Smith et al., 2008

Smith et al., 2008

22 Pages · 2008 · 546 KB · English

tropics, Asia, Pacific Islands, Australia, and Africa to better protocols used for other taxa (Smith et al. migration to Africa (vicariant model) (Fig. 2).

Smith et al., 2008 free download

ORIGINALARTICLE Placing the origin of two speciesrich genera in the late cretaceous with later species divergence in the tertiary: a phylogenetic, biogeographic and molecular dating analysis of Piper and Peperomia (Piperaceae) Jame sFSmith ?Ange la C Steve ns ? Eric J Tepe ?Chris Davidson Received: 13 December 2007 /Accepted: 29 April 2008 /Published online: 28 June 2008 !SpringerVerlag 2008 Abstrac t Nearly all of the species diversity in Piper aceae isencompas sed within Piper and Pepero mia Both genera are pantro pical with area sofdiversi?cat ion in the Neo tropic sand Southeast Asia Piperace ae are less diverse in Africa with only two native species of Piper This stud y examin es the distributio nofboth Piper and Pepero mia with represe ntative sam ples from the Neotropic s, Asia, Pa ci?c Islands ,and Africa Mol ecular dating isused toplac eanage for the crown clades of Piper and Pepero mia as well as ages for diversi? cation within the clad es Both gener ahave ori gins in the late Cret aceous, but species level dive rsi?catio n occurred much later in the Tertiary Biogeo graphy of both genera are correl ated with paleo climate evid ence to better explain the distributio nand dive rsi?catio nofthese large genera Keywor ds Piperace ae !Piper !Pepero mia !g3pd ! trn LF !trn Lintron !ndh F!Biogeo graph y Intro duction The distrib ution of the world?s biota, and how it got to where itpresently is, has been aquestion of systemat ics and evol ution since Darw in (1859 )Moder ntools are able to improv eour abili ty to assess bioge ography Strongly suppor ted phylogenetic hypot heses, and molecu lar seque nce age estimat ions have grea tly enhanced our under standin gofplant and animal distribut ion and move ment (Thorne et al 1998 ;Rambaut and Br omham 1998 ; Kishin oetal 2001 ;Thorne and Kishino 2002 ;ArisBros ou and Yan g2002 ;Sand erson 20 02,2003 ;Drumm ond and Ramba ut 2005 ;Ree etal 2005 ;Smi th 2006 )For exam ple, hypot heses on the puta tive number of plan tintrodu ctions into the Hawaii an islands were initially based on mor pholo gical differ ences and simi larities to pres umed ancestr al groups whi ch often resu lted in sugges tions that many plant taxa had arrive d via multipl e inde pendent introdu ctions (Wagner et al 1990 )Phylogenet ic analyses , howev er, have indicated that asingle introdu ction was more of the norm for most groups where several separ ate introdu ctions had been propos ed (Baldwin et al 1990 , 1991 ;Baldwin and Rob ichaux 1995 ;Givni sh et al 1996 ; Cronk et al 2005 )and in at lea st on ecase, two introduc tions were propos ed whe re only one had been hypot hesized (Howa rth et al 19 97) Recent efforts have been espec ially informat ive regardi ng biogeogra phy ofislands (D eJoode and Wendel 1992 ;Knox et al 1993 ;Wagner and Funk 199 5;Givnish et al 1995 ,1996 ;Howar th et al 1997 ,2003 ;Stues sy et al 1998a ,b;Alic eand Ca mpbell 1999 ;Ganders et al 2000 ; Helfg ott et al 200 0;Wrig ht et al 2000 ,2001 ;Lindqv ist and Alb ert 2002 ;Nep okroeff et al 2003 ;Carine et al 2004 ;Cronk et al 2005 )and norther n hemisp here or temper ate groups (Wen and Stuessy 1993 ;Qiu et al 1995 ; Wen and Janse n1995 ;Lee et al 1996 ;Wen et al 1996 ; Wen and Zim mer 1996 ;Hasebe et al 1998 ;Kim and Jansen 1998 ;Schnabe land Wende l1998 ;Xiang et al 1998 ;Wen et al 1998 ;Kim and Kim 1999 ;Swen son et al JFSmith (&)!A C Stevens Department ofBiological Sciences, Boise State University, 1910 University Drive, Boise, ID 837251515, USA email: [email protected] E JTepe Department ofBiology, University ofUtah, Salt Lake City, UT 84112, USA C Davidson Idaho Botanical Research Foundation, 637 Warm Springs, Boise, ID 83712, USA 123 Plant Syst Evol (2008) 275:9?30 DOI 101007/s0060600800565 2001 ;Bell and Dono ghue 200 5;Alsos et al 2007 ;Bell 2007 ), but attent ion has general ly been lacking for trop ical organ isms This is espec ially true for plan ts with species or clades in So uth America ,Southe ast Asia and Africa (Givnish et al 2000 ,2004 ;Conti et al 2002 ;Dav is et al 2002 ) Vicariance has often been used to expl ain distribut ions that cros smajor barrie rs to dispers al (N elson and Platni ck 1981 ;Humphrie sand Paren ti 1999 ;Conti et al 2002 ;Crisp and Cook 2007 ), especia lly for clades with rel atively older origins Howeve r,much recent data have begun to show that many plan tand animal groups that seemed to ?t a vicariant distribut ion pattern are actual ly the resu lt of more recent, long distance dis persal events (Renne r 2004 ;Sanm art? ?n and Ronquist 2004 ;deQue iroz 2005 ;McG lone 2005 ;Poux et al 2005 ; Winkwo rth et al 2005 ;Smith et al 2006 ;Morr is et al 2007 ) Piperace ae provide ameans of examin ing the biogeog raphy of apantro pical group Piper aceae consi st of only ?ve genera, and the vast majority of the estimated 3,600 species occur nearly equal ly in two genera, Piper and Pepero mia Both gener a are found in tropical regions throughout the wor ld (Miquel 1843 ?1844; Grieg 2004 ; Wanke et al 20 07a ;Quija noAbril et al 2006 )Previous phylo genetic anal yses that have examin ed Piper on the global level iden ti?ed three major clades, one each in the Neotr opics, Southe ast Asia, and Paci?c Islands (Jaramill o and Manos 2001 ;Jaramil lo et al in pres s) The origin ofthe Afri can species of Piper pose an interest ing question Th ere are tw ospecies indi genous to Africa Piper guineense isadioecio us vine ,rem iniscent of the major ity of the species of Southe ast Asia Piper capens eis ashrub with bisexual ?o wers, more simi lar in habit and ?o ral form to species of the Neo tropics The African species imply either asing le ori gin for the genus on the continent and ashift in bree ding system ,ormul tiple origins Likewise unusua listhe pauci ty of species Africa was previous ly dominated by lowland tro pical rainfor est and Neogene aridi?c ation isthought to have resulted in the extincti on of man yAfri can tropical lineage swith the result that the African ?ora is less diverse than that of Asia or South Amer ica (Axel rod 1972 ;Axelrod and Raven 1978 ; Raven and Axelrod 1974 ;Whitmore 1998 )Alte rnativ ely, the low diversity of Piper may re?ect a mor e recent introdu ction event into Africa with insu f?cient time for species diversi ?cation as has been proposed for Rapat ea ceae (Givnish et al 2004 ) Most speci es of Piper tend to occupy relat ively small geogr aphic range s (Ma rquis 2004 ) Some species are widesp read (eg Pi umbellat um ), spre ad on purpos efor their utilit y(Pi nigrum, Pi methys ticum )orhave been distribut ed by accident or have escaped from cultivation (Pi auri tum )Howeve r,most species of Piper are know n only from asingle geogr aphic regi on, island, or continent (Marqui s 2004 )with a few exceptions that are often thoug ht to be recent humanm ediated introdu ctions (eg Pi aduncum ) The distribution al pattern is simi lar in Pepero mia ,and again som e species are geogr aphically widesp read Some of this distribut ion of Pepero mia may have occurred as acom bined resu ltof their wee dy nature and thriving in disturbed areas, humanmedi ated intro ductions and produc ing an abunda nce of small seedlike fruits (eg Pe pellucida )Howeve r,othe rspecies are found in relative ly natural and undisturbed areas, sugges ting that their distributio ns are not the resu ltof human interv ention Pepero mia tetraph ylla issuch an exampl ewith populations know nfrom China, the Neo tropics, Paci?c Islands, Africa, Austr alia, New Zealand, Nepal, India and Indone sia (S Wanke ,personal com munica tion) Itis the goal of this stud ytounder take aphylogenet ic analysi sofPiper and Pepero mia species from the Neo tropic s,Asia, Paci? cIslands ,Aus tralia, and Afri ca to better resolve the biogeogra phic relations within this large pan tropic al group ofplants We have sampled from all geogr aphic regions for Piper and nearly all for Pepero mia, includi ng indivi duals of three Peperomia species from different geographi cregio ns Our phylo genetic anal ysis is based on DNA seque nces of three chlo roplast regi ons: ndh F, trn Lintron, and trn Ltrn Fspacer, and the low copy nuclear gene glyc eraldehyde 3phosphat edehydrogena se (g3pd ;Strand et al 1997 ) Materi als and method s Species used in the analysis, vouche rinformat ion and Genbank accessi on numb ers for all seque nces are listed in Table 1Data are availa ble from the ?rst auth or on reque st DNA was