฀bre;฀8222:฀mic;฀8289:฀mic;฀8300:฀ mic;฀8577:฀meg;฀8595:฀meg;฀9029:฀cau;฀9148:฀obl;฀9251:฀cau;฀9328:฀bre;฀9559:฀pan;฀ P13079:฀mon฀-฀Martin฀s.n.฀bre฀—฀Mathias฀5510:฀mic;฀6007:฀mon฀—฀McDaniel฀ 2570:฀pen;฀17020:฀mic;฀20264:฀mic;฀20677:฀mic;฀20761:฀cau฀—฀McDonagh฀181:฀ pan;฀291:฀wes฀—฀McPherson฀9957:฀pan;฀11046:฀pan;฀14039:฀pan฀—฀Meneces฀395:฀ lei฀—฀Miller฀794:฀meg;฀947:฀pan฀—฀Mitchell฀26-87:฀mon฀—฀Monsalve฀1816:฀nov฀ —฀Monteagudo฀4933:฀obl฀—฀Morawetz฀13-9888:฀pen;฀110-25985:฀obl฀—฀Moretti฀ 12:฀bre฀—฀Mori฀2906:฀pan;฀2918:฀pan;฀5527:฀pan;฀6404:฀pan;฀8758:฀bre;฀15079a:฀bre;฀ 15583:฀bre;฀22721:฀bre;฀22744:฀bre฀—฀Murillo฀565:฀cau฀—฀Murray฀1520:฀pan. Naessany฀106:฀lei฀—฀Nee฀34402:฀mon;฀34992:฀mon;฀35033:฀mon;฀38126:฀lei;฀ 38135:฀lei;฀40971:฀lei;฀41022:฀lei฀—฀Neill฀6359:฀meg;฀7195:฀gra;฀7210:฀meg;฀7568:฀ gra;฀7649:฀nap;฀7883:฀meg;฀8089:฀nap;฀8466:฀meg;฀9088:฀nap;฀9118:฀gra;฀9180:฀nap;฀ 9334:฀lei;฀9589:฀ped;฀9967:฀gra;฀10005:฀meg;฀11048:฀nap;฀11065:฀ped?;฀11095:฀gra;฀ 13201:฀ped?฀—฀Nelson฀680:฀mon;฀763:฀mon฀—฀Núñez฀10146:฀spB;฀10589:฀mon;฀ 10785:฀mon;฀11068:฀mon;฀11202:฀mon;฀12152:฀mon;฀12446:฀spB;฀12951:฀spB;฀16303:฀ mon;฀16898:฀obl;฀19079:฀spB;฀19232:฀spB;฀19240:฀spB;฀19495:฀spB;฀21737:฀spB. Oldeman฀T19:฀ bre;฀ B634:฀ bre;฀ B636:฀ bre;฀ B2165:฀ bre฀ —฀ Oliver฀ 3681:฀ wes฀ —฀Öllgaard฀35223:฀gra;฀38866:฀gra;฀99056:฀gra฀—฀Ownbey฀2715:฀nap. 184 • Chapter 5 Palacios฀783:฀gra;฀1620:฀meg;฀1651:฀gra;฀2212:฀nap;฀2311:฀meg;฀2416:฀gra;฀2492:฀ gra;฀2494:฀meg;฀2871:฀meg;฀2937:฀meg;฀2992:฀gra;฀3270:฀meg;฀3276:฀meg;฀4150:฀nap;฀ 4215:฀meg;฀4261:฀meg;฀4323:฀meg;฀4759:฀gra;฀4902:฀meg;฀5123:฀meg;฀5485:฀nap;฀ 7043:฀gra;฀7602:฀cau;฀8043:฀cau;฀8646:฀ped;฀8925:฀meg;฀10278:฀meg;฀10462:฀gra;฀ 10663:฀meg;฀11016:฀meg฀—฀Pariona฀983:฀obl฀—฀Pennington฀12266:฀nap฀—฀Pérez฀ 832:฀pan฀—฀Phillips฀545:฀mon฀—฀Pinkley฀558:฀meg฀—฀Pipoly฀12331:฀mic;฀12982:฀ cau;฀14155:฀meg;฀16951:฀dol;฀17006:฀dol;฀17073:฀dol฀—฀Pirie฀2:฀lei;฀3:฀lei;฀4:฀mon;฀5:฀ mon;฀7:฀obl;฀33:฀pen;฀35:฀pen;฀54:฀pen;฀57:฀yam;฀58:฀yam;฀59:฀yam;฀60:฀yam;฀66:฀bul;฀68:฀ bul;฀71:฀bul;฀80:฀yam;฀94:฀bul฀—฀Pitman฀195:฀gra฀—฀Plowman฀2551:฀mic;฀7248:฀gra฀ —฀Poeppig฀1750:฀pen;฀2091:฀lon฀—฀Prance฀3527:฀mic;฀3579:฀mic;฀5354:฀mon;฀7885:฀ mon;฀13079฀(=฀Maas฀P13079):฀mon;฀13464:฀mic;฀23783:฀cau;฀24094:฀cau฀—฀Prévost฀ 2284:฀bre;฀2310:฀bre;฀3446:฀bre. Quelal฀191:฀awa฀—฀Quipuscoa฀2324:฀lon;฀2361:฀lon. Raffauf฀102:฀cau฀—฀Ramirez฀4881:฀ped?;฀9517:฀mon฀—฀Revelo฀264:฀meg฀ —฀Revilla฀2265:฀cau;฀3327:฀mic฀—฀Riéra฀668:฀bre;฀685:฀bre;฀1564:฀bre฀—฀Rimachi฀ 1027:฀meg;฀2397:฀cau;฀7588:฀cau;฀7597:฀obl;฀10637:฀lon฀—฀Rodriguez฀724:฀yam;฀ 1112:฀per;฀1152:฀bul;฀1758:฀ped฀—฀Rojas฀0255:฀cen;฀0269:฀cen;฀666:฀mon;฀1014:฀ped฀ —฀Romero-Castañeda฀3369:฀awa฀—฀Romoleroux฀1834:฀gra;฀1883:฀gra฀—฀Rosa฀ 2426:฀mon฀—฀Rubio฀320:฀gra;฀812:฀cau;฀2181:฀awa;฀2397:฀nap฀—฀Rudas฀2260:฀mic;฀ 4514:฀cau฀—฀Rueda฀952:฀mon฀—฀Ruiz,฀H.฀19-28:฀lei;฀19117:฀pen฀—฀Ruíz฀M.,฀J.฀ 1748:฀lon?฀—฀Ruiz,฀M.฀8712:฀mic฀—฀Ruíz฀Z.,฀T.฀3499:฀mac. Sabatier฀486:฀bre;฀1486:฀bre฀—฀Saldías฀530:฀lei฀—฀Sánchez฀323:฀pac;฀415:฀dol฀ —฀Sastre฀503:฀meg;฀5670:฀bre;฀5771:฀bre฀—฀Scharf฀76:฀bre;฀77:฀bre;฀78:฀bre฀—฀Schatz฀ 1002:฀spA;฀1077:฀pan฀-฀Schunke,฀C.฀200:฀cau฀—฀Schunke฀V.,฀J.฀2554:฀mon;฀3320:฀ mon?;฀3828:฀mon?;฀4068:฀mon?;฀5645:฀obl;฀5829:฀obl;฀6412:฀mic;฀7251:฀ped;฀7591:฀ mon?;฀8484:฀mon?;฀10907:฀obl฀—฀Scolnik฀581:฀lei฀—฀Seidel฀2003:฀lei;฀2265:฀lei;฀ 2584:฀mon;฀2674:฀mon;฀2878:฀lei;฀2893:฀lei;฀3543:฀lei;฀4531:฀mon;฀4660:฀mon;฀7234:฀ lei;฀7280:฀mon;฀7290:฀lei฀-฀Silva,฀M.G.฀5398:฀mon฀-฀Silva,฀M.N.฀322:฀mon;฀354:฀mon฀ —฀Silveira฀767:฀mon;฀1464:฀mon฀—฀Smith,฀D.N.฀2031:฀obl;฀2385:฀pen;฀4611:฀ped;฀ 4633:฀ped;฀6613:฀obl;฀6850:฀obl;฀13253:฀lei;฀13834:฀mon;฀14006:฀lei;฀14058:฀lei;฀14148:฀ lei;฀14216:฀mon฀—฀Smith,฀S.F.฀137:฀spB?;฀504:฀spB;฀654:฀spB;฀794:฀spB;฀1577:฀spB;฀ 1578:฀spB฀—฀Soejarto฀1275:฀meg;฀2798:฀ant;฀3586:฀ant฀—฀Solomon฀3523:฀mic;฀6606:฀ lei฀—฀Sperling฀5784:฀mon;฀6198:฀mon฀—฀Sprague฀351:฀meg฀—฀Stahl฀2931:฀ped฀ —฀Stein฀2577:฀meg;฀3036:฀gra฀—฀Steyermark฀94314:฀ven;฀123804:฀mac฀—฀Sugden฀ 613:฀wes฀—฀Sytsma฀2884฀pan. Teixeira฀480:฀mon;฀520:฀mon฀—฀Tessmann฀4176:฀per;฀4748:฀gra฀—฀Timaná฀ 1010:฀spB;฀1537:฀mon;฀1563:฀mon;฀1593:฀mon;฀1617:฀mon;฀1632:฀mon;฀1673:฀mon;฀ 1799:฀mon;฀1822:฀mon;฀1824:฀mon;฀3219:฀mon฀—฀Tipaz฀1428:฀awa;฀1718:฀awa;฀ 1912:฀awa฀—฀Tirado฀591:฀awa;฀1083:฀awa฀—฀Tostain฀33:฀bre฀—฀Tredwell฀22:฀mic฀ —฀Tunqui฀573:฀yam฀—฀Tyson฀3314:฀pan. Valencia฀ 67352:฀ meg;฀ 68807:฀ meg฀ —฀Van฀ der฀Werff฀ 937:฀ mac;฀ 10161:฀ lon;฀ 12045:฀awa;฀13001:฀ped?;฀13022:฀ped;฀13838:฀cau฀—฀Vargas฀C.,฀C.฀1113:฀lei;฀1193:฀lei;฀ Revision and phylogeny of Cremastosperma • 185 1194:฀lei฀—฀Vargas,฀H.฀671:฀meg;฀1025:฀meg฀—฀Vásquez฀1938:฀cau;฀1969:฀meg;฀2994:฀ meg;฀4423:฀gra;฀4559:฀lon;฀6932:฀mon;฀7237:฀meg;฀8010:฀mic;฀8100:฀mic;฀8425:฀cau;฀ 9350:฀mic;฀10088:฀mic;฀10637:฀cau;฀11192:฀meg;฀11423:฀cau;฀11438:฀cau;฀12295:฀mic;฀ 12358:฀cau;฀12655:฀cau;฀12675:฀gra;฀12829:฀mon;฀12862:฀mic;฀13007:฀gra;฀13552:฀mic;฀ 13690:฀mic;฀16885:฀meg;฀17376:฀gra;฀18236:฀cau;฀18990:฀per;฀19055:฀yam;฀19322:฀ spB;฀24428:฀meg?;฀24891:฀bul;฀25605:฀spB฀—฀Villa฀1499:฀meg฀—฀Villiers฀2639:฀bre. Wallnöfer฀112-7488:฀mon;฀13477:฀bre;฀13499:฀bre;฀25-19588:฀mon;฀117-28988:฀ obl฀—฀Warush฀Juwa฀RBAE119:฀cau฀—฀Weberbauer฀4558:฀ped฀—฀White฀913:฀lei฀ —฀Whitmore฀717:฀meg;฀737:฀gra;฀854:฀meg;฀871:฀gra฀—฀Williams,฀Ll.฀670:฀mon;฀ 2197:฀gra;฀4092:฀lon;฀4919:฀mon;฀5085:฀mon;฀5287:฀mon;฀5296:฀mon;฀7423:฀obl฀ —฀Wingfield฀6751:฀mac฀—฀Woytkowski฀7128:฀lon;฀7592:฀meg. Young,฀H.J.฀219:฀spB฀—฀Young,฀K.฀193:฀mon. Zak฀4202:฀meg;฀5272:฀cau฀—฀Zamora฀444:฀meg;฀518:฀meg;฀581:฀gra;฀657:฀meg;฀ 671:฀meg;฀2312:฀spA฀—฀Zuleta฀175:฀meg. 186 • Chapter 5 Index to scientific names Accepted฀taxa฀are฀in฀roman฀type,฀and฀synonyms฀in฀italics.฀Numbers฀refer฀to฀ the฀species฀number฀as฀used฀in฀this฀revision,฀excluded฀names฀are฀indicated฀by฀excl,฀ nomina฀dubia฀by฀nom.dub. Aberemoa฀pedunculata฀Diels฀23 Annona฀nitida฀Ruiz฀&฀Pav.฀6 Cremastosperma฀anomalum฀R.E.Fr.,฀excl. ฀ antioquense฀Pirie฀1 ฀ awaense฀Pirie฀2฀ ฀ brevipes฀(DC.)฀R.E.Fr.฀3 ฀ bullatum฀Pirie฀4 ฀ cauliflorum฀R.E.Fr.฀5 ฀ cenepense฀Pirie฀&฀Zapata฀6 ฀ chococola฀Pirie฀7 ฀ dolichocarpum฀Pirie฀8 ฀ gracilipes฀R.E.Fr.฀9 ฀ guianense฀R.E.Fr.,฀excl. ฀ juruense฀R.E.Fr.฀17 ฀ killipii฀R.E.Fr.฀11 ฀ leiophyllum฀R.E.Fr.฀10 ฀ longicuspe฀R.E.Fr.฀11 ฀ longipes฀Pirie฀12 ฀ macrocarpum฀Maas฀13 ฀ magdalenae฀Pirie฀14 ฀ megalophyllum฀R.E.Fr.฀15 ฀ microcarpum฀R.E.Fr.฀16 ฀ monospermum฀(Rusby)฀R.E.Fr.฀17 ฀ monospermum฀(Rusby)฀R.E.Fr.฀var.฀brachypodum฀R.E.Fr.฀17 ฀ napoense฀Pirie฀18 ฀ novogranatense฀R.E.Fr.฀19 ฀ oblongum฀R.E.Fr.฀20 ฀ pacificum฀R.E.Fr.฀21 ฀ panamense฀Maas฀22 ฀ pedunculatum฀(Diels)฀R.E.Fr.฀23 ฀ pendulum฀(Ruiz฀&฀Pav.)฀R.E.Fr.฀24 ฀ peruvianum฀R.E.Fr.฀25 ฀ poiteaui฀(Diels)฀R.E.Fr.฀3 ฀ polyphlebum฀(Diels)฀R.E.Fr.,฀excl. ฀ stenophyllum฀Pirie฀26 ฀ venezuelanum฀Pirie฀27 Revision and phylogeny of Cremastosperma • 187 ฀ westrae฀Pirie฀28 ฀ williamsii฀R.E.Fr.,฀excl. ฀ yamayakatense฀Pirie฀29 ฀ spec.฀A฀30 ฀ spec.฀B฀31 Cymbopetalum฀monospermum฀Rusby฀17 Klarobelia฀anomala฀(R.E.Fr.)฀Chatrou,฀excl. Guatteria฀brevipes฀DC.฀3 ฀ leiophylla฀Diels฀10 ฀ lucida฀Rusby฀10 ฀ pendula฀Ruiz฀&฀Pav.฀24 ฀ poiteaui฀Diels฀3 ฀ rusbyi฀J.F.Macbr.฀10 ฀ socialis฀J.F.Macbr.,฀nom.dub. Pseudoxandra฀lucida฀R.E.Fr.,฀excl. ฀ polyphleba฀(Diels)฀R.E.Fr.,฀excl. ฀ williamsii฀(R.E.Fr.)฀R.E.Fr.,฀excl. 188 • Chapter 5 Summary This thesis presents results of molecular systematic and biogeographic studies of groups in the flowering plant family Annonaceae, and a taxonomic revision of the genus Cremastosperma, which occurs in the Neotropics. In Chapter 2 preliminary results are presented comparing the timing of diversifications in four predominantly Neotropical genera of Annonaceae; Cremastosperma, Duguetia, Guatteria, and Mosannona. Explanations were sought for the disparity in numbers of species in these four genera. Phylogeny reconstruction was used to assess their monophyly and molecular dating techniques (applying the nonparametric rate-smoothing method) used to arrive at preliminary estimates of the relative ages of their most recent common ancestors (MRCAs). The effects of taxon and character sampling on date estimates in these genera were assessed and compared. Higher sampling of crown group taxa of the species-rich genus Guatteria resulted in significantly older age estimation for its MRCA. This represents a potentially serious bias in a widely used molecular dating method, which in this case made further comparison of species richness in the four genera meaningless. Molecular dating techniques should be assessed for sensitivity to levels of taxon sampling under differing conditions. In Chapter 3 a South American-centred clade was identified (the SAC clade). The SAC clade comprises all the short-branch clade (SBC) genera distributed in South America and mostly only to a limited extent into Central America, but not those endemic to Asia and Central America. In the absence of a fully resolved phylogeny of the SAC clade, this was interpreted to suggest a common origin in South America. The age estimations (produced using three different rate smoothing methods) supported the hypothesis that the SAC clade originated in South America by dispersal across the Boreotropics. Gentry’s hypothesis of a more ancient origin by Gondwanan vicariance was rejected, as was the possibility of later dispersal from Africa. Monophyly was confirmed in Cremastosperma, Malmea, and Mosannona. The monotypic Summary • 189 genus Pseudephedranthus was found to be nested within Klarobelia, the species of which otherwise formed a monophyletic group. The conclusion was drawn that the Andean-centred distribution patterns as observed are not the arbitrary result of the definition of poly- or paraphyletic groups. The ages estimated for the MRCAs of each clade were not significantly different from each other. Although the strength of this test was limited by imprecision in the molecular dating results, these ages appeared to fall within the time frame of the orogeny of the Northern Andes. To test these hypotheses further, species level phylogenies of Cremastosperma, Klarobelia, Mosannona and Malmea are needed. Additional data should also be sought to test the age calibration of the Annonaceae and wider Magnoliales phylogeny. This approach could shed further light on the dynamic processes of the recent invasion of Central America, and the origin of high species diversity in tropical America. In Chapter 4 an ancient paralogue of the widely used chloroplast marker tr nL-F was discovered. The divergence of the paralogous copies was inferred to have taken place in a common ancestor of the Annonaceae. Primers were developed to preferentially amplify and sequence the different paralogues, and the resulting sequences were compared in order to assess their functional homology and phylogenetic utility. Pseudtr nL-F (the paralogue not normally sequenced) evolves at a faster rate than tr nL-F and appears to have complementary phylogenetic signal. It may represent a useful phylogenetic marker in itself. Although the exon appears to be intact, some pseudtr nL-F, intron sequences show signs of disruption of the secondary structure which is otherwise conserved across land plants. They may thus be non-functional. The higher evolutionary rate of pseudtr nL-F could reflect reduced selective pressure due to loss of function. However, it might also be the result of transfer to the nucleus where the rate of change is generally higher. This could have implications for the interpretation of its phylogenetic signal. Future research should attempt to determine both the whereabouts and origin of tr nL-F and pseudtr nL-F in the Annonaceae genome. In Chapter 5 a taxonomic revision and phylogeny of the genus Cremastosperma was presented. Twenty-nine species are recognised. Thirteen have been described as new during this project. C. bullatum Pirie, C. cenepense Pirie & Zapata, and C. yamayakatense Pirie from the basin were described following a field expedition in 2003 to Peru. A new species from northern Venezuela (C. venezuelanum Pirie) was described; one from Amazonian Ecuador (C. napoense Pirie), seven from Colombia and western Ecuador (C. antioquense Pirie, C. awaense Pirie, C. chococola Pirie, C. dolichocarpum Pirie, C. longipes Pirie, C. magdalenae Pirie, and C. stenophyllum Pirie) and one from Panama (C. westrae Pirie). A further two putative species were described informally awaiting availability of further collections. Two species plus one 190 • Summary variety were brought into synonymy: C. killipii R.E.Fr. under C. longicuspe R.E.Fr. and both C. juruense R.E.Fr. and C. monospermum (Rusby) R.E.Fr. var. brachypodum R.E.Fr. under C. monospermum. A resolved phylogeny can be used to infer the ancestral distributions of a clade, and possible factors behind its radiation. Multiple chloroplast DNA markers and pseudtr nL-F were applied to reconstruct the phylogeny of 19 of the 29 species of Cremastosperma. A number of clades were revealed. These included species limited to particular geographic areas. The divergence of the Venezuelan and Guianan lineages occurred prior to those of the clades found