extract ed using Qiagen DNeas yextracti on kits followi ng the man ufacturer? sinstru ctions Ampli?ca tion of ndh F, trn L intron, and the trn Ltrn F space rfollowed protocol sused for other taxa (Smith et al 2004 ) The ampli? cation, cloning and seque ncing of the low copy numb er nuclear gene g3pd used the prime rsof Stra nd etal (1997 )with the addition of BSA (?nal conce ntration of 1lg/lL) and 2% DM SO to the reaction mix Therm al cycling para meters enta iled an initial cycle of 3min at 94"C, followed by 2cycles of 1min at 94"C, 1min at 52"Cand 15 min at72"Cfol lowed by cycles of ste pdown annea ling temperat ures, each repe ated twice ranging from 52 to 40"C with a?nal 24 cycles at 39"C The pGEM T vector system kit (Promeg a) was used to clone g3 pdUpto 20 clon es for some species were sequenced in an attem pt to detect potential paralogs Sequenc es were obtaine dfrom both strands of all gene regions with aLiCor Longre adIR 4200 automa ted sequencer 10 JFSmith etal 123 Table 1Species used inthis analysis, their collection locality, voucher information, and Genbank accession numbers Species Voucher (herbarium acron ym) Collection locality Genbank numbers, g3pd , trnLtrnF,trnLintron, ndh F Pi aduncum L A Bornstein 765 (SEMO) Nicaragua EU519540, EU519630, EU519810, EU519720 Pi aduncum L L A Hahn 438 (SRP) Dominican Republic EU519539, EU519629, EU519809, EU519719 Pi aequale Vahl A Bornstein 716 (SEMO) Honduras EU519550, EU519640, EU519820, EU519730 Pi amalago L A Bornstein 712 (SEMO) Honduras EU519543, EU519633, EU519813, EU519723 Pi arboreum Aubl Aburri Botanical Garden collection number 042 (Aburri Botanical Garden) Cultivated, Ghana EU519525, EU519610, EU519795, EU519705 Pi arboreum Aubl A Bornstein 699 (SEMO) Honduras EU519524, EU519614, EU519794, EU519704 Pi augustum Rudge E Tepe 525 (MU) French Guiana EU519644, EU519634, EU519814, EU519724 Pi auritum H B &K CDavidson 10892 (SRP) La Selva, Costa Rica EU519535, EU519625, EU519805, EU519715 Pi auritum H B &K ARincon 2409 (XAL) Veracruz, Mexico EU519534, EU519624, EU519804, EU519714 Pi austrocaledonicum C DC G McPherson 19190 (MO) New Caledonia EU519508, EU519598, EU519778, EU519688 Pi avellanum C DC E Tepe 616 (MU) French Guiana EU519553, EU519643, EU519823, EU519733 Pi betle L C Davidson 11001 (SRP) Tanzania EU519505, EU519595, EU519775, EU519685 Pi betle L JFSmith 5808 (SRP) Cultivated EU519506, EU519596, EU519776, EU519686 Pi borbonense C DC Accession number 873616 for Conservatoire etJardins Botaniques de Nancy, not vouchered Cultivated, originally from Reunion EU519495, EU519585, EU519765, EU519675 Pi caninum Blume T Flynn 6750 (PTBG) Cultivated, Australia EU519501, EU519591, EU519771, EU519681 Pi capense L f JFSmith 4925 (SCA) Cameroon EU519488, EU519578, EU519758, EU519668 Pi capense L f C Davidson 11009 (SRP) Kenya EU519489, EU519579, EU519759, EU519669 Pi chandocanum C DC Li JM 06182 (PE) Guangxi, China EU519515, EU519605, EU519785, EU519695 Pi colonense C DC A Bornstein 753 (SEMO) Nicaragua EU519538, EU519628, EU519808, EU519718 Pi concepcionis Trel C Davidson 10880 (SRP) Costa Rica EU519545, EU519635, EU519815, EU519725 Pi costatum C DC T Flynn 4261 (PTBG) Cultivated EU519499, EU519589, EU519769, EU519679 Pi ?avi?orum C DC Li JM 06171 (PE) Yunnan, China EU519518, EU519608, EU519788, EU519698 Pi guahamense C DC T Flynn 6748 (PTBG) Cultivated EU519498, EU519588, EU519768, EU519678 Pi guazacapanense Trel & Standl M A Perez Farrera 2923 (HEM) Chiapas, Mexico EU519547, EU519637, EU519817, EU519727 Pi guineense Schumach & Thonn C Davidson 11005 (SRP) Uganda EU519492, EU519582, EU519762, EU519672 Biogeography ofPiper 11 123 Table 1continued Species Voucher (herbarium acron ym) Collection locality Genbank numbers, g3pd , trnLtrnF,trnLintron, ndh F Pi guineense Schumach & Thonn C Davidson 11006 (SRP) Uganda EU519493, EU519583, EU519763, EU519673 Pi guineense Schumach & Thonn C Davidson 11008 (SRP) Kenya EU519494, EU519584, EU519764, EU519674 Pi guineens eSchumach & Thonn JFSmith 4923 (SCA) Cameroon EU519490, EU519580, EU519760, EU519670 Pi guineense Schumach & Thonn JFSmith 4924 (SCA) Cameroon EU519491, EU519581, EU519761, EU519671 Pi hancei Maximowicz Li JM 06211 (PE) Guizhou, China EU519512, EU519602, EU519782, EU519692 Pi hispidum Sw A Rincon 2327 (XAL) Veracruz, Mexico EU519541, EU519631, EU519811, EU519721 Pi hostmannianum (Miq) C DC E Tepe 599 (MU) French Guiana EU519545, EU519636, EU519816, EU519726 Pi humistratum Go?rts &KU Kramer E Tepe 542 (MU) French Guiana EU519549, EU519639, EU519819, EU519729 Pi imperiale (Miq) C DC C Davidson 10882 (SRP) Costa Rica EU519552, EU519642, EU519822, EU519732 Pi methysticum G Forster C W Morden 2957 (HAW) Hawaii EU519496, EU519586, EU519766, EU519676 Pi methysticum G Forster C W Morden 2975 (HAW) Hawaii EU519497, EU519587, EU519767, EU519677 Pi muricatum Blume W SWong 2(SRP) Malaysia EU519526, EU519616, EU519796, EU519706 Pi nigrum L JFSmith 5807 (SRP) Cultivated EU519509, EU519599, EU519779, EU519689 Pi nudifolium C DC C Davidson 10897 (SRP) Costa Rica EU519536, EU519626, EU519806, EU519716 Pi obliquum Ruiz &Pav A Bornstein 738 (SEMO) Nicaragua EU519537, EU519628, EU519807, EU519717 Pi peltatum L C Davidson 10898 (SRP) Costa Rica EU519532, EU519622, EU519802, EU519712 Pi peltatum L E Tepe 589 (MU) French Guiana EU519533, EU519623, EU519803, EU519713 Pi pingbienense Y C Tseng Li JM 06151 (PE) Yunnan, China EU519517, EU519607, EU519787, EU519697 Pi porphyrophyllum N E Br W SWong 3(SRP) Malaysia EU519519, EU519609, EU519789, EU519699 Pi pseudofuligineum C DC A Bornstein 710 (SEMO) Honduras EU519542, EU519632, EU519812, EU519722 Pi puberulum (Benth) Maxim T Flynn 6749 (PTBG) Cultivated EU519500, EU519590, EU519770, EU519680 Pi rothianum FM Bailey JFSmith 6545 (SRP) Queensland, Australia EU519502, EU519592, EU519772, EU519682 Pi sanctum Schltdl ex Miq A Bornstein 744 (SEMO) Nicaragua EU519521, EU519611, EU519791, EU519701 Pi sanctum Schltdl ex Miq H G Domingues 251 (HEM) Chiapas, Mexico EU519520, EU519610, EU519790, EU519700 Pi sanctum Schltdl ex Miq A Rincon 2352 (XAL) Veracruz, Mexico EU519522, EU519612, EU519792, EU519702 Pi sanctum Schltdl ex Miq M A Perez Farrera 2820 (HEM) Chiapas, Mexico EU519523, EU519613, EU519793, EU519703 Pi sarmentosum Roxburgh JFSmith 5806 (SRP) Cultivated EU519510, EU519600, EU519780, EU519690 12 JFSmith etal 123 Table 1continued Species Voucher (herbarium acron ym) Collection locality Genbank numbers, g3pd , trnLtrnF,trnLintron, ndh F Pi semiimmersum C DC Li JM 06161 (PE) Yunnan, China EU519514, EU519604, EU519784, EU519694 Pi submultinerve C DC Li JM 061410 (PE) Yunnan, China EU519516, EU519606, EU519786, EU519696 Pi subpenninerve Ridl W SWong 1(SRP) Malaysia EU519503, EU519593, EU519773, EU519683 Pi thomsonii (C DC) Hook f Li JM 061511 (PE) Yunnan, China EU519513, EU519603, EU519783, EU519693 Pi umbellatum L A Rincon 2325 (XAL) Mexico EU519528, EU519618, EU519798, EU519708 Pi umbellatum L JFSmith 4926 (SCA) Cameroon EU519529, EU519619, EU519799, EU519709 Pi umbellatum L C Davidson 11010 (SRP) Kenya EU519530, EU519620, EU519800, EU519710 Pi umbellatum L C Davidson sn (SRP) Tanzania EU519531, EU519621, EU519801, EU519711 Pi umbellatum L A Bornstein 709 (SEMO) Honduras EU519527, EU519617, EU519797, EU519707 Pi urophyllum C DC ex Pittier C Davidson 10902 (SRP) Costa Rica EU519551, EU519641, EU519821, EU519731 Pi wallichii (Miq) HandelMazzetti Li JM (PE) Yunnan, China EU519511, EU519601, EU519781, EU519691 Pi yucatanense C DC M A Perez Farrera 2898 (SRP) Tabasco, Mexico EU519548, EU519638, EU519818, EU519728 Pi sp Harder 5565 (MO) Vietnam EU519507, EU519597, EU519777, EU519687 Pi sp C Davidson 11000 (SRP) Tanzania EU519504, EU519594, EU519774, EU519684 Peperomia blanda (Jacquin) Kunth Li JM 06262 (PE) Guangxi, China EU519471, EU519561, EU519741, EU519651 Pe blanda (Jacquin) Kunth JE Burrows 8521 (herbarium atBuffelskloof Private Nature Reserve, Mpumalanga,South Africa) South Africa EU519470, EU519560, EU519740, EU519650 Pe caperata Yunck JFSmith 4029 (SRP) Cultivated NA, EU519556, EU519736, EU519646 Pe cavalierii C DC Li JM 06201 (PE) Guangxi, China EU519472, EU519562, EU519742, EU519652 Pe emarginella C DC C Davidson 10867 (CR) Costa Rica EU519486, EU519576, EU519756, EU519666 Pe fernandopoiana C DC JFSmith 4927 (SCA) Cameroon EU519475, EU519565, EU519745, EU519655 Pe glabella (Sw) A Dietr C Davidson 10895 (SRP) Costa Rica EU519484, EU519574, EU519754, EU519664 Pe hernandiifolia Trel C Davidson 10905 (CR) Costa Rica EU519479, EU519569, EU519749, EU519659 Pe hesperomannii Wawra Hawaiian Plant DNA Laboratory 1783 (HAW) Hawaii EU519481, EU519571, EU519751, EU519661 Pe kamerunana C DC JFSmith 4928 (SCA) Cameroon EU519480, EU519570, EU519750, EU519660 Pe leptostachya Hook & Arnott Hawaiian Plant DNA Laboratory 37 (HAW) Hawaii EU519483, EU519573, EU519753, EU519663 Biogeography ofPiper 13 123 Align ment of all sequences was done manual ly Prior to combini ng g3pd and cpDNA seque nces an ILD test (Farris et al 1994 )was run, impleme nted as the partition homo geneity test (PH T) of PAUP* As this test is often sensit ive to low sign al in some of the partitions (Reeve s et al 2001 ;Yoder et al 2001 ),separate max imum pars i mony (MP) boots trap (BS; Felsenst ein 1985 )analyses were run for each of the partitions (cpDNA and g3pd )to search for areas of hard incongrue nce (Seelane n etal 1997 ) Th e data were analy zed using PAUP* 40b10 (Swoff ord 2002 )for MP and BS Indels were either treated as mi ssing data, or rescore d followi ng the methods of Simmo ns and Ochoterena (2000 )Fo rthe MP analy ses, the data wer eanal yzed using the search option of Olms tead and Palmer (1994 )where ?v e searches are performed using 1,000 replicates with nearest neighbor interchange and Mul Trees Off savi ng all shortes ttre es Th eresults of each of these searches are then used as the sta rting trees for asearch usin gtreebise ction reconnect ion (TB R) and MulT rees on The shortes ttrees from all searches were combine dand astric tconse nsus was crea ted Boo tstrap replicat es were performed with 100 searches with 10 rando m replic ates using full heuristic search with TBR and MulT rees on Modelt est Ver sion 327 (Posada and Cranda ll 1998 ) was used to determin ethe best mode lthat ?t the data for Bayesi an inf erence (BI) anal yses Re sults of these tes ts were used as priors in BI analyses with acom plex (di f ferent model for each partition) or simple mode lfor the combine d data using either the likelihoo d ratio test (LRT), or Aikake informat ion content (AIC) cri teria (Table 2) Bayesi an infere nce analyses emp loyed MRBAY ES 30B4 (Huelsenbe ck and Ron quist 2001 ) Ana lyses were run using two milli on generations saving trees every 100 generat ions with aburn in of 25,000 generations Ana lyses Table 1continued Species Voucher (herbarium acron ym) Collection locality Genbank numbers, g3pd , trnLtrnF,trnLintron, ndh F Pe pellucida (L) Kunth Aburri Botanical Garden 040 (Aburri Botanical Garden Herbrium) Ghana EU519477, EU519567, EU519747, EU519657 Pe pellucida (L) Kunth JFSmith 4929 (SCA) Cameroon EU519476, EU519566, EU519746, EU519656 Pe pellucida (L) Kunth E Tepe 578 (MU) French Guiana EU519478, EU519568, EU519748, EU519658 Pe retusa A Dietr JE Burrows 8923 (Buffelskloof herbarium atBuffelskloof Private Nature Reserve, Mpumalanga,South Africa) South Africa EU519473, EU519563, EU519743, EU519653 Pe retusa A Dietr C Davidson 11003 (SRP) South Africa EU519474, EU519564, EU519744, EU519654 Pe sandwicensis Miq Hawaiian Plant DNA Laboratory 1784 (HAW) Hawaii EU519482, EU519572, EU519752, EU519662 Pe serpens Sw E Tepe 548 (MU) French Guiana EU519487, EU519577, EU519757, EU519667 Pe tetraphylla (G Forster) Hook &Arnott C Davidson 10907 (SRP) South Africa NA, EU519557, EU519737, EU519647 Pe tetraphylla (G Forster) Hook &Arnott JE Burrows 8924 (Buffelskloof herbarium atBuffelskloof Private Nature Reserve, Mpumalanga,South Africa) South Africa NA, EU519558, EU519738, EU519648 Pe tetraphylla (G Forster) Hook &Arnott Li JM 061417 (PE) Yunnan, China NA, EU519559, EU519739, EU519649 Pe urocarpa Fisch &CA Mey C Davidson 10889 (SRP) Costa Rica EU519485, EU519575, EU519755, EU519665 Houttuynia cordata Thunb JFSmith 4922 (SRP) Cultivated EU519469, EU519555, EU519735, EU519645 Saururus cernuus L A Bornstein 863 (SEMO) Cultivated NA, EU519554, EU519734, EU519644 NA not ampli?ed 14 JFSmith etal 123 were repeated for each run with the order of species scramb led betwee nruns The Kishino?H asegawa and Shimo daira?Has egawa tests (K H, Kishi no and Has egawa 1989 ;SH, Shimodaira and Hasegawa 1999 )were both used to test trees where constra ints were impose donthe search to make geogr aphic regions (N eotropics and Af rica for Piper ;Asia, Neotr opics, Africa, and Paci?c Islands for Peperomi a)that were not mono phyletic into mono phyleti cgroups ,aswell as con strain Pe ret usa as mono phyletic We include resu lts from both tests butacknow ledge that there are problems with KH tes ts(Go ldman et al 2000 )Paramet ers from the AI C selection criteria of Model test for the com bined data set were used to com pute the likel ihoods since AIC has been shown to provi de better resu lts (Posada and Buc kley 2004 ) Search para meters utilized optim ization with 1,000 repli cate bootstrap sample s Ancest ral distribut ion areas were construc ted with dis persal vicariance analysi s(DIV A Version 11; Ronquist 1996 ,1997 )DIVA minimi zes the numb er of disp ersal and extincti on events neede dtoexplain the patter nofter minal taxa We scored terminal taxa based on their curr ent dis tributio n For the mos tpart this isrelat ively easy as species of Piper tend to be fairly local endem ics (Marqui s2004 ) Howeve r,for some species we include dacce ssions in our phylo genetic analysi sthat were not from their presu med native distri bution (eg, Pi umbellat um )In case swhere we assumed a species had been introdu ced (cultivated species )wescored the ter minals for their native range Piper cani num isknow nfrom both Australia and Malaysia (Chew 1972 )The accession we used was from cultivation and presuma bly Aus tralian in origin, however ,since the species has abroader distrib ution we scored itas pres ent in both Australia and Asia For Pepero mia species, thi sposes agrea ter challenge since som especies are more wide spread and their ancestral native range is not currently know n For Pepero mia we scor ed current distributio n based on the localit y for which the individual in the analysi swas from Multiple analyses usin gdifferent bio geogr aphic regions as alternate tests als o were run Previous phylo genetic anal yses had indicated that Piper ?t three major clades that correspo nded to three disti nct biogeo graphic regi ons; Asia, Paci?c Islands ,and Neo tropic s(Jaram illo and Manos 2001 )We used these as our basis, but sinc eweinclude daddi tional species from China, Malaysi a, Australia, Afri ca, and Reunion, we did addi tional analyses with these areas separ ated from the main three To disti nguish amo ng the effect sofPepero mia and Piper on the results, we also analyzed the two clades independe ntly of each other DIVA als orequires afully bifurca ted tree, theref ore we arbitrar ily selected asing le MP tree with arbitrary relat ionships among the polytomy of Asian species that was reco vered Previous analyses (Jaram illo and Manos 200 1;Jara millo et al in press )have indicated that the Neo tropical species of Piper are likely to be a mono phyletic group To examine the impact of altering the posi tion of Pi sanctum ,Pi aequal e,and Pi urophy llum ,wealso exam ined biogeogra phic patterns in DIVA const raining these three species at the base of the Neotr opical clade with Pi sanctum sister to all remainin g Neotr opical species DIVA does not allow the user to separate the histor ies of the areas that taxa are found in and the lineages of the taxa themse lves (Ree et al 2005 ) Since inf ormation on the geogr aphic regions of Piper aceae is known, it woul dbe valuabl etoinclude the inform ation as part of the model There fore, we also employe dAReA v 21 (Smith 2006 )as a mea ns of re assessing the biogeogra phic histor y of Piperace ae AReA require sanultramet ric tree, therefore, we uti lized the program r8s (Sande rson 2003 )toput dates on our tree In addi tion to an ultramet ric tree, AReA allow sthe user to imple ment priors rega rding potential connec tions betwee ngeographi cregions, which can vary over geolog ical time Other priors include the likel ylongevity of taxa and their vagility We designat ed our global distri bution to compri se ?ve regions: Australia (including New Caledo nia), Asia, Af rica, the Neo tropic s, and the Paci? cIslands (excludi ng New Ca ledonia) We tested two differ ent models of connec tions The ?rst of these isdesignated the dispers al model since itre?ects potential connec tions that may have been pres ent due to dispers al, gener ally longer distanc es and over water barr iers The followi ng connec tions were set to an arbitrary nu mber of 025 prior to Table 2Details ofthe different DNA regions used inthis analysis Size average (range) ndh F trnLintron trnLtrnFspacer g3pd 2,521 (2,320?2,583) 499 (413?533) 423 (357?510) 838 (715?1,007) Aligned size 2,705 617 704 1,209 Number ofpotentially informative characters (MP) 908 211 169 742 Modeltest result (LRT) TVM +G HKY +G+I Modeltest result (AIC) TVM +G TVM +G+I Note that the three cpDNA markers were combined into asingle partition inall analyses Biogeography ofPiper 15 123 60 mya to re?ec tthe lower possibi lity of disp ersal betwee n these regi ons for the disp ersal model: Paci?c Islands to the Neotr opics (Sm ith et al 2006 ),Paci ?c Islands to Australia (Wrigh tetal 2000 ,2001 ),Paci? cIsl ands to Asia (Cronk et al 2005 ),and Aus tralia to Asia (Cronk et al 2005 )The only othe rconnectio ns allowed were Afri ca to the Neo tropic s(Sanm art? ?nand Ron quist 2004 ;Givni sh etal 2004 ; Sa?rkinen etal 2007 )and Afri ca to Australia (Swenson and Bremer 1997 )pri or to 60 mya, which were arbitrar ily set at 10 From 60 mya to the pres ent, the sam esix connec tions were allow ed, but all wi th an arbitr ary valu eof05 The other mode listerme d the vica riant mode land makes connec tions that woul dhave occur red as aresu ltof either vicariance ,ordispers als that coul d have occurred over short distanc es (all land based) Connect ions in this mode l were more struct ured over time and were based on mode ls of continent distributio ns during these time periods (Cox and Moore 2000 )Fo rthe time between 140 and 11 0mya the followi ng arbitrar yvalues wer eused: Neo tropics to Africa 10, Neotr opics to Austr alia (via Antar ctica) 075 For 110?90 mya the connecti on between the Neotropic sto Austr alia was reduced to 06 and from 90 to 40mya the connec tion betwee nthe Neotr opics and Africa was set at 075, betwee nthe Neo tropics and Aus tralia at 05 and a new connectio nbetwee nAfrica and Asia was added at06 From 40 mya to the pres ent, itwas assumed that all con nection swoul dbethe result of disp ersal and were not designat ed Two different runs were comple ted usin gdif ferent arbitrar yvalu es for vagility (001 and 01) bo th with asing learbitrar yvalu efor longe vity (001) for each model As with DIVA, we als oexamin ed the effects while con straining all Neo tropical Piper as monophyl etic Piperace ae have apoor fossil reco rd with afew fossils dating to the Plei stocene (Horn etal 2003 )Howeve r,since nearly all recent molecu lar phylo genetic anal yses have indicated that Piper ales are arelative ly early dive rgent lineage (Qiu et al 1999 ;Soltis et al 2000 ;Doyle and Endres s2000 ;Zanis et al 2002 ;Jaramil lo and Kramer 2007 ),itis extrem ely unlikel ythat the family has only a recent history and itis more likely that the small ?owers and their ?eshy fruits have not been well pres erved The recent discover yofafossil Saurur aceae is encour aging in that Piper aceae foss ils may yet be disco vered (Sm ith and Stock ey 2007 )How ever, the recent age (middl eEocene) and placem ent phylog enetical ly as sister to Saurur us (Smith and Stockey 2007 ),preclude the use of this fossil to calibrat e our tree Appomatox ia has been sugges ted to resemb le the extan tspecies Zippe lia (Pipera ceae) on the basis of shar ed spiny fruits (Friis et al 19 95)However, in the origina ldescrip tion of this fossil Friis et al (1995 )did not think the similarity suf?cien ttoplace itwithin Pipe rales and recent anal yses that have include dAppom atoxia in phylo genetic analyses indicate that amor eparsim onious explanat ion maintai ns its status as a relative of Chlo ranthac eae, although placem ent within Piper ales is only one step longer (Doy le and En dress 2007 )Th erefore, we resolve dtolook outside of Piper aceae and examin eclos ely related families to determin eifany of these had abetter fossil reco rd Lacto ris hasagood foss ilreco rd placing the genus in the late Turonian at912 mya (Zava da and Benson 1987 ;MacPh ail et al 1999 ;Maga llo?n and Sander son 2001 ) Since recent molecu lar data have placed Lacto ris within or clos etoAristolochi aceae (Nickrent et al 2002 ; Borsch et al 2005 ;Neinhuis et al 2005 ;Wanke et al 2007a )and Aristolochi aceae are sister to Saururace ae/ Piperace ae (Nick rent et al 2002 ;Borsch et al 2005 ; Neinhui setal 2005 ;Wanke et al 2007a )Piperace ae/ Saurur aceae woul dhave dive rged from Aristolochi aceae by that time ifnot earlier Th is gives us aminimum stem age for Piper aceae/ Saururace ae, but we are uncertain about the crown age Howeve r,to avoid adding arbitrar ybias, we used the age of 912 mya to cal ibrate our tree We recog nize that this date is potential ly erroneous ,but provides a starting point to begin the inve stigatio nofbiogeogra phy and evol ution wi thin Piperace ae and to establish appro xi mate date sfor the evol ution within this fam ily Penaliz ed likel ihood (PL; San derson 2002 )was used to estimat ethe ages of nodes for Piperace ae Cr oss validati on was conduc ted to ob tain the opti mal smo othing value Increm ents were set at 05 with 14 steps To examin ethe impact of smo othing values on our data, we ran sever al separat eanal yses with smo othing valu es ranging from 2to 100 As with DI VA, we opte d to use afully reso lved topology and we used the same tree used with DI VA Result s Succes sful amp li?cation s were obta ined for all DNA regions for nearly all indivi duals sam pled Glycer aldehyde 3phos phate dehydr ogenas ewas not succe ssfully amp li?ed from any of the accessi ons of Pepero mia tetraph ylla include dinthe analysis, Pe caper ata ,orSau rurus cernu us Phylog enetic analyses wer econduc ted with missing data for thes eindividual s, as well as excludin gthem from the analysi s Exclu sion did not change the topology of the trees, theref ore the results present ed here are based on analyses that include dall accessions Table 2sum marizes inform ation on each of thes eregions as well as the results of Model test 327 (Posa da and Cranda ll1998 )The model selected for the com bined, four gene data set was TVM + G with G = 08658, usin gLRT and AI C The PHT indi cated sign i?cant difference sbetwee nthe two paritions (p=001) However, this is likel ydue to subopt imal trees being found in som eofthe data part itions where low sign al did not resu lt in strongly suppor ted 16 JFSmith etal 123 topologi es When examined separat ely, each partiti on did not exhi bit any area of inco ngruence that had BS [ 75 Since hard incongrue nce was not detect ed (Seelanen et al 1997 ),we com bined data Maxim um pars imony analysis with inde ls treat ed as missing data resu lted intwo equally pars imonious trees each with 8,381 steps, CI = 053, RI = 080 The strict consensus ofthese tre esispres ented inFig 1Rescor ing indels resulted in eight most parsimoni ous trees, each of 8,682 steps, CI = 053, RI = 049 The topologies of the trees were nearly iden tical, but differ ed among the relat ionships within some of the Asian clades and some Neo tropical speci es Fig 1aMajority rule tree based on BI using adifferent model for each ofthe two partitions (cpDNA and g3pd )with AIC showing the part of the tree that includes only Piper Dashed lines represent relationships inthe strict consensus of two MP trees of 8,381 steps each (CI =053, RI =080) with indels scored as missing data Numbers abov ebranches are PP values before the slash and BS values [50% after Asterisks indicate clades that have PP \50%, but are in agreement with the MP trees Constraining all Neotropical specie sof Piper toamonophyletic clade cannot be rejected with SH tests The slashes after Pi augustum and Pi avellanum re?ect the switch ofthe position ofthese two species inthe MP analysis Speci esnames are colored according totheir native distribution: red Africa, Blue Paci?c Island, Purple Asia, Orange Australia/New Caledonia, Green Neo tropics Pi, Piper ;Ch, Chiapas; Ve, Veracruz bMajority rule tree based on BI using adifferent model for each of the two partitions (cpDNA and g3pd )with AIC showing the part of the tree that includes only Peperomia Numbers above branches are PP values before the slash and BS values [50% after Asterisks indicate clades that have PP \50%, but are inagreement with the MP trees Species names are colored according totheir native distribution: red Africa, Blue Paci?c Island, Purple Asia, Green Neotropics Pe, Peperomia Biogeography ofPiper 17 123 Regard less of the scor ing, MP resu lted inboth Pepero mia and Piper as monophyl etic Piper capens eissister toaclade of otherw iseexclusivelyPaci?cIsland speciesPiperguine ense issister to Piborbonense,aspeciesfrom theislandof Re unionandsister tothesetwospeciesisPicaninum,an Au stralian/IndonesianspeciesThiscladelies within wh at wo uldotherwisebe acladeof exclusivelyAsianspecieswith theexceptionof asecondAustralianspecies,Pirothianum wh ichisinacladeofMalaysian species,andPiaustroca ledonicumfrom Ne wCaledoniaTheNe otropicalspeciesare nearly mo nophyleticexcept thatPisanctum,aspeciesfrom Me xico,issistertoallrema iningspeciesof Piper