either in the tropical Andes, or in the Chocó/Darién/western Ecuador region or Central America (i.e. either west or east of the Andes mountain chain). The tropical Andes species all fell into one of two clades. Relationships between these two clades, a further clade including the Central American species Cremastosperma panamense Maas and C. westrae, and two isolated lineages corresponding to accessions from Costa Rica (C. spec. A) and the Magdalena valley of Colombia (C. magdalenae), remain unresolved. These results provide further evidence to suggest the importance of the Andean orogeny as a vicariance event in the history of the evolution of Cremastosperma. Further conclusions await a more resolved phylogeny of the genus. Summary • 191 192 • Summary Nederlandse Samenvatting Dit proefschrift bevat de resultaten van een onderzoek naar enkele Neotropische plantengroepen binnen de Annonaceae, een familie van bloemplanten waarvan de soorten in regenwouden over de hele wereld voorkomen. Het gaat zowel om moleculair systematisch en biogeografisch onderzoek, als om een taxonomische revisie van het genus Cremastosperma. In hoofdstuk 2 worden de eerste resultaten weergegeven van een vergelijkend onderzoek naar de diversificatie van vier voornamelijk Neotropische genera, Cremastosperma, Duguetia, Guatteria en Mosanonna. Met gebruik van methoden zoals fylogenie-reconstructie en moleculaire dateringstechnieken (in dit geval nonparametric rate-smoothing: NPRS) werden verklaringen gezocht voor de verschillen in het aantal soorten binnen deze vier genera. Dit werd gedaan door de monofylie van deze genera vast te stellen en een eerste schatting te maken van de relatieve ouderdom van hun eerste gemeenschappelijke voorouders. Ook werd het effect bepaald van het aantal taxa en kenmerken op de schatting van de ouderdom van deze genera. In het geval van het soortenrijke genus Guatteria had het opnemen van meer soorten uit de kroongroep in de analyse een significant hogere ouderdomsschatting voor de meest recente gemeenschappelijke voorouder tot gevolg. Deze mogelijk ernstige tekortkoming van de veelgebruikte moleculaire dateringsmethode maakt verdere gedetailleerdere vergelijking van soortenrijkdom tussen de genera zinloos. De gevoeligheid van moleculaire dateringsmethoden voor het aantal taxa dat bemonsterd wordt moet verder bestudeerd worden. In hoofdstuk 3 wordt een clade, de zogenaamde SAC-clade (South American centred clade), geïdentificeerd die zich voornamelijk in Zuidamerikaanse gebieden bevindt. Deze bevat alle genera van de zogenaamde short-branch clade (SBC) die in Zuid-Amerika en voor een beperkt deel in Centraal-Amerika voorkomen, maar niet de genera die endemisch zijn voor Azië en CentraalAmerika. Ook al is de fylogenie van het SAC-clade niet helemaal opgelost, Samenvatting • 193 de resultaten suggereren toch een gemeenschappelijk oorsprong van de SACclade in Zuid-Amerika. De schattingen van de ouderdom, die gemaakt zijn met gebruik van drie verschillende ‘rate smoothing’ methoden, ondersteunen de hypothese dat de SAC-clade in Zuid-Amerika is ontstaan na verspreiding via de Bering Straat. Zowel Gentry’s hypothese dat dit patroon ontstaan is door opsplitsing (vicariantie) van Gondwana, als de mogelijkheid van een meer recente verspreiding vanuit Afrika, worden hierdoor verworpen. Monofylie werd bevestigd voor de genera Cremastosperma, Malmea en Mosannona. Het monotypische genus Pseudephedranthus is genest binnen Klarobelia, een genus dat afgezien daarvan een monofyletische groep vormde. Dit leidt tot de conclusie dat de geobserveerde verspreidingspatronen rondom de Andes niet het arbitraire resultaat zijn van de definitie van poly- of parafyletische groepen. Er zijn geen significante verschillen geconstateerd tussen de schattingen van de ouderdom van de meest recente gemeenschappelijke voorouders van elke clade. Alhoewel de kracht van deze test gelimiteerd is door de onnauwkeurige resultaten van de moleculaire datering, lijken deze schattingen binnen de tijdperiode waarin het noordelijk deel van de Andes is ontstaan te vallen. Om deze hypothesen verder te onderzoeken zijn fylogenieën op soortniveau van Cremastosperma, Klarobelia, Mosannona en Malmea nodig. Aanvullende data moet ook worden gegenereerd om de ouderdomsschatting van Annonaceae en andere Magnoliales te testen. Deze benadering zou het dynamische proces van de recente invasie in Centraal-Amerika en de oorsprong van de hoge biodiversiteit in tropisch Amerika kunnen onthullen. In hoofdstuk 4 wordt de ontdekking van een paraloge kopie van de veelgebruikte chloroplast marker tr nL-F beschreven. Deze twee kopieën zijn gedivergeerd in een van de gemeenschappelijke voorouders van de Annonaceae. Primers werden ontwikkeld om de verschillende paraloge kopieën te amplificeren en te sequencen. De gevonden sequenties werden daarna met elkaar vergeleken om de functionele homologie en het fylogenetisch gebruik van de kopieën vast te stellen. Pseudtr nL-F, de paraloge kopie die normaal niet gesequenced wordt, evolueerde sneller dan tr nL-F. Het bevat complementaire fylogenetische signalen en is daardoor zelf ook bruikbaar als fylogenetische marker. Alhoewel het exon intact lijkt, vertonen sommige sequenties van het pseudtr nL-F intron tekenen van verstoring van de secundaire structuur, die in landplanten normaliter geconserveerd is gebleven. Het is dus mogelijk dat deze kopie haar functie verloren heeft. Het is eveneens mogelijk dat de hogere evolutiesnelheid van pseudtr nL-F de selectiedruk weergeeft die ontstaan is door het verlies aan functionaliteit. Deze hogere snelheid zou echter ook het gevolg kunnen zijn van verplaatsing van de paraloge kopie naar de celkern, waar de 194 • Samenvatting mutatiesnelheid in het algemeen hoger is. Dit kan ook implicaties hebben voor de interpretatie van het fylogenetische signaal van deze sequenties. Aanvullend onderzoek zal de plaats van tr nL-F en pseudtr nL-F in het Annonaceae-genoom moeten aantonen. In hoofdstuk 5 worden een taxonomische revisie en een fylogenie van het genus Cremastosperma gepresenteerd. De taxonomische studie heeft geleid tot de erkenning van 29 soorten, waarvan er 13 nieuw zijn voor de wetenschap. De drie soorten C. bullatum Pirie, C. cenepense Pirie & Zapata, and C. yamayakatense Pirie komen allen voor in de vallei van de Río Marañon in Peru, en deze werden beschreven na een verzameltocht naar Peru in 2003. Daarnaast werden één nieuwe soort uit Noord Venezuela (C. venezuelanum Pirie), één uit Amazonisch Ecuador (C. napoense Pirie), zeven uit Colombia en west Ecuador (C. antioquense Pirie, C. awaense Pirie, C. chococola Pirie, C. dolichocarpum Pirie, C. longipes Pirie, C. magdalenae Pirie, and C. stenophyllum Pirie), en één uit Panama (C. westrae Pirie) beschreven. Ook zijn twee mogelijk nieuwe soorten informeel beschreven in afwachting van aanvullende collecties. Twee soorten en een variëteit zijn gereduceerd tot synoniemen en ondergebracht onder de volgende soorten: C. killipii R.E.Fr. onder C. longicuspe R.E.Fr., en C. juruense R.E.Fr. en C. monospermum (Rusby) R.E.Fr. var. brachypodum R.E.Fr. onder C. monospermum. Een opgeloste fylogenie kan gebruikt worden om de distributie van de voorouders in een clade en de mogelijke factoren achter de radiatie te reconstrueren. Hiertoe werd de fylogenie van Cremastosperma gereconstrueerd, aan de hand van meerdere chloroplast DNA merkers en pseudtr nL-F. Een aantal clades binnen Cremastosperma wordt goed ondersteund door bootstrap percentages, en correspondeert in het algemeen met duidelijk omgrensde gebieden in Zuid Amerika. De Venezuelaanse en Guiaanse afstammingslijnen splitsten eerder af dan afstamminglijnen die in de laaglandbossen op de uitlopers van de Andes, en in Centraal Amerika en het gebied Chocó/Darién/westelijk Ecuador voorkomen - met andere woorden, ten oosten en ten westen van de Andes. De verwantschappen tussen twee clades die samen de soorten ten oosten van de Andes bevatten, een derde clade met de Centraalamerikaanse soorten C. panamense en C. westrae, alsmede de soorten C. spec. A (uit Costa Rica) en C. magdalenae (uit de vallei van de Río Magdalena in Colombia) blijven onopgelost. De resultaten suggereren dat de scheiding van populaties door het ontstaan van de Andes een belangrijke rol heeft gespeeld in de evolutionaire historie van Cremastosperma. Een beter opgeloste fylogenie is nodig om verdergaande conclusies te trekken. Samenvatting • 195 196 • Samenvatting Resumen Esta tesis presenta los resultados de la sistemática molecular y la biogeografía de grupos pertenecientes a la familia de plantas con flor, Annonaceae, y la revisión taxonómica del género Cremastosperma, los cuales se encuentran en el Neotrópico. En el capítulo 2 se presentan los resultados preliminares de la comparación de los tiempos de diversificación de cuatro géneros predominante Neotropicales de la familia Annonaceae: Cremastosperma, Duguetia, Guatteria, y Mosannona. Se buscaron explicaciones para la disparidad en el número de especies de estos cuatro géneros. La reconstrucción de la filogenia fue utilizada para determinar su monofilia y se emplearon técnicas de datación molecular (aplicando nonparametric rate-smoothing: NPRS, por su nombre y siglas en ingles) para estimar las edades preliminares relativas de sus ancestros comunes más recientes (MRCAs por sus siglas en ingles). En estos géneros se evaluaron y compararon los efectos del muestreo de taxones y caracteres en las estimaciones de edad. Un muestreo dentro del “crown group” (por su nombre en ingles) del rico genero en especies Guatteria, dio una estimación de la edad significativamente mas vieja para su MRCA. Esto representa una tendencia potencialmente seria de un método extensamente usado de datación molecular, el cual en este caso dio comparaciones adicionales sin sentido de la riqueza de especies en los cuatro géneros. Las técnicas de datación molecular deben ser evaluadas para la sensibilidad de los niveles de muestreo de los taxones bajo diferentes condiciones. En el capítulo 3 se identifico un clado llamado “South American-centred” (el clado SAC). El clado SAC contiene todos los géneros del clado de la rama corta (the short-branch clade, SBC, en ingles) distribuidos en Suramérica y solamente a una extensión limitada en Centro América, pero no aquellos endémicos en Asia y Centro América. En ausencia de una filogenia completamente resuelta del clado SAC, esto se interpreta sugiriendo un origen común en Suramérica. Las estimaciones de edad (producidas usando tres métodos de “rate smoothing”, como su nombre en ingles) apoyaron la hipótesis que el clade SAC se originó Resumen • 197 en Suramérica por la dispersión a través del estrecho Bering. Se rechazo la hipótesis de Gentry de un origen más antiguo por la vicarianza de Gondwana, al igual que la posibilidad de una dispersión posterior desde África. Se confirmo la monofilia de Cremastosperma, Malmea, y Mosannona. El género monotípico Pseudephedranthus se encontró dentro de Klarobelia, especies las cuales de otra manera formaban un grupo monofilético. Se concluyo que los patrones de distribución centro Andina como se observan no son el resultado arbitrario de la definición de grupos poli- o parafileticos. Las edades estimadas de cada clado por los MRCAs no fueron significativamente diferentes el uno del otro. Aunque el poder de esta prueba se vio limitada por la imprecisión en los resultados de datación molecular, estas edades parecen caer dentro del periodo de tiempo de la orogenia de los Andes del norte. Para probar estas hipótesis, son necesarias filogenias a nivel de especie de Cremastosperma, Klarobelia, Mosannona y Malmea. Se debe también buscar datos adicionales para probar la calibración de edad de la familia Annonaceae y de la filogenia más amplia de Magnoliales. Este enfoque podría adicionalmente dar alguna luz en los procesos dinámicos de la reciente invasión de Centro América, y el origen de la alta diversidad de especies en la América tropical. En el capítulo 4 se descubrió un paralogo antiguo del tr nL-F, marcador del cloroplasto extensamente usado. Se deduce que la divergencia de la copias de los paralogos ocurrió en un antepasado común de Annonaceae. Se desarrollaron primers para amplificar y secuenciar los diferentes paralogos, y las secuencias resultantes fueron comparadas para determinar su homología funcional y utilidad filogenética. Pseudtr nL-F (el paralogo no normalmente secuenciado) evoluciona a una rata más rápida que tr nL-F y parece tener una señal filogenética complementaria. Puede representar en sí mismo un marcador filogenéticamente útil. Aunque el exón parece estar intacto, algunas secuencias del intrón del pseudtr nL-F muestran signos de interrupción de la estructura secundaria, la cuál se conserva de otra manera a través de las plantas terrestres. Pueden por lo tanto no ser funcionales. La alta rata evolutiva del pseudtr nL-F podría reflejar la presión selectiva reducida debido a la pérdida de función. Sin embargo, este también puede ser el resultado de la transferencia al