The BI anal yses resulted inessenti ally the same topology regardl ess of the model used to inf er the tree The major ity rule conse nsus of the trees from the analysi susin gacom plex model based on AIC ispresent ed inFig 1Minor differ ences among anal yses usin gLRT or asing le model for all data produc ed asimilar overall topolo gy but differed slight ly in posteri or proba bility (PP) values, none of which was greater than 10 point sdif ferent betwee nanalyses The BI and MP trees are largely in agre ement with each other Both Pepero mia and Piper are monophyl etic Rela tionshi ps amon gPepero mia species differ slight ly in that the reso lution among Pe ret usa , Pe cavalie ri and Pe sandw icensis inBI isunres olved, as are the relations hips among Pe fernand opoiana /Pe kameruna na,Pe pellucida , Pe urocar pa/Pe serpen s,and Pe hernan diifolia and Pe retusa to Pe blanda/Pe leptost achya (Fig 1) Within Piper it is again clear that the native African species are notmonophyl etic and their sister group rela tionshi ps are iden tical betwee nthe MP and BI analyses (Fig 1)The greatest numb erofdifference sbetwee nMPand BI topologi es are the result of grea ter resolut ion within the Old Worl dclade (Fig 1)Relati onships among speci esinthe Neotr opical clade are nearly identical between MP and BI trees, dif fering slight lyinthe placem ent of Pi conce pcio nis , Pi colo nense ,Pi pseudo fuligineu m,Pi hostman nianum , Pi au gustum ,Pi avellan um ,and Pi nudifol ium How ever, the grea test dif ference between the tw otopol ogies isin the placem ent of Pi sanctum Th is species is sister to all remain ing species of Piper inthe MP anal ysis, but issis ter to all Old Worl d species of Piper with BI Likew ise, the Pi aequal e/Pi urophyll um clade whi ch is sister to all remain ing Neo tropical Piper species in MP (exc luding Pi sanctum )issister tothe Pi sanct um /Old World cla de with BI (Fig 1) Both KH and SH tes tsindicated that constra ining Afri can Piper to a mono phyletic clade (with placem ent correspo nding either to the positi on of Pi capens e or Pi guineense of Fig 1)was signi?cantly differ ent (p\ 0001) Likewi se constra ined mono phyly for Asian, Neotr opical, Paci? c, and African Pepero mia each gener ated signi?cantl y longer trees (p\ 0001 ) Onl y two compari sons produced results that were not signi?cant These were the constra int of all Neo tropical species of Piper (p= 0168 and 016 1for KH and SH, respective ly) and amono phyletic Pe retusa (p= 0153 and 0098 for KH and SH, respective ly) Irrespe ctive of the numb er of regions or the tree that was used, DIVA ultimatel yindicated that the ancestor to either Piper or Pepero mia was widesp read (pre sent in all desig nated regions) and required 18 ?19 (all include d Piperace ae), 9?10 (Piper on ly), or 11?12 (Pepero mia only) dispers al events to expl ain the exta nt sampled distributio n Constr aining all Neo tropic al species to asing le clad edid not affect the resu lts RegardlessofwhetherPiper,Peperomia,or allofPiper aceaewere set atadate of 912 mya, the results of r8s were similar Cross validati on gave an op timal smoothing value of 16 which was used for the date spresented here Varyi ng the smo othing value did no tsubstant ially change the dates obtaine dfor nodes on the tree and range dwithin ?ve to ten mya from the values obtaine dwith the value obtained from cross validat ion Fixi ng the age of Piper aceae at 91 2 mya, the crown age of Piper was 7175 mya and Pepero mia was 8889 mya We reiterate that these dates are likel y to be minimum ssince the crown age of Piperace ae islikely to be older than 912 mya, but by an unknow n number of year s Fig 2aUltrametric tree using the topology of the MP analysis of Fig 1aderived from PL placing the age of Piperaceae at912 mya Species names are colored according to their distribution and branches are colored according to consensus of models recreating distribution patterns: red Africa, Blue Paci?c Island, Purple Asia, Orange Australia/New Caledonia, Green Neotropics Where models disagree, colored squares above the lines and before the slash follow distribution patterns determined by AReA using the dispersal model with avagility of 001, colored squares after the slash are based on the dispersal model with avagility of 01 or the vicariance model (with two exceptions) with vagility of01 or001 (see text for amore detailed description of the models used), triangles re?ect the two exceptions determined by the vicariance model that differed from the dispersal model, circles below the branc hes correspond to DIVA Constraining all Neotropical species ofPiper toamonophyletic clade cannot be rejected with SH tests and does not alter the results of AReA as they are presented here Pi, Piper ;Ch, Chiapas; Ve, Veracruz bUltrametric tree using the topology ofthe MP analysis of Fig 1bderived from PL placing the age of Piperaceae at912 mya Species names are colored according to their distribution and branches are colored according to consensus of models recreating distribution patterns: red Africa, Blue Paci?c Island, Purple Asia, Green Neotropics Where models disagree, colored squares above the lines and before the slash follow distribution patterns determined by AReA using the dispersal model with avagility of 001, colored squares after the slash are based on the dispersal mode lwith avagility of01 orthe vicariance model (with two exceptions) with vagility of 01 or 001 (see text for amore detailed description of the models used), triangles re?ect the two exceptions determined by the vicariance model that differed from the dispersal model, circles below the branches correspond toDIVA Pe, Peperomia c 18 JFSmith etal 123 Biogeography ofPiper 19 123 Acom parison of the resu lts of AReA using two differ ent values of vagi lity indicat ed that the higher likelihoo d was obta ined with the vicariant mode land vagility at 001 (ln = 5958) although this did not dif fer greatly from the vicariant model with vagility at01 (ln = 6282) or either of the disp ersal models (ln = 6387, vagi lity = 001, ln = 6519, vagi lity = 01) Th emajor ity of distributio n pattern swithin the tree wer ethe same, however ,some minor difference swere appar ent and these are also pre sented on Fig 2The major ity of the difference swere within Peperomi awith the higher vagility rate (01) indi cating aNeo tropical origin for the genus with subse quent, multipl edispers al events to Af rica, Asia, and the Pa ci?c Islands ,whe reas the lower vagi lity value (001) implied a more widesp read ancestor with aNeo tropical and African origin, eventual extincti on in Africa with subsequent dis persal back to Af rica and multipl edisp ersal events to Asia and the Paci? c Islands (Fig 2) Th e vicariant mode ls resulted in the same distrib ution regardl ess of the vagility value select ed and were nearly in full agre ement with the dispers al mode lwith vagi lity at 01 In cont rast to Pe per omia ,the bioge ograp hy for Piper was relatively consist ent regardl ess of model selected with aNeo tropic al origin for the genus followed by either (1) disp ersal to the Pa ci?c Islands and further disp ersals to Asia, Aus tralia and Africa (disper sal model vagi lity = 001), (2) adispers al to Asia followed by dispers als to Afri ca and Austr alia and asep arate disp ersal to Africa/P aci?c Islands (dispersal model vagility at 01), or (3) dispers al to Af rica with subseq uent dispers al to Asia, Paci ?c Islands ,Aus tralia, and back migra tion to Af rica (vicarian tmode l)(Fig 2)Con straining all Neo tropic al speci es of Piper to asing le clade did not affect the resu lts Discus sion Altho ugh we did not include represe ntatives of Zippe lia, Verhue llia ,orMane kia in our anal yses, we were able to demon strate that the two largest genera of Piperace ae were each monophyl etic and sister to each other in agreem ent with the results of all previous phylo genetic anal yses of the family (Jaramill oand Manos 2001 ;Jara millo et al 2004 ; Wanke etal 2007a ,b;Jaramil lo etal in press ) Scoring of indels and different methods of data anal ysis provided some different resu lts within Piperace ae, but in gener al these were only in areas of weak support Regard less of how the data were anal yzed itisclear that Pepero mia and Piper have different biogeogra phic patter ns (Fig 2)Th is was not fully asurpr ise sinc eprevious phylo genetic anal yses of Piper had indicated cle ar mono phyletic groups that correspo nded to major bioge ographic regions (Jaramill o and Manos 2001 ;Jaramillo et al, in pres s),each of which were also found here (Figs 1,2)Pepero mia in contrast, has several species that are known to be widesp read, theref ore such apatter nwas not expec ted Howeve r, the mono phyly of some of these widesp read species had not yet been tes ted in aphylo genetic analysi s An unusual result of this analysi sisthe placem ent of three Neotr opical speci es: Piper sanctu m, Pi aequale and Pi urophy llum Bayesian anal yses placed thes ethree spe cies as sister to the Old Worl dclades of Piper and MP analyses rooted the tree with Pi sanct um sister to all remain ing species of Piper (Fig 1) Although thes eare interest ing resu lts, itshould be note dthat KH and SH tes ts could not reject amono phyletic cla de of all Neo tropic al species of Piper that include dPi sanct um, Pi aequal e,and Pi urophy llum Theplacem ent of these three species may be the resu ltof limited sampli ng Inclusion of additio nal, and cri tical speci es may result in the placem ent of these species within, or at least sister to an exclusi vely Neotr opical cla de of Piper A stud ythat includes over 300 species and over 600 acce ssions of Piper provides evidence that Pi sanctum is sister to Pi cine reum C DC and this clade is sister to the remainder of Neotr opical species excludi ng species in section Schi lleria ,few of which were include dinthe present anal ysis (Jaramill oetal in press ) The exclusi on of Pi cinereum and species of sectio n Schilleri ainclude dinJaramil lo etal (in press )islikely the best expl anation for the unusua lposition of Pi sanct um in the present stud y How ever, since the impact of con straining all Neo tropical species of Piper into asingle clade did not affect the overall results of the bioge ograp hic analyses ,we will discuss our resu lts using the trees obtaine ddirectly from the analyses of our data The biogeogra phic origin of Piper Speciat ion within Piper has cle arly followed adisti nct biogeo graphic patter nwith radiat ions occurring in three main biographi cregions as previo usly outlin ed by Jaramillo and Manos (2001 )and Jara millo et al (in pres s) These regions are Asia, the Paci?c, and Neotropic swith support for the sister group relat ionship of the Paci? cand Asian clades Th eresults from AReA imply that Piper has an origin in the Neotropic swith adispers alto either the Pa ci?c Islands ,orthe Pa ci?c Isl ands/Africa and Asia usin gour dispers albased mode l(Fig 2)An alternative perspectiv eis present ed with connec tions based on vica riance and land connec tions betwee nthe Neo tropics and Africa Unde rthis model, Piper dispersed ?rst from the Neo tropics to Africa and then to the Paci ?c and Asia with subse quent dispers als from Asia to Australia and back to Africa (Fig 2)DIVA analysi sdoes not suppor tthese mode ls and implies a widesp read ancestor of Piper followed by later extinct ions and dispersals (Fig 2)Howeve r,DIVA does not allow for 20 JFSmith etal 123 any prior know ledge of histories of area stobeconsi dered independe ntly of the histor yoflineage s(Ronqui st 1996 , 1997 ;Ree etal 2005 )In cont rast, AreA mak es use of prior assumpt ions assigned by the user related to bioge ograp hic connec tions and the ages ofclades to determin eadistri bution patter n The dispers al mode lused in AReA allowed for aconnection betwee nthe Neotrop ics and the Pa ci?c Islands that was grea ter than that between the Neotr opics and Asia but only prior to 60 mya Howeve r, this time period pre dates the grea ter dive rsi?catio nthat occurred within Piper and Pep eromia (Fig 2)During the time per iod when lineage diversi? cation was occur ring, the dispers al model used in AReA allowed for all connec tions to be equal The disp ersal mode lopti mized an initial dis persal event to the Paci?c Isl ands whe nvagi lity is low (001) with subsequent disp ersals from the Pa ci?c Islands to Africa and Asia and further disp ersal events from Asia to Africa, New Caled onia, and Aus tralia (Fig 2) With an increas einvagility (01), the dispersal model has aslightly different optimiza tion and allows for adirect connec tion betwee nthe Neotropic sand Asia, as well as adisp ersal to both Af rica and the Pa ci?c Islands Subsequ ent dispers als from Asia are near ly equival ent for either valu es of vagility under the dispersal mode l,the excep tion bein ganAsian ancestor for the Pi guineense /Pi borbone nse/Pi caninum clade with low vagility fol lowed by dispers al to Africa for the ances tor to Pi borbonense and Pi guineense With a higher vagi lity valu e, this same clade is proposed to be African in origin (albeit from within an Asian clade) and dispers al of Pi cani num back to Aus tralia/As ia sinc ethis species occurs in both Austr alia and Indonesi a(Fig 2)The vicariant mode lproduc ed anear ly identical distributio n pattern to the dispersal mode lwith highe rleve lsof vagility with two main excep tions: (1) the ?rst clade outside of the Neotr opics in Piper was African in origin, and (2) the dis persal to the Paci?c was from Africa, not vice versa as implied with the dispers al mode l(Fig 2) The stem and crown ages for the nonN eotropical clade are 530 and 4186 mya ,respect ively These age range sjust barely ?t into the mode lwhe rethe connec tion betwee nAfrica and the Neotr opics was still highe r(075) than betwee nall areas being equal and is the most likely explanatio nfor these results This connec tion is intrigui ng and the simi larity betwee nhabi tand breeding system of species of Piper in the Paci? c, Neotro pical species and Pi capense argue for this closer connection How ever, the low diversity of native Piper speci es in Africa argue sagai nst this If the African origin ofall nonN eotropical cla des of Piper iscorr ect then either widesp read extincti on or areduction in the rate of speciat ion in Afri ca would be necessar y to explain the depaup erate Piper ?ora there In light of this, itseems more likely that dispers al to Af rica occurred inde penden tly from Asia and the Paci ?c, not from the Neotropic s Piper bioge ography: Asia/ Paci?c Itisyet to be fully reso lved ifthe sister group relationship betwee nthe Asian and Paci? ccla des wi llbe maint ained, or if further samp ling will reve al that on eisderived from within the other Studies with grea ter species sampling have maintai ned the sister group relations hip betwee nthese clades (Jaramill oetal in press )Paci? cIslands are gen erally volc anic in ori gin and have amore recent origin relative to the Asian mainl and As aresu lt, mos tstud ies that have examin ed plan tgroups with sign i?cant Pa ci?c and mainland distribut ions have show n that the Pa ci?c species are derived from mainland clades (Baldwin et al 1990 ,1991 ;Wright et al 2000 ,2001 ;Cronk et al 2005 ) From the data we have here ,itappea rs that the Aus tralian and New Ca ledonian species are also part of the Asian radiation Australia has only 13 species of Piper (Aust ralia Plant Names Index 2007 )and we were only able to ob tain mater ial from tw owhi ch appea rasinde pendent introdu ctions (Fig 2)We have only one of 16 species of Piperace ae from New Caled onia (Morat 1993 ),which also appea rs to be aseparat eintrodu ction event from the Aus tralian introdu ctions and isclearly not apart of the Paci?c clade Howeve r,thes econclu sions should be rega rded as prelimi nary until we have better sampling from these regions Piper bioge ography: Af rica Itisclear that the native Af rican species are not the result of asing le introdu ction Piper capens eisapart of the Paci?c clade and repr esents an introdu ction into Africa (but see discussi on above regarding the resu lts of the vicari ant model) inde penden tofPi guineense (Fig s1?2)Th elatter species issister to Pi borbone nse ,aspecies from Reunio n The proxi mity of Reunion tothe East African coast sugges ts that since the sister to these two species isPi cani num ,an Indone sian/Austra lian species (our sample presuma bly is from Aus tralia) ,that dispersal wen tfrom Asia to Australia to Reunion and Af rica How ever, the limi ted sampling of Asian species in ouranalysis preclude sthis conclu sion Further sam pling of Asian species will be esse ntial to resolve whether this pattern isretained in the fut ure The phylo genetic analysi sdoes reso lve an interesting questio nregardi ng the habit, breeding system, and dive r sity of Piper in Africa Piper capense isan erect shrub with hermaph roditic ?owers, whe reas Pi guine ense is adioe cious vine Since the major ity of Neo tropic al species are shrub swith herm aphroditic ?owe rs and Asian species are dioecio us vines, ithad been consider ed apossibi lity that the two African species were, resp ectively, from the Neo tropic sand Asia (Jaramill oand Marquis 2004 ) If both species were from only one region, then ashift of habit