núcleo donde la rata de cambio es generalmente más alta. Esto podría tener implicaciones para la interpretación de su señal filogenética. Futura investigaciones deben intentar determinar tanto el lugar como el origen de tr nL-F y del pseudtr nL-F en el genoma de la familia Annonaceae. En el capítulo 5 se presenta la revisión taxonómica y la filogenia del género Cremastosperma. Se reconocen veinte nueve especies. Trece se han descrito como nuevas durante este proyecto. C. bullatum Pirie, C. cenepense Pirie & Zapata, and C. yamayakatense Pirie fueron descritas después de una expedición a campo al Perú en el 2003. Se describió una nueva especie del norte de Venezuela (C. venezuelanum Pirie); una de la Amazonia Ecuatoriana 198 • Resumen (C. napoense Pirie); siete de Colombia y del occidente del Ecuador (C. antioquense Pirie, C. awaense Pirie, C. chococola Pirie, C. dolichocarpum Pirie, C. longipes Pirie, C. magdalenae Pirie y C. stenophyllum Pirie) y una de Panamá (C. westrae Pirie). Dos supuestas especies más fueron descritas informalmente esperando la disponibilidad de más colecciones. Dos especies más una variedad fueron sinonimia: C. killipii R.E.Fr. bajo C. longicuspe R.E.Fr. y tanto C. juruense R.E.Fr. como monospermum (Rusby) R.E.Fr. var. brachypodum R.E.Fr. bajo C. monospermum. Una filogenia resuelta puede ser usada para inferir la distribución ancestral de un clado, y los posibles factores detrás de su radiación. Los múltiples marcadores de ADN del cloroplasto y el pseudtr nL-F fueron aplicados para la reconstrucción de la filogenia de 19 de las 29 especies de Cremastosperma. Se revelaron un número de clados. Éstos incluyeron especies limitadas a unas áreas geográficas particulares. La divergencia de los linajes venezolanos y de la Guyana ocurrieron antes de los aquellos ocurridos en los clados encontrados en los Andes tropicales, o en la región de Chocó/Darién/Ecuador occidental o América Central (ej. a cualquiera de los dos lados de la Cordillera de los Andes al occidente o al oriente). Todas las especies de los Andes tropicales caen dentro de uno de dos clados. Permanecen sin resolver las relaciones entre estos dos clados, un clado de las especies de Centro América C. panamense Maas y C. westrae, y dos linajes aislados correspondientes a las especies de Costa Rica (C. spec. A) y a la del valle del Magdalena en Colombia (C. magdalenae). Estos resultados proporcionan evidencia adicional para sugerir la importancia de la orogenia andina como un evento de vicariance en la historia de la evolución de Cremastosperma. Otras conclusiones aguardan una filogenia resuelta del género. Resumen • 199 200 • Resumen References Altschul, S. F., Madden, T. L., Schaf fer, A. A., Zhang, J., & Zhang, Z. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucl. Acids Res. 25: 3389-3402. APG II. 2003. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG II. Bot. J. Linn. Soc. 141 (4): 399-436. Aylif fe, M. A., Scott, N. S., & Timmis, J. N. 1998. Analysis of plastid DNA-like sequences within the nuclear genomes of higher plants. Molec. Biol. Evol. 15 (6): 738745. Bakker, F. T. Culham, A., Marais, E. M., & Gibby, M. 2005. Nested radiation in Cape Pelargonium. Pages 75-100 in Plant species-level systematics: new perspectives on patter n and process (Bakker, F. T., Chatrou, L. W., Gravendeel, B., and Pelser, P. B., eds.). Regnum Vegetabile 143, A. R. G. Gantner Verlag, Liechtenstein. Bakker, F. T., Culham, A., Gomez-Martinez, R., Carvalho, J., Compton, J., Dawtrey, R., & Gibby, M. 2000. Patterns of nucleotide substitution in angiosperm cpDNA tr nL (UAA)- tr nF (GAA) regions. Molec. Biol. Evol. 17 (8): 1146-1155. Barraclough, T. G., & Reeves, G. 2005. The causes of speciation in flowering plant lineages: species-level DNA trees in the African genus Protea. Pages 31-46 in Plant species-level systematics: new perspectives on patter n and process (Bakker, F. T., Chatrou, L. W., Gravendeel, B., and Pelser, P. B., eds.). Regnum Vegetabile 143, A. R. G. Gantner Verlag, Liechtenstein. Besendahl, A., Qiu, Y.-L., Lee, J., Palmer, J. D., & Bhattacharya, D. 2000. The cyanobacterial origin and vertical transmission of the plastid tRNA LEU group-I intron. Cur. Genet. 37: 12-23. Borsch, T., Hilu, K. W., Quandt, D., Wilde, V., Neinhuis, C., & Barthlott, W. 2003. Noncoding plastid tr nT-tr nF sequences reveal a well resolved phylogeny of basal angiosperms. J. Evol. Biol. 16: 558-576. Bremer, B., Jansen, R. K., Oxelman, B., Backlund, M., Lantz, H., & Kim, K.J. 1999. More characters or more taxa for a robust phylogeny - case study from the coffee family. Syst. Biol. 48 (3): 413-435. Bremer, K. 1994. Branch support and tree stability. Cladistics 10: 295-304. References • 201 Bremer, K. 2000. Early Cretaceous lineages of monocot flowering plants. Proc. Natl. Acad. Sci. USA 97 (9): 4707-4711. Bur nham, R. J., & Graham, A. 1999. The history of Neotropical vegetation: new developments and status. Ann. Missouri Bot. Gard. 86: 546-589. Bygrave, P. 2000. Molecular Systematics of Annonaceae Juss. PhD Thesis. The University of Reading, UK. Cech, T. R., Damberger, S. H., & Gutell, R. 1994. Representation of the secondary and tertiary structure of group I introns. Nat. Struct. Biol. 1: 273-280. Chase, M. W., & Cox, A. V. 1998. Gene sequences, collaboration and analysis of large data sets. Austral. Syst. Bot. 11: 215-229. Chatrou, L. W. 1997. Studies in Annonaceae XXVIII. Macromorphological variation of recent invaders in northern Central America: the case of Malmea (Annonaceae). Amer. J. Bot. 84 (6): 861-869. Chatrou, L. W. 1998. Changing Genera: Systematic studies in Neotropical and West African Annonaceae. PhD Thesis. Utrecht University, the Netherlands. Chatrou, L. W., Koek-Noorman, J., & Maas, P. J. M. 2000. Studies in Annonaceae XXXVI. The Duguetia alliance: where the ways part. Ann. Missouri Bot. Gard. 87: 234-245. Chatrou, L. W., & Pirie, M. D. 2005. Three new rarely collected or endangered species of Annonaceae from Venezuela. Blumea 50: 33-40. Chatrou, L. W., Rainer, H., & Maas, P. J. M. 2004. Annonaceae. Pages 18-20 in Flowering Plants of the Neotropics (Smith, N., Mori, S. A., Henderson, A., Stevenson, D. W., and Heald, S. V., eds.). Princeton, New Jersey, USA. Chesters, K. I. M. 1955. Some plant remains from the Upper Cretaceous and Tertiary of West Africa. Ann. Mag. Nat. Hist. 12 (8): 489-504. Clegg, M. T., & Zurawski, G. 1992. Chloroplast DNA and the study of plant phylogeny: present status and future prospects. Pages 1-13 in Molecular Systematics of Plants (Soltis, P. S., Soltis, D. E., and Doyle, J. J., eds.). Chapman and Hall, New York. Crepet, W. L., & Nixon, K. C. 1998. Two new fossil flowers of magnoliid affinity from the Late Cretaceous of New Jersey. Amer. J. Bot. 85 (9): 1273-1288. Crepet, W. L., Nixon, K. C., & Gandolfo, M. A. 2004. Fossil evidence and phylogeny: the age of major angiosperm clades based on mesofossil and macrofossil evidence from Cretaceous deposits. Amer. J. Bot. 91 (10): 1666-1682. Cuénoud, P., Savolainen, V., Chatrou, L. W., Powell, M., Grayer, R. J., & Chase, M. W. 2002. Molecular phylogenetics of Caryophyllales based on nuclear 18S rDNA and plastid rbcL, atpB, and matK DNA sequences. Amer. J. Bot. 89 (1): 132-144. Cummings, M. P., Nugent, J. M., Olmstead, R. G., & Palmer, J. D. 2003. Phylogenetic analysis reveals five independent transfers of the chloroplast gene rbcL to the mitochondrial genome in angiosperms. Cur. Genet. 43: 131-138. Davies, T. J., Barraclough, T. G., Chase, M. W., Soltis, P. S., Soltis, D. E., & Savolainen, V. 2004. Darwin’s abominable mystery: Insights from a supertree of the angiosperms. Proc. Natl. Acad. Sci. USA 101 (7): 1904-1909. 202 • References Davis, C. C., Bell, C. D., Mathews, S., & Donoghue, M. J. 2002. Laurasian migration explains Gondwanan disjunctions: Evidence from Malpighiaceae. Proc. Natl. Acad. Sci. USA 99 (10): 6833-6837. Davis, P. H., & Heywood, V. H. 1963. Principles of angiosperm taxonomy. Van Nostrand, New York. Davis, S. D., Heywood, V. H., Herrera-MacBryde, O., Villla-Lobos, J., & Hamilton, A. C. 1997. Centres of Plant Diversity. A guide and strategy for their conservation. WWF and IUCN, Cambridge, U.K. De Jussieu, A. L. 1789. Genera Plantarum. Herissant et Barrois, Paris, France. Diels, L. 1906. Annonaceae andinae. Bot. Jahrb. Syst. 37: 408-410. Diels, L. 1931. Annonaceae novae. Notizbl. Bot. Gart. Berlin-Dahlem 11: 73-86. Dilcher, D. L., & Crane, P. R. 1984. Archaeanthus: an early angiosperm from the Cenomanian of the Western Interior North America. Ann. Missouri Bot. Gard. 71: 351-383. Dobes, C. H., Mitchell-Olds, T., & Koch, M. A. 2004. Extensive chloroplast haplotype variation indicates Pleistocene hybridization and radiation of North American Arabis drummondii x divaricarpa, and A. holboelli (Brassicaceae). Molec. Ecol. 13 (2): 349-370. Donoghue, M. J., & Mathews, S. 1998. Duplicate genes and the root of angiosperms, with an example using phytochrome sequences. Molec. Phylog. Evol. 9 (3): 489. Doyle, J. A., Bygrave, P., & Le Thomas, A. 2000. Implications of molecular data for pollen evolution in Annonaceae. Pages 259-284 in Pollen and spores: morphology and biology (Harley, M. M., Morton, C. M., and Blackmore, S., eds.). Royal Botanic Garden Kew, UK. Doyle, J. A., & Le Thomas, A. 1996. Phylogenetic analysis and character evolution in Annonaceae. Adansonia 18: 279-334. Doyle, J. A., & Le Thomas, A. 1997. Phylogeny and geographic history of Annonaceae. Géographie physique et Quarter naire 51 (3): 353-361. Doyle, J. A., Sauquet, H., Scharaschkin, T., & Le Thomas, A. 2004. Phylogeny, molecular and fossil dating, and biogeographic history of Annonaceae and Myristicaceae (Magnoliales). Int. J. Pl. Sci. 165 (4 suppl.): S55-S67. Doyle, J. J., & Doyle, J. L. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phyt. Bull. 19: 11-15. Dunal, M. F. 1817. Monographie de la famille des Annonacées. Treuttel & Würtz, Paris. Fay, M. F., Bayer, C., Alverson, S., de Bruijn, A. Y., & Chase, M. W. 1998. Plastid rbcL sequence data indicate a close affinity between Diegodendron and Bixa. Taxon 47: 43-50. Fay, M. F., Swensen, S. M., & Chase, M. W. 1997. Taxonomic affinities of Medusagyne oppositifolia (Medusagynaceae). Kew Bull. 52: 111-120. Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39 (4): 783-791. References • 203 Fitch, W. M. 1971. Toward defining the course of evolution: minimum change for a specified tree topology. Syst. Zool. 20: 406-416. Fries, R. E. 1930. Revision der Arten einiger Anonaceen-Gattungen I. Acta Horti Berg. 10 (1): 1-128. Fries, R. E. 1931. Revision der Arten einiger Anonaceen-Gattungen II. Acta Horti Berg. 10 (2): 129-341. Fries, R. E. 1934. Revision der Arten einiger Anonaceen-Gattungen III. Acta Horti Berg. 12 (1): 1-220. Fries, R. E. 1937. Revision der Arten einiger Annonaceen-Gattungen IV. Acta Horti Berg. 12 (2): 221-288. Fries, R. E. 1939. Revision der Arten einiger Annonaceen-Gattungen V. Acta Horti Berg. 12 (3): 289-577. Fries, R. E. 1948. Annonaceae from Tropical America. Kongl. Svenska Vetenskapsakad. Handl. 24 (10): 3-5. Fries, R. E. 1950. Contributions to the knowledge of the Annonaceae in northern South America. Ark. Bot. 1 (6): 329-331. Fries, R. E. 1959. Annonaceae. Pages 1-171 in Die Natürlichen Pflanzenfamilien, ed. 2, Band 17a II (Engler, A., and Prantl, K., eds.). Duncker & Humblot, Berlin. Fulton, T. M., Van der Hoeven, R., Eannetta, N. T., & Tanksley, S. D. 2002. Identification, analysis, and utilization of conserved otholog set markers for comparative genomics in higher plants. Pl. Cell 14 (7): 1457-1467. Gaut, B. S. 1998. Molecular clocks and nucleotide substitution rates in higher plants. Pages 93-120 in Evolutionary Biology (Hecht, M. K., MacIntyre, R. J., and Clegg, M. T., eds.). Plenum Press, New York. Gentry, A. H. 1982. Neotropical floristic diversity: phytogeographical connections between Central and South America, Pleistocene climatic fluctuations, or an accident of the Andean orogeny? Ann. Missouri Bot. Gard. 69: 557-593. Geyer, C. J. 1991. Markov chain Monte Carlo maximum likelihood. Pages 156-163 in Computing sciene and statistics: proceedings of the 23rd symposium on the inter face (Keramidas, M., ed.) Interface Foundation of North America, Fairfax, Virginia, USA. Gregory-Wodzicki, K. M. 2000. Uplift history of the Central and Northern Andes: A review. GSA Bull. 112 (7): 1091-1105. Haf fer, J. 1969. Speciation in Amazonian forest birds. Science 165: 131-137. Hamilton, M. B. 1999. Four primer pairs for the amplification of chloroplast intergenic regions with intraspecific variation. Molec. Ecol. 8: 521-523. Hedges, B. S., & Kumar, S. 2004. Precision of molecular time estimates. Trends Genet. 20 (5): 242-247. Hennig, W. 1966. Phylogenetic Systematics. University of Illinois Press, Urbana. Hillis, D. M., & Bull, J. J. 1993. An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis. Syst. Biol. 42 (2): 182-192. 