Biogeography ofPiper 21 123 and/or bree ding system occur red within Africa These consider ations, however ,ignored the Paci? cgroup which also are general ly shrub s(Smith 1975 )The data pres ented here clearly indicat eseparate introductions and itislikely that both species have retai ned characteri stics of their parenta lclade with regard to habit and breeding syst em Ithas als obeen surprising that Af rica ishome toonly two endemi cspecies of Piper wherea sthe Neotropic sare home to appro ximatel ytwothirds of the 1,000 species and Asia (inclu ding Aus tralia and East Indian Islands) has about 300 species (G rieg 2004 ;Jaramil lo and Mar quis 2004 ) The much sma ller land mass of the Paci ?c Islands is home to only about ten speci es all of which corr espond to sectio n Macro piper (Sm ith 1975 ;Jara millo et al in pres s)Expla nations for the paucity of Afri can speci es have invok ed extincti on that coincid ed with aridi?c ation of muc hofthe African cont inent and loss of suitable Piper habitat (Axel rod 1972 ;Raven and Axelrod 1974 ;Axelrod and Raven 1978 ),or recent introdu ctions In light of the phylogenet ic results, and the crown ages of Pi capens eand Pi guineense atappro ximatel y8mya ,itseems much more likely that the low species dive rsity of Piper in Africa is the result of a relative lyrecent arrival and the large species level radiation that isseen in othe rareas where Piper isfound has not had suf?c ient time to occur This scenari ohas been invoked for African Rapat eaceae and Bromeliac eae (G ivnish et al 2000 ,2004 )Givni sh etal (2004 )conclude that, despi tethe similar ity in habi tats between Maschal ocepha lus and its closest relat ives in So uth Amer ica and the fact that the two regions were likely adjacent prior to the openi ng of the Atlanti c,the presence of Rapat eaceae in Afri ca isthe result of arecent (73 mya; Givni sh et al 2000 ,2004 )long dis tance dispers al event Likewi se, the pres ence of asing le species of Pit cairnia (B romeliaceae )isalso propos ed to be the result of arelatively rece nt (12 mya )long distanc e dispers al event (Givnish et al 2004 ) Multip le introductions into Africa are rare with other tropic al groups that have agloba ldistribut ion Davis et al (2002 )found that the Afri can acridoc arpoid Mal pighiaceae had asing le origin from So uth Amer ica with asubse quent long distanc e dispersal into New Caledon ia Likewise Mo?ller and Cronk (2001 )found asingle origin for all sample dspecies of African Gesneriac eae and have specu lated that asing le origin from Asia is likely the best explanat ion for the family in Afri ca Con tiet al (2002 ) utilize dphylogenet ic relations hips to show that three small African famili es previous ly placed within Crypt eroniaceae (Olinia ceae, Penaea ceae, and Rhynchoc alycace ae) formed amono phyletic group In cont rast, de Groot et al (2006 ) have demonst rated up to ?ve inde penden tintrodu ctions into Africa for Aris tolochia Piper umbellat um : Species of Piper that are not in cultivat ion are general ly only found in a sing le biogeo graphic regi on (A sia, Africa, Paci?c Islands , Neotr opics) The one excep tion to thi sinour analy ses is Pi umbe llatum Th is stud yinclude sindividual sfrom the Neotr opics and Africa, but othe rstudies have sam pled indivi duals from Asia (Jaram illo and Manos 2001 )In all studies ,all indivi duals have formed amono phyletic group (Jaram illo and Manos 2001 ;Figs 1,2) The most com mon explanat ion for this widesp read distributio nhas been recent human introdu ctions Trans fer of plant material by human siseither intent ional becau se the plan thas utilit y, or accident al as acontam inant in othe rmateria ls The former likel y expl ains the distribut ion of Pi nigrum , Pi betle ,Pi methysticum ,and Pi auri tum which are economi cally important and widel yculti vated in tropic al regions thr oughout the wor ld Piper umbe llatum is a weedy species typical of distur bed habitats, and its curr ent distribut ion outs ide of the Neotr opics has been consi dered to be the result of accident al introductions Howeve r,this species also hasmedicinal uses in Came roon (J F Smith, person al observ ation) and loca lbota nists consider the species to benative (Chri sFouminya m, pers onal com munica tion) The data presented here cle arly show that the African individual sofPi umbe llatum are mono phyletic and nested within an otherw ise exclusivel y Neo tropic al clade Howeve r, the age of the African Pi umbe llatum clade is estimat ed at 1088 mya which would prec lude human transf er (Fig 2) Thus Pi umbellat um re?ects a shift in dispersal abili ty within Piper that is more simi lar to that of Pepero mia or an unusual disp ersal event for this species Placing dates oncla des relie sonsmo othing the substitut ion rate from one branch to the next without allowing for major shifts in rate change (Sande rson 1997 , 2002 ) A range of smoothing valu es never brough tthe date for the Pi umbellat um clade within a range that could be attributed to human actions How ever, the weedy habit of Pi umbe llatum may allow for agreater subst i tution rate as aresult ofsmall foundi ng populations and a shorter generation time As such, the greater dive rgence among the seque nces of Pi umbella tum include dinthis study may est imate the age of this clade to be older than it actual ly is Pepero mia biogeogra phy Some species of Peperomia are known to have species distribut ed thr oughout the tropics Howeve r, the mono phyly of these widel ydistributed species had notyet been tested with rig orous molecu lar phylo genetic methods (Wank eetal 2006 )As itturns out, at least one species examin ed here is not monophyl etic, Pe retusa Anindi vidual of thi sspecies from So uth Africa was sister to Pe cavalieri ifrom China Pepero mia retusa and Pe cav alierii are consi dered close, and possibl ymay represent 22 JFSmith etal 123 subspe cies rather than distinct species (C heng et al 1999 ) Another indivi dual of Pe retusa from So uth Af rica was sister to the Asian acce ssion of Pe blanda and Pa ci?c accessi on of Pe leptostach ya (Figs 1,2) However, KH and SH tes tscould not rej ect amono phyleti cPe retusa and the results found here may simpl yre?ect divergen ce of the seque nces sam pled amo ng different populations Alte rna tively, since aglobal revision of Pepero mia has not been conduc ted recently, itmay simply be that amore care ful and diligent examin ation of morp hological charact ers will reveal distinct difference samong thes epopulations that may mer it their separ ation as disti nct species Similar studies of Piper have found that some of the widesp read species may act ually represe nt anumber of distinct and different lineage sasmultiple popul ations from differ ent regions have not resu lted in mono phyletic species in sev eral instances (Jaramill oetal in pres s) In contrast to Pe retusa ,three additional species were include d with sample s from differ ent bioge ograp hic regions, (Pe bland a,Pe pellucid a,and Pe tetraph ylla ), each of whi ch prove dtobemonophyl etic with ob vious dispers als to other regions followi ng speciat ion (Fig 2)As with Pi umbell atum ithad been speculated that some of these species may repr esent humanm ediated introdu ctions as they are gener ally found in disturbed habitats (Pe pel lucida )Howeve r,two of these species ,with sample sfrom either Africa and the Neotropic s(Pe pelluci da)orfrom Africa and Asia (Pe tetraph ylla )have intras peci?c dive r gence sthat are near ly as old as the major ity of clades leading to differ ent speci es within the genus (Fig 2)The occurren ce of several species with popul ations on differ ent contine nts implies that Pepero mia species are capab le of long distance disp ersal The fact that the popul ations on the different cont inents are distinct at the molecular level and that the distinct ion can be dated to 4941?53 75 mya for at least two of the species indicates that stro ng and simi lar selection pres sures are maint aining morphol ogical simi lariti es of popul ations that are widely disjunc tand have been disjunc tfor long periods Alte rnativ ely, itispossi ble that the ages for Pepero mia are distorted bythe shorter generat ion time and pres umably smaller founde rpopula tions of Pepero mia in com parison to Piper which may affect rate heteroge neity Likewi se, itmay be possibl ethat these species have dispers ed mul tiple times to differ ent biogeo graphic regions and our sampling here by chanc e re?ects som eofthe olde rdispers al events Gr eater sam pling may demonst rate more recent dispersal events and it ispossi ble that gene ?ow among the newer dispersal events and