204 • References Hilu, K. W., Borsch, T., Müller, K., Soltis, D. E., Soltis, P. S., Savolainen, V., Chase, M. W., Powell, M. P., Alice, L. A., Evans, R., Sauquet, H., Neinhuis, C., Slotta, T. A. B., Rohwer, J. G., Campbell, C. S., & Chatrou, L. W. 2003. Angiosperm phylogeny based on matK sequence information. Amer. J. Bot. 90 (12): 1758-1776. Hooghiemstra, H., & van der Hammen, T. 1998. Neogene and Quaternary development of the neotropical rain forest: the forest refugia hypothesis, and a literature overview. Earth-Sci. Rev. 44 (3-4): 147. Hoot, S. B., Culham, A., & Crane, P. R. 1995. The utility of atpB gene sequences in resolving pylogenetic relationships: comparison with rbcL and 18S ribosomal DNA sequences in the Lardizabalaceae. Ann. Missouri Bot. Gard. 82 (2): 194-207. Huelsenbeck, J. P. 1997. Is the Felsenstein zone a fly trap? Syst. Biol. 46: 69-74. Huelsenbeck, J. P. 2000. MrBayes: Bayesian inference of phylogeny. Distributed by the author. Department of Biology, University of Rochester. Huelsenbeck, J. P., Ronquist, F., Nielsen, R., & Bollback, J. P. 2001. Bayesian inference of phylogeny and its impact on evolutionary biology. Science 294: 2310-2314. Iturralde-Vinent, M. A., & McPhee, R. D. E. 1999. Paleogeography of the Caribbean region: Implications for Cenozoic biogeography. Bull. Amer. Mus. Nat. Hist. 238: 195. Johnson, D. M., & Murray, N. A. 1995. Synopsis of the tribe Bocageeae (Annonaceae), with revisions of Cardiopetalum, Froesiodendron, Trigynaea, Bocagea, and Hor nschuchia. Brittonia 47 (3): 248-319. Kim, S., Park, C.-W., Kim, Y.-D., & Suh, Y. 2001. Phylogenetic relationships in family Magnoliaceae inferred from ndhF sequences. Amer. J. Bot. 88 (4): 717-728. Koch, M. A., Dobes, C., Matschinger, M., Bleeker, W., Vogel, J., Kiefer, M., & Mitchell-Olds, T. 2005. Evolution of the tr nF(GAA) gene in Arabidopsis relatives and the Brassicaceae family: monophyletic origin and subsequent diversification of a plastidic pseudogene. Molec. Biol. Evol. 22 (4): 1032-1043. Koek-Noor man, J., Van Setten, A. K., & Van Zuilen, C. M. 1997. Studies in Annonaceae. XXVI. Flower and fruit morphology in Annonaceae. Their contribution to patterns in cluster analysis. Bot. Jahrb. Syst. 119 (2): 213-230. Kojoma, M., Kurihara, K., Yamada, K., Sekita, S., Satake, M., & Iida, O. 2002. Genetic identification of cinnamon (Cinnamomum spp.) based on tr nL-tr nF chloroplast DNA. Planta Med. 68 (1): 94-96. Lundberg, J. G., & Chernof f, B. 1992. A miocene fossil of the Amazonian fish Arapaima (Teleostei, Arapaimidae) from the Magdalena river region of Colombia - biogeographic and evolutionary implications. Biotropica 24 (1): 2-14. Maas, P. J. M. 1983. Project systematics of Annonaceae. Taxon 32: 528-529. Maas, P. J. M., Mennega, E. A., & Westra, L. Y. T. 1994. Studies in Annonaceae. XXI. Index to species and infraspecific taxa of neotropical Annonacea. Candollea 49 (2): 389-481. References • 205 Maas, P. J. M., & Westra, L. Y. T. 1984. Studies in Annonaceae. II. A monograph of the genus Anaxagorea A.St.Hil., part 1. Bot. Jahrb. Syst. 105 (1): 73-134. Maas, P. J. M., & Westra, L. Y. T. 1985. Studies in Annonaceae. II. A monograph of the genus Anaxagorea A.St.Hil., part 2. Bot. Jahrb. Syst. 105 (2): 145-204. Maas, P. J. M., Van Heusden, E. C. H., Koek-Noorman, J., Van Setten, A. K., & Westra, L. Y. T. 1986. Studies in Annonaceae. VII: New species from the tropics and miscellaneous notes. Proc. Kon. Ned. Akad. Wetensch. Ser. C 89 (3): 249-278. Maas, P. J. M., & Westra, L. Y. T. 1992. Flora Neotropica Monograph 57: Rollinia (Annonaceae). The New York Botanical Garden, New York. Maas, P. J. M., & Westra, L. Y. T. 2003. Revision of the Neotropical genus Pseudoxandra (Annonaceae). Blumea 48: 201-259. Maas, P. J. M., Westra, L. Y. T., & Chatrou, L. W. 2003. Flora Neotropica Monograph 88: Duguetia (Annonaceae). The New York Botanical Garden, New York. Macbride, J. F. 1929. In: Spermatophytes, mostly Peruvian 3. Some Peruvian Annonaceae with a new Guatteria. Publ. Field Columbian Mus., Bot Ser. 4 (7): 171-172. Mathews, S., & Donoghue, M. J. 1999. The root of angiosperm phylogeny inferred from duplicate phytochrome genes. Science 286: 947-950. Michel, F., & Westhof, E. 1990. Modeling of the three-dimensional structure of group I catalytic introns based on comparative sequence analysis. J. Mol. Biol. 216: 585-610. Millen, R. S., Olmstead, R. G., Adams, K. L., Palmer, J. D., Lao, N. T., Heggie, L., Kavanagh, T. A., Hibberd, J. M., Gray, J. C., Morden, C. W., Calie, P. J., Jer miin, L. S., & Wolfe, K. H. 2001. Many parallel losses of infA from chloroplast DNA during angiosperm evolution with multiple independent transfers to the nucleus. Pl. Cell 13 (3): 645-658. Mols, J. B. 2004. From Miliusa to Miliuseae to Miliusoid: identifying clades in Asian Annonaceae. PhD thesis, Universiteit Leiden. Mols, J. B., Co, D. L. V., Gravendeel, B., Chatrou, L. W., Pirie, M. D., van der Ham, R. W. J. M., van Marle, E. J., & Kessler, P. J. A. submitted. Morphological character evolution in the miliusoid clade (Annonaceae). Int. J. Pl. Sci.. Mols, J. B., Gravendeel, B., Chatrou, L. W., Pirie, M. D., Bygrave, P. C., Chase, M. W., & Kessler, P. J. A. 2004. Identifying clades in Asian Annonaceae: monophyletic genera in the polyphyletic Miliuseae. Amer. J. Bot. 91 (4): 590-600. Morley, R. J. 2003. Interplate dispersal paths for megathermal angiosperms. Perspectives Pl. Ecol. Evol. Syst. 6: 5-20. Muller, J., & Muller, K. 2004. Treegraph: automated drawing of complex tree figures using an extensible tree description format. Molec. Ecol. Notes 4 (4): 786-788. Murray, N. A. 1993. Revision of Cymbopetalum and Porcelia (Annonaceae). Syst. Bot. Monogr. 40: 1-121. Myers, N., Mittermeier, C. G., da Fonseca, G. A. B., & Kent, J. 2000. Biodiversity hotspots for conservation priorities. Nature 403: 853-858. Nakazono, M., & Hirai, A. 1993. Identification of the entire set of transfered chloroplast DNA sequences in the mitochondrial genome of rice. Molec. Gen. Genet. 236: 341-346. 206 • References Nee, S., Holmes, E. C., May, R. M., & Harvey, P. H. 1994. Extinction rates can be estimated from molecular phylogenies. Phil. Trans., Ser. B 344: 77-82. Nichols, R. 2001. Gene trees and species trees are not the same. Trends Ecol. Evol. 16 (7): 358-364. Nixon, K. C. 1999. The Parsimony Ratchet, a New Method for Rapid Parsimony Analysis*1. Cladistics 15 (4): 407-414. Notsu, Y., Masood, S., Nishikawa, T., Kubo, N., Akiduki, N., Nakazono, M., Hirai, A., & Kadowaki, K. 2002. The complete sequence of the rice (Oryza sativa L.) mitochondrial genome: frequent DNA sequence acquisition and loss during the evolution of flowering plants. Molec. Genet. Genomics 268: 434-445. Olmstead, R. G., Michaels, H. J., Scott, K. M., & Palmer, J. D. 1992. Monophyly of the Asteridae and identification of major lineages inferred from DNA sequences of rbcL. Ann. Missouri Bot. Gard. 79: 249-265. Olmstead, R. G., & Sweere, J. A. 1994. Combining data in phylogenetic systematics: an empirical approach using three molecular data sets in the Solanaceae. Syst. Biol. 43 (4): 467-481. Pennington, R. T., & Dick, C. W. 2004. The role of immigrants in the assembly of the South American rainforest tree flora. Phil. Trans., Ser. B 359: 1611-1622. Pennington, R. T., Lavin, M., Prado, D. E., Pendry, C. A., Pell, S. K., & Butterworth, C. A. 2004. Historical climate change and speciation: neotropical seasonally dry forest plants show patterns of both Tertiary and Quaternary diversification. Phil. Trans., Ser. B 359: 515-538. Pindall, J. L., Cande, S. C., Pitman, W. C., Rowley, D. B., Dewey, J. F., LaBrecque, J., & Haxby, W. 1988. A plate-kinematic frame-work for models of Caribbean evolution. Tectonophysics 155: 121-138. Pirie, M. D. 2005. New species of Cremastosperma (Annonaceae) from Colombia, Ecuador, and Panama. Blumea 50: 41-60. Pirie, M. D., & Zapata C., M. 2004. Three new endemic species of Cremastosperma (Annonaceae) from the Río Marañon basin, Amazonas, Peru. Ar naldoa 11(2): 7-20. Posada, D., & Crandall, K. A. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14 (9): 817-818. Qiu, Y. L., Lee, J., Ber nasconi-Quadroni, F., Soltis, D. E., Soltis, P. S., Zanis, M., Zimmer, E. A., Chen, Z., Savolainen, V., & Chase, M. W. 2000. Phylogeny of basal angiosperms: Analysis of five genes from three genomes. Int. J. Pl. Sci. 161: Suppl. 6: 3-27. Quandt, D., Muller, K., Stech, M., Frahm, J.-P., Frey, W., Hilu, K., & Borsch, T. 2004. Molecular evolution of the chloroplast tr nL-F region in land plants. Ann. Missouri Bot. Gard.: 13-37. Quandt, D., & Stech, M. 2004. Molecular evolution of the tr nT UGU- tr nF GAA region in bryophytes. Plant Biology 6: 545-554. Reisz, R. R., & Muller, J. 2004. Molecular timescales and the fossil record: a paleontological perspective. Trends Genet. 20 (5): 237-241. References • 207 Renner, S. S. 2004. Bayesian analysis of combined chloroplast loci, using multiple calibrations, supports the recent arrival of Melastomataceae in Africa and Madagascar. Amer. J. Bot. 91 (9): 1427-1435. Renner, S. S., Clausing, G., & Meyer, K. 2001. Historical biogeography of Melastomataceae: the roles of Tertiary migration and long-distance dispersal. Amer. J. Bot. 88 (7): 1290-1300. Richardson, J. E., Chatrou, L. W., Mols, J. B., Erkens, R. H. J., & Pirie, M. D. 2004. Historical biogeography of two cosmopolitan families of flowering plants: Annonaceae and Rhamnaceae. Phil. Trans., Ser. B 359: 1495-1508. Rogstad, S. H. 1989. The biosystematics and evolution of the Polyalthia hypoleuca complex (Annonaceae) of Malesia, I. Systematic treatment. J. Ar nold Arb. 70: 153-246. Ruiz Lopez, H., & Pavon, J. A. 1959. Flora Peruviana, et chilensis. Anales Inst. Bot. Cavanilles 17 (1): 411-415, 418-432. Ruiz Lopez, H., & Pavón, J. A. 1798. Systema vegetabilium florae peruvianae et chilensis. Gabrielis de Sancha, Madrid. Rusby, H. H. 1910. New species from Bolivia, collected by R.S. Williams - I. Bull. New York Bot. Gard. 6: 504-505. Rusby, H. H. 1927. Descriptions of new genera and species of plants collected on the Mulford biological exploration of the Amazon Valley, 1921-1922. Mem. New York Bot. Gard. 7: 242-247. Rutschmann, F. 2004. Bayesian molecular dating using PAML/multidivtime. A step-by-step manual. University of Zürich, Switzerland. Sanderson, M. J. 1997. A nonparametric approach to estimating divergence times in the absence of rate constancy. Molec. Biol. Evol. 14 (12): 1218-1231. Sanderson, M. J. 2002a. Estimating absolute rates of molecular evolution and divergence times: a penalised likelihood approach. Molec. Biol. Evol. 19 (1): 101-109. Sanderson, M. J. 2002b. r8s, version 1.50. Distributed by the author, Section of Evolution and Ecology, University of California, Davis. Sanderson, M. J., & Doyle, J. A. 2001. Sources of error and confidence intervals in estimating the age of Angiosperms from rbcL and 18S rDNA data. Amer. J. Bot. 88 (8): 1499-1516. Sanderson, M. J., & Shaf fer, H. B. 2002. Troubleshooting molecular phylogenetic analyses. Annu. Rev. Ecol. Syst. 33: 49-72. Sanderson, M. J., Thor ne, J. L., Wikstrom, N., & Bremer, K. 2004. Molecular evidence on plant divergence times. Amer. J. Bot. 91 (10): 1656-1665. Sauquet, H., Doyle, J. A., Scharaschkin, T., Borsch, T., Hilu, K., Chatrou, L. W., & Le Thomas, A. 2003. Phylogenetic analysis of Magnoliales and Myristicaceae based on multiple data sets: implications for character evolution. Bot. J. Linn. Soc. 142: 125-186. Savolainen, V., Cuénoud, P., Spichiger, R., Martinez, M. D. P., Crèvecoeur, M., & Manen, J.-F. 1995. The use of herbarium specimens in DNA phylogenetics: evaluation and improvement. Pl. Syst. Evol. 197: 87-98. 208 • References Shaw, J., Lickey, E. B., Beck, J. T., Farmer, S. B., Liu, W., Miller, J., Siripun, K. C., Winder, C. T., Schilling, E. E., & Small, R. L. 2005. The tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. Amer. J. Bot. 92 (1): 142-166. Shields, R. 2004. Pushing the envelope on molecular dating. Trends Genet. 20 (5): 221222. Siddall, M. E., & Whiting, M. F. 1999. Long-branch abstractions. Cladistics 15 (1): 924. Sikes, D. S., & Lewis, P. O. 2001. PAUPRat: PAUP* implementation of the parsimony ratchet., version 1, beta. Distributed by the authors. Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, USA. Simmons, M. P., & Miya, M. 2004. Efficiently resolving the basal clades of a phylogenetic tree using Bayesian and parsimony approaches: a case study using mitogenomic data from 100 higher teleost fishes. Molec. Phylog. Evol. 31 (1): 351-362. Simmons, M. P., & Ochoterena, H. 2000. Gaps as characters in sequence-based phylogenetic analysis. Syst. Biol. 49 (2): 369-381. Small, R. L., Rybur n, J. A., Cronn, R. C., Seelanan, T., & Wendel, J. F. 1998. The tortoise and the hare: choosing between noncoding plastome and nuclear Adh sequences for phylogeny reconstruction in a recently diverged plant group. Amer. J. Bot. 85 (9): 1301-1315. Smith, A. B., & Peterson, K. J. 2002. Dating the time of origin of major clades: Molecular Clocks and the Fossil Record. Annu. Rev. Earth Planet. Sci. 30 (1): 65-88. Smith, N., Mori, S. A., Henderson, A., Stevenson, D. W., & Heald, S. V. 2004. Flowering Plants of the Neotropics. Princeton University Press, Princeton, New Jersey, USA. Sole de Porta, N. 1971. Algunos generos nuevos de polen precedentes de la formación Guaduas (Maastrichtiense-Paleocene) de Colombia. Stud. Geol. Salamanca 2: 133-143. Soltis, P. S., Soltis, D. E., Savolainen, V., Crane, P. R., & Barraclough, T. G. 2002. Rate heterogeneity among lineages of tracheophytes: Integration of molecular and fossil data and evidence for molecular living fossils. Proc. Natl. Acad. Sci. USA 99 (7): 4430-4435. Sorenson, M. D. 1999. TreeRot, version 2. Boston University. Suzuki, Y., Glazko, G. V., & Nei, M. 2002. Overcredibility of molecular phylogenies obtained by Bayesian phylogenetics. Proc. Natl. Acad. Sci. USA 99 (25): 16138-16143. Swof ford, D. L. 2000. PAUP*. Phylogenetic analysis using parsimony (* and other methods). Sinauer. Swof ford, D. L., Olsen, G. J., Waddell, P. J., & Hillis, D. M. 1996. Phylogeny reconstruction. Pages 407-514 in Molecular systematics (Hillis, D. M., Moritz, C., and Mable, B. K., eds.). Sinauer Associates, Inc., Sunderland, Massachusetts. Taberlet, P., Gielly, L., Pautou, G., & Bouvet, J. 1991. Universal primers for amplification of the three non-coding regions of chloroplast DNA. Pl. Mol. Biol. 17: 1105-1109. References • 209 Tavare, S. 1986. Some probabilistic and statistical problems on the analysis of DNA sequences. Lectures on Mathematics in the Life Sciences 17: 57-86. Ter Steege, H., Sabatier, D., Castellanos, H., van Andel, T. R., Duivenvoorden, J., Adalardo de Oliveira, A., Ek, R. C., Lilwah, R., Maas, P. J. M., & Mori, S. A. 2000. An analysis of the floristic composition and diversity of Amazonian forests including those of the Guiana Shield. J. Trop. Ecol. 16 (6): 801-828. Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F., & Higgins, D. G. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucl. Acids Res. 25 (24): 4876-4882. Thorne, J., Kishino, H., & Painter, I. 1998. Estimating the rate of evolution of the rate of molecular evolution. Molec. Biol. Evol. 15 (12): 1647-1657. Thorne, J. L., & Kishino, H. 2002. Divergence time and evolutionary rate estimation with multilocus data. Syst. Biol. 51: 689-702. Thornton, J. W., & DeSalle, R. 2000. Gene family evolution and homology: genomics meets phylogenetics. Annu. Rev. Genomics Hum. Genet. 1: 41-73. Tuomisto, H., & Ruokolainen, K. 1997. The role of ecological knowledge in explaining biogeography and biodiversity in Amazonia. Biodiv. Cons. 6: 347-357. Valencia, R., Balslev, H., & Paz y Miño, G. 1994. High tree alpha-diversity in Amazonian Ecuador. Biodiv. Cons. 3: 21-28. Van Heusden, E. C. H. 1992. Flowers of Annonaceae: morphology, classification, and evolution. Blumea Suppl. 7: 1-218. Van Setten, A. K., & Koek-Noor man, J. 1992. Studies in Annonaceae. XVII. Fruits and seeds of Annonaceae: morphology and its significance for classification. Biblioth. Bot. 142: 1-101. Vijverberg, K., & Bachmann, K. 1999. Molecular evolution of a tandemly repeated tr nF (GAA) gene in the chloroplast genomes of Microseris (Asteraceae) and the use of structural mutations in phylogenetic analyses. Molec. Biol. Evol. 16 (10): 1329-1340. Von Martius, C. F. P. 1841. Annonaceae. Flora Brasiliensis 13 (1): 1-64. Wagner, A. 2002. Selection and gene duplication: a view from the genome. Genome Biology 3 (5): reviews1012.1 - reviews1012.3. Walker, J. M. 1971. Pollen morphology, phytogeography and phylogeny of the Annonaceae. Contr. Gray Herb. 202: 1-133. Wikström, N., Savolainen, V., & Chase, M. W. 2001. Evolution of the angiosperms: calibrating the family tree. Proc. Roy. Soc. Lond. Biol. 1482: 2211-2220. Yang, Z. 1997. PAML: a program package for phylogenetic analysis by maximum likelihood. CABIOS 13: 555-556. Zuker, M. 2003. Mfold web server for nucleic acid folding and hybridization prediction. Nucl. Acids Res. 31 (13): 3406-3415. 210 • References Dankwoord Tijdens฀de฀vier฀jaren฀(en฀nog฀wat)฀van฀mijn฀promotie฀onderzoek฀in฀Utrecht฀ heb฀ik฀veel฀geleerd฀en฀meegemaakt.฀Ik฀heb฀er฀ook฀ontzettend฀veel฀plezier฀in฀ gehad,฀waarvoor฀ik฀natuurlijk฀veel฀mensen฀te฀danken฀heb.฀Hier฀wil฀ik฀graag฀ enkelen฀van฀hen฀noemen. Bij฀het฀herbarium฀had฀ik฀de฀geluk฀om฀terecht฀te฀komen฀in฀een฀hechte฀ groep,฀rijk฀aan฀opgebouwde฀kennis.฀In฀Paul฀Maas฀had฀ik฀een฀promotor฀die฀mij฀ de฀ ruimte฀ gaf฀ om฀ het฀ onderzoek฀ dat฀ ik฀ wilde฀ doen฀ uit฀ te฀ voeren,฀ maar฀ die฀ mij฀ook฀steunde฀met฀zijn฀inzicht฀in฀en฀begrip฀voor฀de฀planten฀waarmee฀ik฀ vaak฀worstelde.฀Paul,฀bedankt฀voor฀alles.฀Mijn฀enige฀spijt฀is฀dat฀ik฀de฀kans฀niet฀ gepakt฀heb฀om฀met฀je฀op฀veldexpeditie฀te฀gaan.฀Misschien฀dat฀ik฀mezelf฀nog฀ mee฀kan฀smokkelen฀bij฀een฀van฀je฀toekomstige฀verzameltochten฀-฀tenminste,฀als฀ ik฀er฀niet฀te฀oud฀voor฀word฀om฀je฀nog฀bij฀te฀kunnen฀houden.฀Lubbert฀Westra฀ en฀Jifke฀Koek-Noorman฀hebben฀het฀ook฀druk,฀zonder฀recht฀op฀verlof.฀Hun฀ kennis฀en฀advies฀heb฀ik฀heel฀dankbaar฀kunnen฀benutten.฀The฀generosity฀of฀Bep฀ Mennega,฀in฀the฀guise฀of฀the฀Alberta฀Mennega฀Stichting,฀has฀allowed฀me฀to฀ attend฀a฀number฀of฀important฀congresses.฀That฀of฀the฀Miquel฀Fonds฀also฀helped฀ me฀on฀my฀way฀to฀Peru.฀Heel฀erg฀bedankt฀daarvoor.฀Verder฀heb฀ik฀bijna฀iedereen฀ in฀het฀herbarium฀een฀keertje฀over฀iets฀lastig฀gevallen.฀Marion฀Jansen-Jacobs฀ over฀collecties,฀Hans฀ter฀Steege฀en฀Paddy฀Haripersaud฀over฀het฀fotograferen฀ van฀herbarium฀exemplaren,฀Guda฀Bernard-Glijnis฀over฀boeken,฀en฀Rob฀Bakker,฀ Erik-Jan฀van฀Marle,฀en฀Erik฀Simonis฀over฀van฀alles฀en฀nog฀wat.฀Alle฀andere฀ mensen฀die฀ooit฀“rondhangen”฀op฀de฀19de฀en฀21ste฀verdieping,฀hartelijk฀bedankt฀ voor฀alle฀gezelligheid฀tijdens฀de฀koffiepauzes฀en฀daarbuiten. In฀het฀moleculaire฀lab฀wordt฀iedereen฀erg฀goed฀begeleid฀en฀geholpen฀door฀ onze฀manager,฀Jan฀Maas.฀Dankzij฀Jan฀blijft฀de฀boel฀altijd฀draaien,฀en฀samen฀met฀ Herman฀ en฀ Cees,฀ zorgt฀ hij฀ ook฀ voor฀ een฀ gezellige฀ sfeer.฀ Dat฀ er฀ in฀ Utrecht฀ toch฀ maar฀ enkele฀ mensen฀ bezig฀ zijn฀ met฀ fylogeniereconstructie,฀ geeft฀ onze฀ samenwerking฀met฀de฀Onderzoekschool฀Biodiversiteit,฀en฀de฀andere฀‘takken’฀ van฀het฀Nationaal฀Herbarium฀Nederland฀een฀extra฀groot฀gewicht.฀‘Molecular฀ Taskforce’฀ meetings฀ met฀ collega’s฀ van฀ Leiden฀ en฀ Wageningen฀ zijn฀ altijd฀ Dankwoord • 211 ontzettend฀interessant฀en฀edukatief฀geweest,฀en฀de฀verdere฀opbouw฀van฀het฀ Annonaceae฀project฀binnen฀de฀NHN฀heeft฀tot฀een฀zeer฀prettige฀samenwerking฀ geleid.฀My฀special฀thanks฀to฀Johan฀Mols,฀Freek฀Bakker,฀and฀James฀Richardson฀ who฀not฀only฀helped฀me฀in฀my฀own฀work,฀but฀helped฀me฀along฀by฀involving฀ me฀in฀theirs. Met฀de฀projecten฀van฀veel฀mensen฀die฀stage฀bij฀ons฀kwamen฀lopen฀heb฀ik฀ me฀lekker฀bemoeid.฀Bij฀sommige฀was฀dat฀ook฀min฀of฀meer฀de฀bedoeling.฀MariaPaula฀Balcázar฀began฀the฀attack฀for฀a฀molecular฀phylogeny฀of฀the฀Unonopsis฀ clade,฀and฀in฀doing฀so฀was฀beset฀by฀the฀kind฀of฀fascinating฀problems฀you฀can฀ only฀really฀enjoy฀if฀someone฀else฀is฀doing฀the฀pippetting.฀The฀poisoned฀chalice฀ has฀since฀been฀passed฀on,฀but฀the฀solution฀to฀that฀particular฀piece฀of฀genetic฀ trickery฀developed฀into฀Chapter฀4฀of฀this฀thesis.฀Not฀only฀did฀you฀come฀back฀ for฀the฀M.Sc.฀module,฀but฀you฀sacrificed฀yourself฀selflessly฀to฀countless฀requests฀ for฀translations,฀ranging฀from฀emails฀to฀my฀Spanish฀summary฀here.฀Many฀thanks฀ again!฀Danaë฀Rozendaal฀maakte฀er฀een฀bijzondere฀stage฀waar฀iedereen฀wat฀van฀ geleerd฀heeft.฀Ook฀jou฀werden฀de฀mogelijke฀complicaties฀van฀het฀bakken฀van฀ moleculaire฀fylogenieën฀niet฀bespaard.฀Je฀bent฀er฀geslaagd฀doorgekomen. Tijdens฀mijn฀project฀heb฀ik฀het฀geluk฀gehad฀om฀zelfs฀twee฀veldexpedities฀ naar฀ tropisch฀ Amerika฀ te฀ kunnen฀ maken.฀ Dit฀ waren฀ geweldige,฀ intensieve,฀ en฀ indrukwekkende฀ ervaringen.฀Voor฀ de฀ eerste,฀ in฀ Bolivia,฀ mocht฀ ik฀ met฀ Lars฀ mee฀ om฀ mijn฀ eerste฀ Cremastosper ma’s฀ in฀ het฀ echt฀ te฀ zien,฀ en฀ om฀ verzamelvaardigheden฀te฀leren฀van฀een฀expert.฀Eén฀van฀de฀belangrijkste฀lessen฀ voor฀mij฀was฀om฀te฀zien฀hoe฀cruciaal฀de฀hulp฀van฀collega’s฀in฀het฀bezochte฀ land฀is฀voor฀het฀succes฀van฀zo’n฀expeditie.฀Thanks฀are฀due฀to฀Stephan฀Beck฀and฀ the฀staff฀of฀the฀Herbario฀Nacional฀de฀Bolivia,฀to฀Mario฀Saldías฀and฀staff฀of฀the฀ Herbarium฀of฀the฀Universidad฀Autónoma฀Gabriel฀René฀Moreno,฀and฀to฀Pierre฀ Ibisch฀and฀Christoph฀Nowicki,฀and฀Israel฀Vargas฀of฀the฀Fundación฀Amigos฀de฀la฀ Naturaleza.฀Special฀thanks฀to฀René฀Boot,฀at฀that฀time฀at฀PROMAB฀(Programa฀ Manejo฀de฀Bosques฀de฀la฀Amazonía฀Boliviana)฀in฀Riberalta,฀where฀we฀were฀ also฀joined฀by฀Tanya฀Scharaschkin.฀We฀greatly฀enjoyed฀a฀most฀successful฀stay฀ thanks฀to฀the฀help฀of฀René,฀and฀that฀of฀Jacaranda฀van฀Rheenen,฀Roel฀Brienen,฀ René฀Aramayo,฀Adhemar฀Saucedo,฀Don฀Nico฀Divico,฀and฀Calixto฀Salas. Toen฀ het฀ onmogelijk฀ was฀ voor฀ Lars฀ op฀ mijn฀ tweede฀ reis,฀ naar฀ Peru,฀ mee฀te฀gaan,฀werd฀ik฀gered฀door฀Marleen฀Botermans฀en฀Robin฀van฀Velzen.฀ Zonder฀ jullie฀ was฀ het฀ een฀ eenzame฀ en฀ zeer฀ stressvolle฀ tocht฀ geweest.฀ Soms฀ was฀het฀toch฀even฀doorbikkelen,฀maar฀ik฀ben฀trots฀op฀wat฀wij฀samen฀hebben฀ gepresteerd.฀Tenemos฀una฀deuda฀de฀gratitude฀con฀una฀multitud฀de฀personas฀ que฀de฀una฀u฀otra฀forma฀hicieron฀mas฀facil฀nuestro฀pasaje฀por฀Lima฀a฀traves฀ de฀Annonaceae฀en฀Pasco,฀Amazonas฀y฀San฀Martin.฀A฀nuestra฀llegada฀al฀Museo฀ de฀Historia฀Natural฀(Universidad฀Nacional฀Mayor฀de฀San฀Marcos,฀Lima,฀Peru)฀ fuimos฀tratados฀con฀mucha฀amabilidad,฀generosidad฀y฀paciencia.฀Queremos฀dar฀ gracias,฀especialmente฀a฀Magda฀Chanco,฀Oscar฀Tovar,฀Asunción฀Cano,฀Haydee฀ 212 • Dankwoord Montoya,฀y฀Hamilton฀Beltrán.฀Marybel฀Morales฀nos฀acompaño฀desde฀Lima฀a฀ traves฀de฀Pasco฀donde฀fuimos฀extremamente฀afortunados฀de฀recibir฀la฀ayuda฀y฀ la฀guia฀de฀Rocío฀Rojas฀y฀del฀resto฀del฀personal฀en฀la฀base฀de฀Oxapampa฀del฀ Missouri฀Botanical฀Garden฀y฀del฀personal฀y฀guardaparques฀de฀INRENA฀en฀ el฀Parque฀Nacional฀Yanachaga-Chemillen.฀฀El฀resto฀del฀tiempo฀en฀el฀campo฀ estuvimos฀acompañados฀por฀Mario฀Zapata,฀de฀la฀Universidad฀Privada฀Antenor฀ Orrega,฀Trujillo฀ un฀ gran฀ compañero฀ de฀ viaje฀ y฀ consumado฀ naturalista.฀ En฀ Trujillo,฀nuestros฀agradecimientos฀son฀para฀Abundio฀Sagástegui฀(en฀particular฀ por฀ compartir฀ a฀ Mario฀ de฀ UPAO)฀ y฀ al฀ personal฀ del฀ Herbario฀Truxillense฀ especialmente฀a฀Eric฀Rodríguez฀y฀Victor฀Medina฀Ibañez฀por฀sus฀sugerencias฀ acerca฀de฀las฀colectas฀en฀Amazonas.฀En฀Amazonas฀fuimos฀privilegiados฀de฀tener฀ el฀ permiso฀ de฀ estar฀ y฀ trabajar฀ en฀ la฀ comunidad฀ de฀Aguaruna฀ de฀Yamayakat,฀ Imaza.฀Agradecemos฀a฀Apu฀Ephrain฀Wisum฀Yagkug฀y฀a฀nuestros฀guias฀expertos฀ Ricardo฀Apanú฀Nampin,฀y฀Julio฀Saan฀Kasen฀pues฀en฀sus฀bosques฀colectamos฀ especimenes฀de฀dos฀especies฀nuevas฀de฀Cremastosper ma.฀Finalmente฀en฀San฀ Martín฀ fuimos฀ ayudados฀ por฀ Marco฀ León฀ y฀ Jim฀Vasquez฀ del฀ herbario฀ y฀ por฀ Tomas฀Diaz฀de฀la฀estacion฀de฀campo฀“Biodiversidad”฀la฀localidad฀de฀la฀nueva฀ especie฀descrita฀Pseudoxandra฀angustifolia฀Maas),฀de฀la฀Universidad฀Nacional฀ San-Martín,฀Tarapoto,฀Peru.฀ Hendrik฀ Rypkema฀ heeft฀ de฀ grootste฀ deel฀ van฀ alle฀ ฀ tekeningen฀ in฀ dit฀ proefschrift฀gemaakt,฀van฀de฀kaart฀op฀de฀omslag฀tot฀de฀‘Nijntje’฀tekeningen฀in฀ de฀sleutel.฀Om฀het฀boekje฀mooi฀in฀elkaar฀te฀zetten,฀heb฀ik฀het฀verder฀getroffen฀ met฀Pieter฀van฀Dorp฀van฀Vliet.฀Arme฀Pieter฀werd฀deze฀keer฀dus฀het฀slachtoffer฀ bij฀grafische฀vormgeving฀die฀de฀formaterings฀nachtmerrie฀van฀een฀systematisch฀ proefschrift฀in฀handen฀kreeg.฀Ik฀voel฀me฀ontzettend฀trots฀dat฀mijn฀eigen฀naam฀ op฀zo’n฀mooi,฀door฀jullie฀vormgegeven฀produkt฀mag฀pronken. Voor฀mijn฀paranimfen,฀Roy฀Erkens฀en฀Jeanine฀Berk,฀is฀het฀werkje฀pas฀net฀ begonnen฀-฀bedankt฀alvast!฀Roy฀in฀het฀bijzonder฀heeft฀me฀heel฀vaak฀geholpen฀ sinds฀ik฀hier฀aan฀kwam฀vier฀jaar฀geleden,฀en฀we฀hebben฀ook฀samen฀veel฀lol฀ gehad.฀Uwe฀Scharf฀is฀er฀ook฀regelmatig฀geweest฀sinds฀het฀begin,฀en฀was฀onder฀ andere฀een฀inspiratie฀door฀zijn฀snel฀leren฀van฀het฀Nederlands.฀Finally,฀saved฀ (almost)฀till฀last,฀I’d฀like฀to฀thank฀my฀co-promotor฀Lars฀Chatrou.฀It฀has฀been฀ a฀ real฀ pleasure฀ to฀ roll฀ into฀ work฀ each฀ day฀ and฀ know฀ that฀ there’d฀ always฀ be฀ interesting฀things฀to฀discuss฀with฀you.฀It’s฀also฀been฀a฀reassurance฀(and฀not฀just฀ for฀me)฀to฀know฀that฀in฀you฀there฀was฀a฀broad฀pair฀of฀shoulders฀onto฀which฀ problems฀could฀summarily฀be฀dumped฀when฀it฀all฀got฀too฀much.฀I฀hope฀I฀didn’t฀ take฀too฀much฀advantage,฀too฀often! Ik฀wil฀graag฀een฀extra฀‘dank฀je฀wel’฀zeggen฀tegen฀iedereen฀hier฀die฀heeft฀ gezorgd฀dat฀ik฀me฀de฀laatste฀jaren฀zo฀thuis฀heb฀gevoeld.฀Het฀is฀als฀buitenlander฀ erg฀ makkelijk฀ om฀ jezelf฀ een฀ beetje฀ buitengesloten฀ te฀ voelen,฀ juist฀ door฀ het฀ feit฀dat฀iedereen฀Engels฀kan฀spreken.฀Jullie฀zijn฀erg฀geduldig฀met฀mij฀geweest,฀ en฀ik฀ben฀er฀erg฀dankbaar฀voor.฀My฀folks฀at฀home฀have฀always฀been฀a฀great฀ Dankwoord • 213 support,฀despite฀barely฀getting฀to฀see฀me฀over฀the฀last฀few฀years.฀Er฀blijft฀op฀ deze฀lijst฀natuurlijk฀één฀iemand฀over,฀voor฀wie฀geduld฀hebben฀met฀mij฀een฀ kunst฀is฀geworden,฀zeker฀in฀de฀laatste฀maanden฀tijdens฀het฀opschrijven฀van฀dit฀ proefschrift.฀Geertje,฀je฀bent฀gewoon฀geweldig.฀Dank฀je฀wel฀voor฀dat. 214 • Dankwoord Publications Pirie, M. D., & Goldsmith, F. B. 1999. Changes in Hertfordshire Ponds. Trans HNHS 33 (5): 482-493. Chatrou, L. W., & Pirie, M. D. 2003. Two new species of Annonaceae from Bolivia. Rev. Soc. Bol. Bot. 4 (1): 25-30. Mols, J. B., Gravendeel, B., Chatrou, L. W., Pirie, M. D., Bygrave, P. C., Chase, M. W., & Kessler, P. J. A. 2004. Identifying clades in Asian Annonaceae: monophyletic genera in the polyphyletic Miliuseae. Amer. J. Bot. 91 (4): 590-600. Pirie, M. D., & Zapata C., M. 2004. Three new endemic species of Cremastosperma (Annonaceae) from the Río Marañon basin, Amazonas, Peru. Ar naldoa 11(2): 7-20. Richardson, J. E., Chatrou, L. W., Mols, J. B., Erkens, R. H. J., & Pirie, M. D. 2004. Historical biogeography of two cosmopolitan families of flowering plants: Annonaceae and Rhamnaceae. Phil. Trans., Ser. B 359: 1495-1508. Chatrou, L. W., & Pirie, M. D. 2005. Three new rarely collected or endangered species of Annonaceae from Venezuela. Blumea 50: 33-40. Pirie, M. D. 2005. New species of Cremastosperma (Annonaceae) from Colombia, Ecuador, and Panama. Blumea 50: 41-60. Pirie, M. D., Chatrou, L. W., Erkens, R. H. J., Maas, J. W., Van der Niet, T., Mols, J. B., & Richardson, J. E. 2005. Phylogeny reconstruction and molecular dating in four Neotropical genera of Annonaceae: the effect of taxon sampling in age estimations. Pages 149-174 in Plant species-level systematics: new perspectives on patter n and process (Bakker, F. T., Chatrou, L. W., Gravendeel, B., and Pelser, P. B., eds.). Regnum Vegetabile 143, A. R. G. Gantner Verlag, Liechtenstein. Chatrou, L. W., Pirie, M. D., & Maas, P. J. M. in press. Annonaceaein Catalogue of Vascular Plants of Bolivia (Jørgensen, P. M., ed.) Missouri Botanical Garden, St. Louis. Mols, J. B., Co, D. L. V., Gravendeel, B., Chatrou, L. W., Pirie, M. D., van der Ham, R. W. J. M., van Marle, E. J., & Kessler, P. J. A. submitted. Morphological character evolution in the miliusoid clade (Annonaceae). Int. J. Pl. Sci. Publications • 215 Pirie, M. D., Chatrou, L. W., Mols, J. B., Erkens, R. H. J., & Oosterhof, J. submitted. ‘Andean-centred’ genera in the short-branch clade of Annonaceae: Testing biogeographic hypotheses using phylogeny reconstruction and molecular dating. J. biogeog. Pirie, M. D., Balcázar Vargas, M. P., Boter mans, M., Bakker, F. T., & Chatrou, L. W. submitted. An ancient paralogue of the cpDNA trnL (UAA)-trnF (GAA) region in Annonaceae and its application in phylogeny reconstruction. Molec. Phylog. Evol. 216 • Publications Curriculum Vitae Michael฀ D.฀ Pirie฀ was฀ born฀ in฀ London฀ on฀ the฀ 18 th฀ of฀August฀ 1977.฀ In฀ 1995฀he฀completed฀his฀secondary฀school฀education฀at฀King฀Edward’s฀School,฀ Witley,฀with฀A฀levels฀in฀biology,฀chemistry,฀and฀music.฀In฀October฀of฀that฀year฀ he฀ began฀ further฀ study฀ in฀ biology฀ at฀ University฀ College,฀ London.฀ During฀ the฀three฀year฀B.Sc.฀programme฀he฀followed฀courses฀including฀‘Genetics฀and฀ Evolution’,฀‘Plant฀Reactions฀to฀Environmental฀Stress’,฀‘Population฀Genetics’,฀ and฀‘Conservation฀ Biology’.฀ In฀ 1998฀ he฀ graduated฀ with฀ upper฀ second฀ class฀ honours,฀having฀submitted฀a฀thesis฀based฀on฀a฀biological฀survey฀(results฀of฀which฀ were฀subsequently฀presented฀in฀Pirie฀&฀Goldsmith,฀1999).฀In฀November฀and฀ January฀of฀1998/99฀and฀February฀and฀March฀of฀1999฀he฀volunteered฀as฀a฀field฀ worker฀at฀the฀Australian฀Trust฀for฀Conservation฀Volunteers฀in฀New฀South฀Wales฀ and฀Victoria,฀and฀at฀the฀South฀Slough฀Estuarine฀Research฀Reserve,฀in฀Oregon,฀ USA,฀respectively.฀In฀October฀1999฀he฀began฀the฀course฀‘The฀Biodiversity฀and฀ Taxonomy฀of฀Plants’฀at฀the฀University฀of฀Edinburgh฀and฀Royal฀Botanic฀Garden฀ Edinburgh,฀gaining฀the฀degree฀of฀Master฀of฀Science฀in฀September฀2000,฀with฀ a฀thesis฀entitled฀‘Biogeography฀and฀Systematics฀of฀Centrolobium฀(Fabaceae)’.฀ In฀November฀2000฀he฀moved฀to฀the฀Netherlands฀to฀begin฀research฀for฀a฀Ph.D.฀ as฀‘Assistent฀in฀Opleiding’฀at฀the฀Utrecht฀University฀branch฀of฀the฀National฀ Herbarium฀of฀the฀Netherlands฀(NHN).฀He฀followed฀courses฀both฀at฀the฀NHN฀ (Systematics฀ of฀ Neotropical฀ Plant฀ Families฀ -฀ Utrecht;฀ Advanced฀Topics฀ in฀ Phylogeny฀Reconstruction฀-฀Leiden)฀and฀at฀the฀University฀of฀Copenhagen,฀ Denmark฀(Phylogenetic฀Systematics฀and฀Historical฀Biogeography).฀His฀contract฀ was฀ extended฀ for฀ 3฀ months฀ in฀ 2004฀ in฀ order฀ to฀ help฀ the฀ preparation฀ and฀ teaching฀ of฀ a฀ new฀ course฀‘Advanced฀ Evolutionary฀ Biology’฀ for฀ the฀ Utrecht฀ MSc฀programmes฀‘Plant฀Biology’฀and฀‘Biogeology’.฀As฀part฀of฀his฀Ph.D.฀research฀ he฀made฀collecting฀trips฀to฀Bolivia฀(October-December฀2001,฀with฀Chatrou)฀ and฀ Peru฀ (October-December฀ 2003),฀ resulting฀ in฀ the฀ publication฀ of฀ new฀ species฀(Chatrou฀&฀Pirie,฀2003;฀Pirie฀&฀Zapata฀C.,฀2004)฀and฀a฀contribution฀ to฀the฀Catalogue฀of฀the฀Vascular฀Plants฀of฀Bolivia฀(Chatrou฀et฀ al.,฀in฀press).฀ Further฀publications฀arising฀from฀his฀Ph.D.฀research฀include฀Pirie฀et฀al.฀(2005฀-฀ Curriculum Vitae • 217 Chapter฀2;฀submitted฀-฀Chapters฀3฀&฀4).฀He฀collaborated฀actively฀with฀workers฀ of฀ the฀Annonaceae฀ project฀ across฀ the฀ NHN฀ (Mols฀ et฀ al.,฀ 2004;฀ Richardson฀ et฀ al.,฀ 2004;฀ Mols฀ et฀ al.,฀ submitted)฀ and฀ presented฀ results฀ at฀ the฀ Nordic฀ Botany฀meeting฀(Utrecht฀2001),฀Young฀Systematists฀meeting฀(London฀2002),฀ Annonaceae฀ workshop฀ (Malaga,฀ 2002)฀ and฀ Systematics฀Association฀ biennial฀ meeting฀(Dublin฀2003).฀He฀has฀been฀awarded฀a฀bursary฀to฀attend฀and฀present฀ results฀at฀the฀next฀SA฀meeting฀in฀Cardiff฀(August฀2005).฀ In฀ July฀ 2005฀ he฀ will฀ begin฀ a฀ three฀ year฀ postdoc฀ position฀ researching฀ systematics฀ and฀ biogeography฀ of฀ the฀ grass฀ subfamily฀ Danthonioideae฀ at฀ the฀ Institute฀for฀Systematic฀Botany,฀Zürich,฀Switzerland. 218 • Curriculum Vitae Appendix A Details of accessions sampled for DNA sequence data. Table 1: Accession details and rbcL, tr nL-tr nF, matK, and psbA-tr nH genbank accession numbers. Table 2: ndhF, tr nT-tr nL, tr nS-tr nG, atpB-rbcL and pseud tr nL-F genbank accession numbers. Unless otherwise indicated, sequences were published as original results of chapters of this thesis. DNA ID refers to UDNA database identification number, unless prefixed with JB (Mols, Leiden) or MWC (Chase, Jodrell lab, Royal Botanic Gardens Kew). Voucher specimens for DNA samples extracted at the National Herbarium of the Netherlands (NHN) are held at herbaria as indicated by the following acronyms: NHN - Leiden University (L); New York Botanical Garden (NY); Missouri Botanical Garden (MO); NHN Utrecht University (U); United States National Herbarium (US); and NHN - Wageningen University (WAG). Appendix A • 219 • Origin Coelocaryon Gabon Appendix A preussii Warb. DNA ID Wieringa, J.J. 20 rbcL AY743437 trnL-trnF matK AY743456 AY743475 psbA-trnH et al. 3640 (WAG) Kmeria yunnanensis China Liriodendron g China chinense Sargent Magnolia kobus DC. Voucher AY158183a unknown Chatrou, L.W. 521 AY841593 520 AY743438 b AY841670b AY841424 et al. 279 (U) g Japan Chatrou, L.W. AY743457 AY743476 AY841425 et al. 278 (U) Cinnamomum cassia Blume China AB054241 Izu experimental AB054233c station 21 Persea americana Mill. g Neotropics cv. accnum Galbulimima belgraveana 484 AY841592 AY841669b 479 (U) Australasia (F. Muell.) Sprague Eupomatia bennettii Chatrou, L.W. b AY220415d Qiu, Y-L. 90034 (NCU) Australia Prov. 50/51d unknown F.Muell. Degeneria roseiflora Fiji J.M. Milller Alphonsea boniana Vietnam Finet & Gagnep. J.M. Miller AY220414 1189 (SUVA) AY220361d Kessler, P.J.A. AY318965e AY319077e AY518809f 113 AY743439 AY743458 AY743477 AY841427 498 AY238952b AY231284 AY238960 AY841426 3116 (L) Anaxagorea Brazil - Espirito Maas, P.J.M. silvatica R.E.Fr. Santo et al. 8836 (U) Anaxagorea phaeocarpa Ecuador - Napo Maas, P.J.M. Mart. JB57 et al. 8592 (U) AY238944d • 1 A p p e n d i x A : Ta b l e 1 220 Taxon Taxon Origin Voucher Annickia chlorantha Gabon Sosef, M.S.M. (Oliv.) Setten & Maas Annickia kummeriae Tanzania trnL-trnF matK psbA-trnH 976 AY841594b AY841671b AY841393 AY841442 Johnson, D.M. MWC7004 AY238959b AY231285 AY238961d AY841443 AY238945d 1942 (OWU) Gabon Setten & Maas Annona muricata L. rbcL 1877 (WAG) (Engl. & Diels) Setten & Maas Annickia pilosa (Exell) DNA ID Sosef, M.S.M. 977 AY743450 AY743469 AY743488 AY841444 525 AY743440 AY743459 AY743478 AY841428 529 AY238953b AY231286 AY238962d AY841429 1803 (WAG) g Neotropics Chatrou, L.W. 468 (U) Artabotrys hexapetalus g India (L.f.) Bhandari Asimina triloba (L.) Dunal Chatrou, L.W. AY238946d 470 (U) g North America 517 AY743441 AY743460 AY743479 AY841430 190 AY841600b AY841678b DQ018262 AY841445 332 AY841601b AY841679b Costa Rica - Limón Chatrou, L.W. et al. 93 (U) 239 AY841602b AY841680b AY841394 AY841431 Gabon Wieringa, J.J. et al. 198 AY841603b AY841681b AY841395 AY841432 Chatrou, L.W. et al. 276 (U) Bocageopsis multiflora Guyana (Mart.) R.E.Fr. Bocageopsis pleiosperma Maas Jansen-Jacobs, M.J. et al. 5789 (U) Brazil - Amazonas Miralha, J.M.S. AY841446 et al. 300 (U) Appendix A Cananga odorata (Lam.) Hook.f & Thomson Cleistopholis glauca Engl. & Diels 3278 (WAG) • 221 Cremastosperma brevipes French Guiana Scharf, U. 76 (U) 601 AY743527 AY743573 AY743550 AY841447 Peru - Amazonas Pirie, M.D. et al. 71 (U) 1100 AY743537 AY743583 AY743560 AY841459 (DC.) R.E.Fr. Cremastosperma bullatum Pirie 2 • 222 • Taxon Origin Voucher Cremastosperma bullatum Pirie Peru - Amazonas Pirie, M.D. et al. DNA ID Appendix A rbcL trnL-trnF matK psbA-trnH 1102 DQ018235 DQ018188 DQ018276 DQ018256 598 AY743525 AY743571 AY743548 DQ018240 312 AY743519 AY743565 AY743542 AY841448 DQ018277 DQ018257 94 (U) Cremastosperma cauliflorum Brazil - Acre R.E.Fr. Cremastosperma cauliflorum 9029 (U) Peru - Loreto R.E.Fr. Cremastosperma cenepense Maas, P.J.M. et al. Chatrou, L.W. et al. 224 (U) Peru - Amazonas Rojas, R. 255 (U) 1269 DQ018236 Ecuador - Napo Chatrou, L.W. et al. 491 AY743521 AY743567 AY743544 Pirie & Zapata Cremastosperma gracilipes R.E.Fr. Cremastosperma leiophyllum 267 (U) Bolivia - Santa Cruz Pirie, M.D. et al. 2 (U) 594 AY743523 AY743569 AY743546 AY841449 Venezuela - Falcón Wingfield, R. 6751 (U) 741 AY743528 AY743574 AY743551 AY841450 1270 AY841521 AY841535 DQ018279 AY841460 87 AY743517 AY743563 AY743540 490 AY743520 AY743566 AY743543 DQ018239 493 AY743522 AY743568 AY743545 AY841451 82 AY743516 AY743562 AY743539 R.E.Fr. Cremastosperma macrocarpum Maas Cremastosperma magdalenae Colombia - Antioquia Escobar, L.A. de Pirie 3309 (U) Cremastosperma megalophyllum Ecuador - Napo Maas, P.J.M. et al. R.E.Fr. 8595 (U) Cremastosperma megalophyllum Ecuador - Napo Chatrou, L.W. et al. R.E.Fr. 268 (U) Cremastosperma megalophyllum Ecuador - Napo Chatrou, L.W. et al. R.E.Fr. 259 (U) Cremastosperma microcarpum R.E.Fr. Peru - Loreto Maas, P.J.M. et al. 8289 (U) • 3 Taxon Origin Voucher Cremastosperma microcarpum Peru - Loreto Chatrou, L.W. et al. R.E.Fr. DNA ID rbcL trnL-trnF matK psbA-trnH 303 AY319058e AY319172e AY518874f AY841452 AY841453 208 (U) Cremastosperma monospermum Bolivia - Pando Pirie, M.D. et al. 4 (U) 596 AY743524 AY743570 AY743547 Peru - Loreto Killip, E.P. 28961 (NY) 753 AY743533 AY743579 AY743556 Ecuador - Napo Neill, D. 7649 (U) 737 DQ018224 DQ018177 DQ018265 Devia A., W. 5335 (MO) 744 AY743529 AY743575 AY743552 Pirie, M.D. et al. 7 (U) 1097 DQ018232 DQ018185 DQ018273 DQ018253 Maas, P.J.M. et al. 600 AY743526 AY743572 AY743549 DQ018241 (Rusby) R.E.Fr. Cremastosperma monospermum (Rusby) R.E.Fr. Cremastosperma napoense Pirie Cremastosperma novogranatense Colombia - El Valle DQ018242 R.E.Fr. Cremastosperma oblongum Peru - Pasco R.E.Fr. Cremastosperma oblongum Brazil - Acre R.E.Fr. 9148 (U) Peru - Loreto Morawetz, W. 25985 (U) 739 DQ018225 DQ018178 DQ018266 DQ018243 Panama Miller, J.S. 947 (U) 747 AY743530 AY743576 AY743553 DQ018246 Peru - San Martín Gentry, A.H. 45510 (U) 760 AY743538 AY743584 AY743561 DQ018251 Cremastosperma pedunculatum Ecuador - Zamora- Gentry, A.H. 80904 (U) 754 DQ018228 DQ018181 DQ018269 DQ018248 (Diels) R.E.Fr. Chinchipe Cremastosperma pendulum Peru - Pasco Pirie, M.D. et al. 33 (U) 1098 DQ018233 DQ018186 DQ018274 DQ018254 Cremastosperma oblongum R.E.Fr. Cremastosperma panamense Appendix A Maas Cremastosperma pedunculatum (Diels) R.E.Fr. • 223 (Ruiz & Pav.) R.E.Fr. 4 • 224 • Appendix A Taxon Origin Voucher Cremastosperma pendulum Peru - Huánuco Morawetz, W. 9888 (U) Peru - Ucayali Peru - Amazonas DNA ID rbcL trnL-trnF matK psbA-trnH 749 AY743531 AY743577 AY743554 DQ018247 Graham, J.G. 635 (U) 751 AY743532 AY743578 AY743555 AY841454 Rodríguez R., E. 755 AY743534 AY743580 AY743557 745 DQ018227 DQ018180 DQ018268 245 AY841522 AY841536 (Ruiz & Pav.) R.E.Fr. Cremastosperma pendulum (Ruiz & Pav.) R.E.Fr. Cremastosperma peruvianum R.E.Fr. Cremastosperma sp. 1112 (U) Peru - San Martín Woytkowski, F. DQ018245 7128 (MO) Cremastosperma spec. A Costa Rica - Osa Chatrou, L.W. et al. AY841455 103 (U) Cremastosperma spec. B Peru - Madre de Dios Foster, R.B. 3418 (US) 758 DQ018230 DQ018183 DQ018271 DQ018250 Cremastosperma spec. B Peru - Madre de Dios Smith, S.F. 1577 (US) 756 DQ018229 DQ018182 DQ018270 DQ018249 Cremastosperma venezuelanum Venezuela - Carababo Steyermark, J.A. 757 AY743536 AY743582 AY743559 AY841457 Oliver, R.L. 3681 (MO) 761 DQ018231 DQ018184 DQ018272 DQ018252 Pirie, M.D. et al. 57 (U) 1099 AY743535 AY743581 AY743558 AY841458 Vásquez, R. 19055 (MO) 742 DQ018226 DQ018187 DQ018267 DQ018244 Pirie, M.D. et al. 80 (U) 1101 DQ018234 DQ018179 DQ018275 DQ018255 Chatrou, L.W. et al. 203 AY841523 AY841537 DQ018258 AY841434 Pirie Cremastosperma westrae Pirie 94314 (NY) Panama Cremastosperma yamayakatense Peru - Amazonas Pirie Cremastosperma yamayakatense Peru - Amazonas Pirie Cremastosperma yamayakatense Peru - Amazonas Pirie Cymbopetalum sp. Costa Rica 44 (U) • 5 Taxon Origin Cymbopetalum sp. Bolivia - Santa Cruz Voucher Maas, P.J.M. et al. DNA ID rbcL trnL-trnF matK 98 AY841609b AY841687b 210 AY743442 AY743461 AY743480 psbA-trnH AY841433 8811 (U) Cymbopetalum torulosum Costa Rica - La Selva G.E.Schatz Chatrou, L.W. et al. 54 (U) Dasymaschalon sootepense Craib Thailand Kessler, P.J.A. 3201 (L) 998 AY743443 AY743462 AY743481 Desmopsis microcarpa R.E.Fr. Chatrou, L.W. et al. 234 AY319059e AY319173e AY518804f 475 AY738152 AY740564 AY740532 339 AY738153 AY740565 AY740533 348 AY738154 AY740566 AY740534 357 AY738155 AY740567 AY740535 361 AY841613b AY841691b AY740536 362 AY738156 AY740568 AY740536 365 AY738157 AY740569 AY740537 Chatrou, L.W. et al. 42 (U) 201 AY738158 AY740570 AY740538 372 AY738159 AY740571 AY740539 Costa Rica - Limón AY841461 85 (U) Duguetia bahiensis Maas Brazil - Bahia Amorim, A.M. et al. 800 (U) Duguetia cadaverica Huber Guyana Jansen-Jacobs, M.J. et al. 5868 (U) Duguetia calycina Benoist Guyana Jansen-Jacobs, M.J. et al. 5661 (U) Duguetia cauliflora R.E.Fr. Guyana Jansen-Jacobs, M.J. et al. 5687 (U) Duguetia chrysea Maas Brazil - Amazonas Maas, P.J.M. et al. Appendix A 8052 (U) Duguetia chrysea Maas Brazil - Amazonas Maas, P.J.M. et al. 8053 (U) Duguetia confinis Gabon Wieringa, J.J. & • 225 (Engl. & Diels) Chatrou F.I. van Nek 3290 (WAG) Duguetia confusa Maas Costa Rica Duguetia echinophora R.E.Fr. Brazil - Amazonas Maas, P.J.M. et al. 8046 (U) 6 • AY841435 226 • Taxon Origin Voucher Duguetia guianensis R.E.Fr. Guyana University of Guyana, DNA ID Appendix A rbcL trnL-trnF matK 153 AY738160 AY740572 AY740540 288 AY738161 AY740573 AY740541 732 AY738162 AY740574 AY740542 401 AY738163 AY740575 AY740543 psbA-trnH course Neotropical Botany 33 (U) Duguetia hadrantha Peru - Loreto (Diels) R.E.Fr. Duguetia lucida Urb. Chatrou, L.W. et al. 181 (U) Bolivia - Beni Chatrou, L.W. et al. 367 (U) Duguetia macrocalyx R.E.Fr. Guyana Jansen-Jacobs, M.J. et al. 3011 (U) Duguetia macrophylla R.E.Fr. Peru - Loreto Maas, P.J.M. et al. 403 AY738164 AY740576 AY740544 Duguetia marcgraviana Mart Bolivia - Santa Cruz Chatrou, L.W. et al. 729 AY738165 AY740577 AY740545 406 AY738166 AY740578 AY740546 408 AY738167 AY740579 AY740547 410 AY738168 AY740580 AY740548 302 AY738169 AY740581 AY740549 242 AY738170 AY740582 AY740550 476 AY738171 AY740583 AY740551 327 (U) Duguetia megalocarpa Maas Brazil - Amazonas Maas, P.J.M. et al. 8045 (U) Duguetia moricandiana Mart. Brazil - Bahia Carvalho, A.M.V. de Duguetia neglecta Sandwith Guyana Jansen-Jacobs, M.J. et al. 3322 (U) 5655 (U) Duguetia odorata (Diels) Peru - Loreto J.F.Macbr. Duguetia panamensis Standl. Chatrou, L.W. et al. 207 (U) Costa Rica - Limón Chatrou, L.W. et al. 97 (U) Duguetia peruviana (R.E.Fr.) J.F.Macbr. Ecuador - Guayas Maas, P.J.M. et al. 8571 (U) • 7 Taxon Origin Duguetia pycnastera Sandwith Brazil - Amazonas Voucher Miralha, J.M.S. DNA ID rbcL trnL-trnF matK 416 AY738172 AY740584 AY740552 261 AY738173 AY740585 AY740553 424 AY738174 AY740586 AY740554 426 AY738175 AY740587 AY740555 427 AY738176 AY740588 AY740556 428 AY738177 AY740589 AY740557 854 AY738178 AY740590 AY740558 451 AY738179 AY740591 AY740559 455 AY738180 AY740592 AY740560 335 AY738181 AY740593 AY740561 457 AY738182 AY740594 AY740562 461 AY738183 AY740595 AY740563 et al. 241 (U) Duguetia quitarensis Benth. Peru - Madre de Dios Chatrou, L.W. et al. 123 (U) Duguetia riedeliana R.E.Fr. Brazil - Rio de Janeiro Maas, P.J.M. et al. 8819 (U) Duguetia salicifolia R.E.Fr. Brazil - São Paulo Cordeiro, I. et al. 915 (U) Duguetia sessilis (Velloso) Maas Brazil - Rio de Janeiro Maas, P.J.M. et al. 8838 (U) Duguetia sooretamae Maas Brazil - Espirito Santo Maas, P.J.M. et al. 8827 (U) Duguetia staudtii Cameroon (Engl. & Diels) Chatrou Duguetia stelechantha (Diels) Andel, T.R. van et al. 3290 (U) Brazil - Amazonas R.E.Fr. Maas, P.J.M. et al. 8058 (U) Appendix A Duguetia surinamensis R.E.Fr. Brazil - Amazonas Maas, P.J.M. et al. 8057 (U) Duguetia ulei (Diels) R.E.Fr. Brazil - Amazonas Miralha, J.M.S., 243 (U) • 227 Duguetia uniflora (DC.) Mart. Brazil - Amazonas Coêlho, D. INPA 3711 (U) Duguetia yeshidan Sandwith Guyana Jansen-Jacobs, M.J. et al. 6129 (U) 8 • psbA-trnH 228 • Appendix A Taxon Origin Voucher Ephedranthus parviflorus Brazil - Prance, G.T. et al. S.Moore Mato Grosso 19246 (U) Ephedranthus sp. Brazil - Maas, P.J.M. et al. Espirito Santo Ephedranthus sp. Peru - Loreto DNA ID rbcL trnL-trnF matK 31 AY841615b AY841693b 105 AY841616b AY841694b 284 AY319061e AY319175e 1020 AY743444 AY743463 AY743482 287 AY743445 AY743464 AY743483 Kessler, P.J.A. 3188 (L) 1019 AY743446 AY743465 AY743484 Jongkind, C.C.H. et al. 958 AY743451 AY743470 AY743489 959 AY841524 AY841538 221 AY740958 AY741007 AY740909 psbA-trnH AY841462 AY841396 AY841463 8826 (U) Chatrou, L.W. et al. AY841464 173 (U) Fissistigma glaucescens Merrill Hong Kong Law, C.L & Kendrick, R. (L) Fusaea peruviana R.E.Fr. Peru - Loreto Chatrou, L.W. et al. AY841436 179 (U) Goniothalamus griffithii Thailand Hook.f. & Thoms. Greenwayodendron oliveri Ghana (Engl.) Verdc. Greenwayodendron suaveolens Gabon Engl. & Diels var. suaveolens Guatteria aeruginosa Standl. AY841465 1795 (WAG) McPherson, G. AY841466 15802 (WAG) Costa Rica Chatrou, L.W. et al. 66 (U) Guatteria alata Maas & Panama Mori, S.A. 2894 (U) 776 AY740959 AY741008 AY740910 Panama Mori, S.A. 2952 (U) 777 AY740960 AY741009 AY740911 Bolivia - Chatrou, L.W. et al. 578 AY740961 AY741010 AY740912 Santa Cruz 339 (U) van Setten Guatteria allenii R.E.Fr. Guatteria alutacea Diels • 9 Taxon Origin Voucher DNA ID rbcL trnL-trnF matK psbA-trnH AY841437 Guatteria anomala R.E.Fr. Mexico Ishiki, M. et al. 2233 (U) 150 AY740962 AY741011 AY740913 Guatteria atra Sandwith aff. Guyana Gopaul, D. & Maas, 715 AY740963 AY741012 AY740914 512 AY740964 AY741013 AY740915 785 AY740965 AY741014 AY740916 P.J.M. 2791 (U) Guatteria australis Brazil - Lobão, A.Q. & Fiaschi, A.St.-Hil. aff. Rio de Janeiro P. 499 (U) Guatteria blainii (Griseb.) Urb. Dominican Republic Maas, P.J.M. 6443 (U) Guatteria brevicuspis R.E.Fr. Brazil - Amazonas Prance, G.T.16328 (U) 801 AY740966 AY741015 AY740917 Guatteria caribea Urb. Dominican Republic Tuxill, J. 89 (U) 786 AY740967 AY741016 AY740918 Guatteria chiriquiensis R.E.Fr. Costa Rica Chatrou, L.W. et al. 202 AY740968 AY741017 AY740919 233 AY740969 AY741018 AY740920 552 AY740970 AY741019 AY740921 43 (U) Guatteria diospyroides Baill. Costa Rica - Limón Chatrou, L.W. et al. 84 (U) Guatteria discolor R.E.Fr. aff. Brazil - Acre Maas, P.J.M. et al. 9030 (U) Appendix A Guatteria dumetorum R.E.Fr. Panama FLORPAN 2497 (U) 779 AY740971 AY741020 AY740922 Guatteria elata R.E.Fr. Peru - Loreto Chatrou, L.W. et al. 328 AY740972 AY741021 AY740923 Gentry, A.H. 56948 (U) 833 AY740973 AY741022 AY740924 Chatrou, L.W. et al. 575 AY740974 AY741023 AY740925 99 AY740975 AY741024 AY740926 560 AY740976 AY741025 AY740927 252 (U) Guatteria elegantissima R.E.Fr. Colombia - El Valle Guatteria foliosa Benth. Bolivia - Santa Cruz 325 (U) • 229 Guatteria glabrescens R.E.Fr. Guatteria guianensis (Aubl.) R.E.Fr. 10 • Brazil - Maas, P.J.M. et al. Rio de Janeiro 8816 (U) Brazil - Amazonas Webber, A.C. et al. 1884 (U) AY841438 230 • Appendix A Taxon Origin Voucher DNA ID Guatteria heterotricha R.E.Fr. Colombia - El Valle Monsalve B., M. 1262 (U) 829 Guatteria hyposericea Diels aff. Bolivia - Beni Chatrou, L.W. et al. 580 rbcL trnL-trnF matK AY740977 AY741026 AY740928 AY740978 AY741027 AY740929 psbA-trnH 375 (U) Guatteria inuncta R.E.Fr. Costa Rica Liesner, R.L. 14631 (U) 780 AY740979 AY741028 AY740930 Guatteria inundata Mart. Peru - Loreto Chatrou, L.W. et al. 291 AY740980 AY741029 AY740931 191 (U) Guatteria jefensis Barringer Panama Valdespino, I.A. 685 (U) 781 AY740981 AY741030 AY740932 Guatteria latifolia Brazil - Lobão, A.Q. 544 (U) 510 AY740982 AY741031 AY740933 (Mart.) R.E.Fr. Rio de Janeiro Guatteria liesneri D.M. Brazil - Amazonas Cid F., C.A. 8403 (U) 792 AY740983 AY741032 AY740934 Guatteria macropus Mart. Brazil - Bahia Pirani, J.R. 2725 (U) 790 AY740984 AY741033 AY740935 Guatteria maypurensis Kunth Guyana Jansen-Jacobs, M.J. 182 AY740985 AY741034 AY740936 309 AY740986 AY741035 AY740937 822 AY740987 AY741036 AY740938 Johnson & N.A. Murray et al. 5416 (U) Guatteria megalophylla Diels Peru - Loreto Chatrou, L.W. et al. 216 (U) Guatteria multivenia Diels Ecuador - Napo Maas, P.J.M. 8611 (U) Guatteria oligocarpa Mart. Brazil - Bahia Maas, P.J.M. 7006 (U) 811 AY740988 AY741037 AY740939 Guatteria olivacea R.E.Fr. Peru - Loreto Chatrou, L.W. et al. 304 AY740989 AY741038 AY740940 Chatrou, L.W. et al. 80 (U) 230 AY740990 AY741039 AY740941 Scharf, U. 85 (U) AY740991 AY741040 AY740942 209 (U) Guatteria oliviformis Costa Rica - Donn.Smith Monteverde Guatteria ouregou French Guiana 679 (Aubl.) Dunal • 11 Taxon Origin Voucher Guatteria parvifolia R.E.Fr. Brazil - São Paulo Gottsberger, G.K. DNA ID rbcL trnL-trnF matK 809 AY740992 AY741041 AY740943 819 AY740993 AY741042 AY740944 249 AY740994 AY741043 AY740945 Molino, J.F. 1593 (U) 482 AY740995 AY741044 AY740946 Chatrou, L.W. et al. 283 AY740996 AY741045 AY740947 217 AY740997 AY741046 AY740948 573007 (U) Guatteria pittieri R.E.Fr. Colombia - El Valle Werff, H.H. van der 9767 (U) Guatteria pudica Costa Rica - Osa N.Zamora & Maas Guatteria punctata Chatrou, L.W. et al. 107 (U) French Guiana (Aubl.) R.A. Howard Guatteria puncticulata R.E.Fr. Peru - Loreto 172 (U) Guatteria recurvisepala R.E.Fr. Costa Rica - La Selva Chatrou, L.W. et al. 61 (U) Panama - San Blas Mori, S.A. 5531 (U) 783 AY740998 AY741047 AY740949 Brazil - Minas Gerais CFCR 4116 (U) 803 AY740999 AY741048 AY740950 Guyana Scharf, U. 60 (U) 696 AY741001 AY741050 AY740952 Guatteria scytophylla Diels Brazil - Amazonas Maas, P.J.M. et al. s.n. (U) 559 AY741002 AY741051 AY740953 Guatteria sellowiana Brazil - Rio de Janeiro Lobão, A.Q. 557 (U) 511 AY741003 AY741052 AY740954 Panama - Panama 784 AY741004 AY741053 AY740955 Guatteria rotundata Maas & van Setten Guatteria rupestris Mello-Silva & Pirani Appendix A Guatteria schomburgkiana Mart. • 231 Schltdl. aff. Guatteria sessilicarpa Maas & van Setten 12 • McPherson, G. 2599 (U) psbA-trnH Taxon 232 • Guatteria sp. Appendix A Guatteria tonduzii Diels Origin Voucher Brazil - Lobão, A.Q. et al. Rio de Janeiro 565 (U) Costa Rica - Osa Chatrou, L.W. et al. DNA ID rbcL trnL-trnF matK 513 AY741000 AY741049 AY740951 259 AY741005 AY741054 AY740956 AY740957 psbA-trnH 121 (U) Guatteria villosissima Brazil - Bahia Lobão, A.Q. 630 (U) 564 AY741006 AY741055 Peru - Madre de Dios Gentry, A.H. et al. 1029 AY841525 AY841539 AY841467 277 AY841627b AY841705b AY841468 301 AY743452 AY743471 AY743490 AY841469 486 AY319062e AY319176e AY518866f AY841470 1027 AY841526 AY841540 AY841471 253 AY841628b AY841706b AY841472 145 AY841629b AY841707b 260 AY238955b AY231288 A.St.-Hil. Klarobelia candida Chatrou 43466 (U) Klarobelia cauliflora Chatrou Peru - Loreto Chatrou, L.W. et al. 161 (U) Klarobelia inundata Chatrou Peru - Loreto Chatrou, L.W. et al. 205 (U) Klarobelia megalocarpa Ecuador - Esmeraldas Chatrou Klarobelia sp. Klarobelia stipitata Chatrou Maas, P.J.M. et al. 8521 (U) Venezuela - Meier, W. & S. Llamozas Distr. Federal 3685 (U) Costa Rica - Osa Chatrou, L.W. et al. 113 (U) Letestudoxa bella Pellegr. Gabon Wieringa, J.J. & T.Nzabi 2797 (WAG) Malmea dielsiana R.E.Fr. Peru - Madre de Dios Chatrou, L.W. et al. 122 (U) AY238964d AY841473 AY238948d Malmea dimera Chatrou Panama - Panama Croat, T.B. 34626 (U) 39 AY841631b AY841709b Malmea sp. Peru - Loreto Chatrou, L.W. et al. 8 (U) 197 AY841527 AY841541 AY841474 AY841397 AY841475 • 13 Taxon Origin Voucher Malmea surinamensis Chatrou Suriname Jansen-Jacobs, M.J. DNA ID rbcL trnL-trnF matK psbA-trnH 850 AY743453 AY743472 AY743491 AY841476 AY841477 et al. 6207 (U) Mezzettia parviflora Becc. Indonesia Okada 3388 (L) JB32 AY318983e AY319095e Monanthotaxis whytei g Chatrou, L.W. 475 (U) 533 AY841635b AY841713b Africa (Stapf) Verdc. Monocarpia euneura Miq. Indonesia Slik, F. 2931 (L) JB106 AY318998e AY319111e AY518865f Monodora myristica g Chatrou, L.W. 477 (U) 535 AY743447 AY743466 AY743485 Chatrou, L.W. 237 AY743510 AY743496 AY743503 AY841479 186 AY743511 AY743497 AY743504 AY841480 Garwood, N.C. 3129 (U) 50 AY743512 AY743498 AY743505 AY841481 Maas, P.J.M. et al. 487 AY743513 AY743499 AY743506 AY841482 Ivory Coast (Gaertn.) Dunal Mosannona costaricensis R.E.Fr. Costa Rica - Limón et al. 90 (U) Mosannona discolor (R.E.Fr.) Guyana Chatrou Jansen-Jacobs, M.J. et al. 6000 (U) Mosannona garwoodii Panama - Barro Chatrou & Welzenis Colorado Island Mosannona pacifica Chatrou Ecuador - Esmeraldas 8531 (U) Appendix A Mosannona papillosa Chatrou Ecuador - Napo Pitman, N. (U) 974 AY743514 AY743500 AY743507 AY841483 Mosannona sp. Costa Rica - Chatrou, L.W. et al. 226 AY743509 AY743495 AY743502 AY841478 Monteverde 71 (U) Peru - Loreto - Chatrou, L.W. et al. 313 AY319064e AY319178e AY518869f AY841484 Yanamono 226 (U) Gabon Wieringa, J.J. et al. 54 AY743448 AY743467 AY743486 Mosannona vasquezii Chatrou • 233 Neostenanthera myristicifolia (Oliv.) Exell 14 3566 (WAG) • 234 Taxon Origin Voucher Onychopetalum amazonicum Brazil - Para Sperling, C.R. et al. • Appendix A R.E.Fr. Onychopetalum periquino DNA ID rbcL trnL-trnF matK psbA-trnH 972 DQ018222 DQ018175 DQ018261 DQ018237 17 AY319065e AY319179e AY518876f AY841485 629 AY841639b AY841717b 268 AY319066e AY319180e 326 AY841640b AY841718b AY841488 27 AY841528 AY841542 AY841489 254 AY841641b AY841719b AY841490 300 AY841642b AY841720b AY841491 103 AY841529 AY841543 AY841492 308 AY841643b AY841721b AY841493 80 AY841644b AY841722b AY841494 5925 (U) Bolivia - Beni Chatrou, L.W. et al. 425 (U) (Rusby) D.M.Johnson & N.A.Murray Oxandra asbeckii (Pulle) R.E.Fr. Guyana University of Guyana, AY841486 course Neotropical Botany UG-NB-55 (U) Oxandra espintana Peru - Chatrou, L.W. et al. (Spruce ex Benth.) Baill. Madre de Dios 133 (U) Oxandra euneura Diels Peru - Loreto Chatrou, L.W. et al. DQ018260 AY841487 249 (U) Oxandra laurifolia (Sw.) A.Rich Dominican Republic Maas, P.J.M. et al. 8375 (U) Oxandra longipetala R.E.Fr. Costa Rica - Osa Chatrou, L.W. et al. 114 (U) Oxandra macrophylla R.E.Fr. Peru - Loreto Chatrou, L.W. et al. 204 (U) Oxandra nitida R.E.Fr. Oxandra polyantha R.E.Fr. Brazil - Maas, P.J.M. et al. Rio de Janeiro 8821 (U) Peru - Loreto Chatrou, L.W. et al. 215 (U) Oxandra sphaerocarpa R.E.Fr. Peru - Loreto Maas, P.J.M. et al. 8226 (U) • 15 Taxon Origin Voucher Oxandra venezuelana R.E.Fr. Costa Rica - Osa Chatrou, L.W. et al. DNA ID rbcL trnL-trnF matK psbA-trnH 258 AY841645b AY841723b AY841495 280 AY841646b AY841724b AY841496 954 AY841647b AY841725b AY841497 505 AY743454 AY743473 120 (U) Oxandra xylopioides Diels Peru - Loreto Chatrou, L.W. et al. 165 (U) Piptostigma fasciculata Ghana (De Wild.) Boutique Piptostigma mortehani De Wild. Jongkind, C.C.H. et al. 1862 (WAG) Gabon Wieringa, J.J. et al. AY743492 AY841498 2779 (WAG) Piptostigma pilosum Oliv. Cameroon Wieringa, J.J. 2030 (WAG) 956 AY841648b AY841726b e AY319130e AY518827f AY518872f Polyalthia celebica Miq. Indonesia Mols, J.B. 9 (L) JB AY319016 Polyalthia discolor Diels Papua New Guinea Takeuchi & Ama JB114 AY319021e AY319135e JB17 AY319023e AY319137e AY841499 AY841500 16394 (L) Polyalthia glauca (Hassk.) Boerl. Indonesia Polyalthia suberosa (Roxb.) g India Mols, J.B. 20 (L) Chatrou, L.W. 480 (U) 536 AY238956 b Thwait. AY231289 AY841501 AY238965d AY841502 AY518873f AY841503 AY238949d Malaysia SAN 143918 (L) JB22 AY319039e AY319153e Pseudephedranthus fragrans Venezuela - Maas, P.J.M. et al. 33 AY841651b AY841729b (R.E.Fr.) Aristeg. Amazonas 6878 (U) Pseudomalmea diclina Peru - Loreto - Chatrou, L.W. et al. 305 AY319068e AY319128e (R.E.Fr.) Chatrou Yanamono 211 (U) Pseudomalmea diclina Peru - Chatrou, L.W. et al. 270 AY841530 AY841544 (R.E.Fr.) Chatrou Madre de Dios 136 (U) Polyalthia sumatrana Appendix A (Miq.) King AY841504 AY841398 AY841506 • 235 16 • AY841505 236 • Appendix A Taxon Origin Voucher Pseudoxandra bahiensis Maas Brazil - Bahia Mori, S.A. 12051 (U) Pseudoxandra longipes Maas Colombia - El Valle Monsalve B., M. DNA ID rbcL trnL-trnF matK psbA-trnH 1008 AY841653b AY841731b AY841508 1007 AY841532 AY841546 AY841509 307 AY319076e AY319190e 1005 AY841531 AY841545 AY841511 81 AY841654b AY841732b AY841512 110 AY841533 AY841547 88 AY841657b AY841735b 211 AY319051e AY319165e 222 AY319069e AY319183e AY841516 28 AY841658b AY841736b AY841439 318 AY743449 AY743468 1168 (U) Pseudoxandra lucida R.E.Fr. Peru - Loreto Chatrou, L.W. et al. AY319190f AY841510 213 (U) Pseudoxandra pacifica Maas aff. Colombia - El Valle Rooden, J. van et al. 565 (U) Pseudoxandra polyphleba Peru - Loreto (Diels) R.E. Fr. Maas, P.J.M. et al. 8227 (U) Pseudoxandra spiritus-sancti Brazil - Maas, P.J.M. et al. Maas Espirito Santo 8833 (U) Ruizodendron ovale Ecuador - Napo Maas, P.J.M. et al. (Ruiz & Pav.) R.E.Fr. AY841513 AY841514 8600 (U) Sapranthus viridiflorus Costa Rica - Chatrou, L.W. et al. G.E.Schatz La Selva 55 (U) Stenanona costaricensis R.E.Fr. Costa Rica - Chatrou, L.W. et al. Braulio Carillo 67 (U) Brazil - Amazonas Stevenson, D.W. et al. Tetrameranthus duckei R.E.Fr. AY841399 AY743493 AY841515 1002 (U) Trigynaea lanceipetala Peru - Loreto D.M.Johnson & N.A.Murray Chatrou, L.W. et al. AY743487 234 (U) Unknown genus sp. Peru - Huánuco Tello 3416 (NY) 750 AY841520 AY841534 Unonopsis elegantissima R.E.Fr. Peru - Loreto Chatrou, L.W. et al. 327 DQ018223 DQ018176 AY841456 DQ018263 DQ018238 250 (U) • 17 Taxon Unonopsis pittieri Saff. Unonopsis rufescens Origin Voucher DNA ID rbcL trnL-trnF matK psbA-trnH 223 AY841661b AY841739b DQ018264 AY841517 Costa Rica - Chatrou, L.W. et al. Braulio Carillo 68 (U) French Guiana Orava, C.E. 9 (U) 714 AY743455 AY743474 AY743494 AY841518 Peru - Loreto Chatrou, L.W. et al. 329 AY841662b AY841740b AY841400 AY841519 538 AY238957 AY231290 AY238966d AY841440 (Baill.) R.E.Fr. Unonopsis stipitata Diels 253 (U) Uvaria lucida Benth. g West Africa subsp. virens (N.E.Br.) Verdc. Xylopia frutescens Aubl. Botanische Tuinen 84GR00334 (U) Costa Rica - Osa Chatrou, L.W. et al. AY238950d 248 AY841667b AY841745b 106 (U) a Wang, Y.-L. et al. unpublished; b Chatrou et al. unpublished; (submitted); g Cultivated in Utrecht University Botanic Garden Appendix A • 237 18 • c Kojoma et al. (2002); d Sauquet et al. (2003); e Mols et al. (2004); f Mols et al. AY841441 238 • Annickia chlorantha (Oliv.) DNA ID Appendix A ndhF trnT-trnL trnS-trnG atpB-rbcL 976 AY841401 AY841571 AY841550 AY841370 977 AY841402 AY841572 AY841551 AY841371 pseudtrnL-F Setten & Maas Annickia pilosa (Exell) Setten & Maas Bocageopsis multiflora 190 DQ018199 (Mart.) R.E.Fr. Cananga odorata (Lam.) 239 AY841403 AY841548 AY841372 198 AY841404 AY841549 AY841373 601 AY841405 AY841573 AY841552 AY841374 1102 DQ018142 DQ018172 DQ018221 1100 DQ018140 DQ018170 DQ018220 598 DQ018125 DQ018150 DQ018206 312 AY841406 AY841574 Hook.f & Thomson Cleistopholis glauca Engl. & Diels Cremastosperma brevipes DQ018191 (DC.) R.E.Fr. Cremastosperma bullatum Pirie Cremastosperma bullatum Pirie Cremastosperma cauliflorum R.E.Fr. Cremastosperma cauliflorum AY841553 AY841375 DQ018192 R.E.Fr. Cremastosperma cenepense 1269 DQ018173 Pirie & Zapata • 19 A p p e n d i x A : Ta b l e 2 Taxon Taxon DNA ID ndhF trnT-trnL trnS-trnG atpB-rbcL pseudtrnL-F 594 DQ018123 DQ018148 DQ018193 741 DQ018129 DQ018154 DQ018194 1270 DQ018143 DQ018174 493 DQ018122 DQ018147 490 DQ018121 DQ018146 303 DQ018120 DQ018145 DQ018204 596 DQ018124 DQ018149 DQ018205 Cremastosperma napoense Pirie 737 DQ018127 DQ018152 DQ018208 Cremastosperma novogranatense 744 Cremastosperma leiophyllum R.E.Fr. Cremastosperma macrocarpum Maas Cremastosperma magdalenae Pirie Cremastosperma megalophyllum R.E.Fr. Cremastosperma megalophyllum R.E.Fr. Cremastosperma microcarpum R.E.Fr. Cremastosperma monospermum (Rusby) R.E.Fr. DQ018156 Appendix A R.E.Fr. Cremastosperma oblongum R.E.Fr. 600 DQ018126 DQ018151 DQ018207 • 239 Cremastosperma oblongum R.E.Fr. 739 DQ018128 DQ018153 DQ018209 Cremastosperma oblongum R.E.Fr. 1097 DQ018137 DQ018167 DQ018217 Cremastosperma panamense Maas 747 DQ018131 DQ018158 DQ018211 Cremastosperma pedunculatum 760 DQ018135 DQ018165 DQ018215 (Diels) R.E.Fr. 20 • Taxon 240 Cremastosperma pedunculatum DNA ID ndhF trnT-trnL • Appendix A 754 DQ018161 749 DQ018159 trnS-trnG atpB-rbcL pseudtrnL-F (Diels) R.E.Fr. Cremastosperma pendulum (Ruiz & Pav.) R.E.Fr. 751 DQ018132 DQ018160 DQ018212 1098 DQ018138 DQ018168 DQ018218 Cremastosperma sp. 745 DQ018130 DQ018157 DQ018210 Cremastosperma spec. A 245 DQ018119 DQ018144 DQ018203 Cremastosperma spec. B 758 Cremastosperma pendulum (Ruiz & Pav.) R.E.Fr. Cremastosperma pendulum (Ruiz & Pav.) R.E.Fr. DQ018164 Cremastosperma spec. B 756 DQ018133 DQ018162 DQ018213 Cremastosperma venezuelanum 757 DQ018134 DQ018163 DQ018214 761 DQ018136 DQ018166 DQ018216 1101 DQ018141 DQ018171 1099 DQ018139 DQ018169 Pirie Cremastosperma westrae Pirie Cremastosperma yamayakatense Pirie Cremastosperma yamayakatense DQ018219 Pirie Cremastosperma yamayakatense 742 DQ018155 Pirie Ephedranthus sp. 105 AY841407 AY841575 AY841554 AY841376 Greenwayodendron oliveri 958 AY841408 AY841576 AY841555 AY841377 (Engl.) Verdc. • 21 Taxon DNA ID ndhF trnT-trnL trnS-trnG atpB-rbcL pseudtrnL-F Klarobelia inundata Chatrou 301 AY841409 AY841577 AY841556 AY841378 Malmea dielsiana R.E.Fr. 260 AY841410 AY841578 AY841557 AY841379 DQ018195 Malmea sp. 197 AY841411 AY841579 AY841558 AY841380 DQ018196 Malmea surinamensis Chatrou 850 Monocarpia euneura Miq. DQ018197 JB106 AY841412 AY841580 AY841559 AY841381 Mosannona costaricensis R.E.Fr. 237 AY841413 AY841581 AY841560 AY841382 Onychopetalum amazonicum R.E.Fr. 972 AY841414 AY841582 AY841561 AY841383 Onychopetalum periquino (Rusby) 17 DQ018198 D.M.Johnson & N.A.Murray Oxandra espintana 268 DQ018189 (Spruce ex Benth.) Baill. Piptostigma mortehani De Wild. Polyalthia discolor Diels Polyalthia suberosa (Roxb.) 505 AY841415 AY841583 AY841562 AY841384 JB114 AY841416 AY841584 AY841563 AY841385 536 AY841417 AY841585 AY841564 AY841386 JB22 AY841418 AY841586 AY841565 AY841387 305 AY841419 AY841587 AY841566 AY841388 Thwait. Polyalthia sumatrana (Miq.) King Appendix A Pseudomalmea diclina (R.E.Fr.) Chatrou • 241 Pseudoxandra lucida R.E.Fr. 307 AY841420 AY841588 AY841567 AY841389 Pseudoxandra spiritus-sancti Maas 110 AY841421 AY841589 AY841568 AY841390 Sapranthus viridiflorus G.E.Schatz 211 AY841422 AY841590 AY841569 AY841391 Unonopsis elegantissima R.E.Fr. 327 Unonopsis pittieri Saff. 223 Unonopsis stipitata Diels 329 22 • DQ018200 DQ018201 AY841423 AY841591 AY841570 AY841392 DQ018202 242 • Appendix A Appendix A • 243 244 • Appendix A Appendix A • 245 246 • Appendix A Appendix A • 247 248 • Appendix A Appendix B F i g . 1 ( C h a p t e r 2 , F i g . 2 . ) ( a ) Cremastosperma leiophyllum R.E.Fr. (Pirie et al. 2, Bolivia), flower before anthesis; (b) Cremastosperma leiophyllum R.E.Fr. (Pirie et al. 2, Bolivia), immature fruit; (c) Guatteria megalophylla Diels (Chatrou et al. 387, Bolivia), flower before anthesis; (d) Guatteria megalophylla Diels (Chatrou et al. 387, Bolivia), almost mature fruit; (e) Mosannona vasquezii Chatrou (Chatrou et al. 226, Peru), flower before anthesis; (f) Mosannona sp. (Chatrou et al. 72, Costa Rica), immature fruit; (g) Duguetia marcgraviana Mart. (Chatrou et al. 327, Bolivia), immature fruit; (h) Duguetia marcgraviana Mart. (Chatrou et al. 329, Bolivia), flower at anthesis. Photographs by L.W. Chatrou. Appendix B • 249 a b F i g . 2 . Cremastosperma brevipes (DC.) R.E.Fr. a. flower buds; b. fruit (French Guiana; photos: S. Mori) a b F i g . 3 . Cremastosperma bullatum Pirie a. leaf base, illustrating bullate (‘bubbled’) appearance and long golden hairs; b. flowers (a: Pirie et al. 71; b: Pirie et al. 94, Peru - Amazonas; photos: MDP) 250 • Appendix B a b F i g . 4 . Cremastosperma cauliflorum R.E.Fr. a. branching inflorescence; b. illustrating closure of inner petals on maturing of flower (Maas et al. 9021, Brazil - Acre; photos P.J.M. Maas) a b F i g . 5 . Cremastosperma leiophyllum R.E.Fr. flower at anthesis: a. illustrating closure of inner petals on maturing of flower; b. colour change in stigmas (compare with Appendix 2.1a) (Pirie et al. 2, Bolivia - Santa Cruz; photos L.W. Chatrou) Appendix B • 251 b a F i g . 6 . Cremastosperma megalophyllum R.E.Fr. a. habit; b. mature fruit (a: Maas et al. 8577; b: Maas et al. 8595, both Ecuador - Napo, photos P.J.M. Maas) a b c F i g . 7 . Cremastosperma microcarpum R.E.Fr. a. & b. flower; c. fruit (a & b: Maas et al. 8222; c: Maas et al. 6281, both Peru - Loreto, photos P.J.M. Maas) 252 • Appendix B b a c F i g . 8 . Cremastosperma monospermum (Rusby) R.E.Fr. a. habit; b. leaf (illustrating often cuspidate leaf apex); c. flower bud (Pirie et al. 5, Bolivia - Pando; photos L.W. Chatrou) a b F i g . 9 . Cremastosperma oblongum R.E.Fr. a. flower; b. fruit (Maas et al. 9148, Brazil - Acre; photos P.J.M. Maas) Appendix B • 253 F i g . 1 0 . Cremastosperma pendulum (Ruiz & Pav.) R.E.Fr. habit and flower bud (Pirie et al. 33, Peru - Pasco; photo MDP) 254 • Appendix B a c b F i g . 1 1 . Cremastosperma yamayakatense Pirie a. flower bud; b. habit and fruit; c. mature fruit (a: Pirie et al. 57; b: Pirie et al. 58; c: Pirie et al. 60; all Peru - Amazonas, photos MDP) F i g . 1 2 . a. Bolivia - Pando, Nov. 2001. From left: A. Saucedo; L.W. Chatrou, T. Scharaschkin (then at the University of California, Davis); N. Divico (Don Nico); C. Salas; Appendix B • 255 F i g . 1 2 b. Peru - San Martín: Estación ‘Biodiversidad’, Dec. 2003. From left: T. Diaz F.(Universidad Nacional San-Martín, Tarapoto); M. Zapata C. (Universidad Privada Antenor Orrego, Trujillo); R. van Velzen and M. Botermans (NHN - Wageningen); M. Gonzales I.; MDP. F i g . 1 2 c. Peru - Amazonas: Comunidad Yamayakat, Nov. 2003. R. Apanu N. (left); J.S. Kasen; displaying local craft. 256 • Appendix B