establish ed popul ations is maint aining mor phologic al unity across cont inents Coale scence of all genes ,includi ng the ones sampled here for phylogenet ic anal yses, simpl y has either yet to occur, or has coalesc ed to the copies from the earlier immigrant s The sampling ofPepero mia in this study was low, but suf?c ient to show that major biogeo graphic patterns within the genus are not as cle arcut as they are in Piper Mul tiple introdu ctions into Africa, Asia, and the Paci ?c Islands are necessar yfor either mode lused in AReA for Pepero mia with an ancestor either in the Neo tropics, or in the Neo tropic sand Africa (Fig 2)DIVA indicat es awidesp read ancestor for Pepero mia with subseq uent extincti ons and re introdu ctions to explain the current distributio nofspecies (Fig 2) As wi th AReA, DI VA requires multipl eint ro ductions into all geogr aphic regions excep tthe Neotr opics Further sampling of Pe peromia species may yet reve al some patterns of bioge ography It is likel ythat the sma ller fruit sofPepero mia allow for agreater abili ty to disp erse than the larger ,heavier fruits of Piper All of the kno wn fruit disp ersal in Piper is carried out by bats primari ly (Fleming 1988 ,2004 ), and birds secon darily (Gorch ov et al 1993 ,1995 ) In both cases, the fruit sare ingested and, thus ,are only carrie dover arelat ively short distanc ebefore they are expel led Fu r therm ore, the fruits of mos tspecies of Pe peromia are sticky and readily stick to the feet and feathers of birds (Ridley 1930 )In fact, one of the longest docum ented instanc es of seed dispers al by birds isin Pepero mia (V aldebenito et al 1990 )Not only does itappea rthat birds carrie dPepero mia seeds from sout hern South Amer ica to the Paci ?c Juan Ferna ndez Islands ,600 km from the coast of Chile, but then from the Juan Ferna ndez Islands to Tris tan de Cunha Island in the South Atla ntic; avoyage of over 5,000 km (Vald ebenito et al 1990 ) Numerous bir ds are know nto?yamo ng these island groups (Vald ebenito et al 1992 ) An addi tional explanat ion is that the epiphytic habit of Pepero mia allow sfor grea ter likelihoo dofestabl ishment on different cont inents than would occur ifthe species were directly root ed in the soil Itmay be that dispers al ofPiper fruits is as likel yasthos eofPepero mia and that compe tition restrict smost species from becoming establish ed If compet ition among epip hytes wer elower than that of ter restrial species ,itmay be possible that Pepero mia could dispers eand establish at agreater rate than Piper and thus the sam e species can be found on several contine nts Lastly, itmay be that the widesp read species of Pepero mia are mor etolerant, or even require ,more distur bed habitats than species of Piper This is de?nitely true of Pe pellu cida which general ly isfound in weedy, disturbed habitats in many tropical parts of the wor ld (S Wanke ,pers onal commun ication) Diver gence times Ideally we would have studied divergen ce times within Piperace ae usin gthe phylogenet ic history of the family in Biogeography ofPiper 23 123 conjun ction with multipl efossil calibrat ion points Unf or tunately, Piperace ae have apoor foss ilreco rd with only a handf ul of recen tfossils attributed to the family (Horn etal 2003 )and its sister family, Sau ruraceae (Smith and Stoc key 2007 )Ra ther than avoid estimat ing dive rgence times we opted to use the fossil record of Lacto ris (Zava da and Benson 1987 ;MacPhail et al 1999 )asastarting point Recent phylo genetic analyses have placed La ctoris within Piperal es, and more speci ?cally, part of Ar istolochiaea e, the sister to Piperace ae/Sauru raceae (Nick rent et al 2002 ; Borsch et al 2003 ;Jara millo et al 2004 ;Jaramil lo and Kram er 2007 ;Wanke etal 2007a ,b)Thus the pres ence of Lacto ris in the Turonian implies that at the minimum, its sister clade, Piper aceae/Sau ruraceae, mus t also have diverged at this point Itis possible that our dates under estimat ethe divergen ce points within Piperaceae sinc eit may be that this clade diverged well befor ethe origin of Lacto ris :however ,sinc ewedonot know whe nSaurura ceae and Piperace ae dive rged, itmay be that ourestimates are grea ter than what actu ally occur red Until other fossils are discover ed that can con?rm the divergen ce points that are estimat ed here ,these dates provi de amiddle ground on which we can start to form hypothese sfor fut ure testing It may be assumed that since Piper aceae are amo ng some of the olde st lineages of ?owe ring plants (Qiu et al 1999 ;Soltis et al 2000 ;Doyle and En dress 2000 ;Zanis etal 2002 ;Jaramil lo and Kram er 2007 )that dispersals and radiation swithin the famil ywould also be anci ent Our data indicate that with the stem ages of Piper and Pe per omia de?ned at 912 mya ,the crown ages for the two genera are still both in the Cr etaceous at 7175 and 8889 mya ,resp ectively (Fig 2) Despite the older crown ages for both Piper and Pep eromia ,most extant clades within these genera do not diverge until much later with major diversi? cation of Piper into sever alNeo tropic alcla des, aPaci? c/Pi capens e,and a predo minantly Asian cla de at 3733 mya (Fig 2) Like wise, although the crown age of Pepero mia is plac ed at 8889 mya ,the majority of diversi? cation within the genus occurred between 5709 and 3133 mya (Fig 2)Both of these resu lts may be an artefact of limit ed species sam pling, but broader phylogenet ic analyses indicate that Pi sanctum isone of the earliest dive rgent lineages within Piper and that othe rearly divergen tcla des with the excep tion of sectio nSchilleria are represente dhere as well (Jaram illo et al in pres s) Likewi se, sam pling within Pepero mia represents som eofthe earliest divergen tclades (Wank eetal 2006 )Therefor e, additional sampling may reveal branc hes of earli er dates, but itwould be unlikely to discover many branches divergin gpri or to the ones dis covered here Recent radiat ions and dispers als are not in con?ict with biogeo graphic and ph ylogenetic stud ies ofother groups that have been able to estimate dates on radiat ions Won and Renner (2006 )found that although the fossil record indicates adiverse and widesp read Cret aceous Gne tum, the current species ?distribu tions are the result of dive rgence that occur red in the Upper Oligo cene, Mio cene, and Plio cene Similar case sofrelat ively rece nt radiat ions in anci ent lineage shave been found for Ephedra (Huang and Price 2003 :Icke rtBond and Wojcie chowski 2004 )and Equise tum (Des Mar ais etal 2003 ),as well as several angiosperm taxa with ancient and widel ydistribut ed foss ilrecords such as Chlor anthacea e(Zhang and Renner 2003 ;Eklund et al 2004 ),Calyc anthacea e(Zhou et al 2006 ),Illici um (Morris et al 2007 ), Anisoph ylleac eae (Zhang et al 2007 ), and Nympha eace ae (Yoo et al 2005 ) A pattern of an ancient stem lineage with amore recen t crown age is not restricted to taxa with origins in the Cretaceo us or earlier Renner etal (2001 )proposed astem age for Melastom atacea eand Mem ecylace ae at 53mya based on foss ildata This age marked atrichotomy in the tree among Memecy laceae ,tribe Kibessie ae of Mel as tomatace ae and the remai nder of Melastom atacea e Using fossils to establ ish tw ocalibr ation points acro ss the tree and assumin gaclock, the crown ages for the three groups are 133? 145, 92?100, and 45?49 mya, resp ectively Altho ugh the crown age for the major ity of the Mel as tomatace ae seems relat ively recent (4?8 mya after the stem origin) ,itshould be emphasized that thi sisthe point where major lineage swithin the fam ily dive rge and that species radiation s within these lineage s occur more recently (10?15 mya) Penningto netal (2004 )also found apre Pleisto cene crown age (11?20 )for adive rsity of taxo nomica lly unrelat ed groups that occur in seasonal ly dry tropic al forest s(Cou rsetia ,Poiss onia ,Ruprec htia ,Loxop terygi um and Cha etocalyx/ Nisso lia ) The occur rence of old stem ages and recent crown ages implies long periods of stasis, high leve ls of extincti on among earli er dive rgent lineage s(cr eating ??naked ??stems), or acom bination of both Lack of fossil sfor Piper aceae preven tsus from dete rmining which of these hypothese sis more likel yfor thi sfamily Either hypoth esis is possi ble and data from differ ent taxo nomic groups suppor teither hypot hesis Some groups have extensive foss ilrecords and demon strate what appea rs to be early extinct species arguing for extincti ons that have produc ed long stem lin eages (Melastom ataceae: Renner etal 2001 ,Juglandac eae: Manc hester 1989 ,1991 ;Manc hester and Wheeler 1993 ) In contrast,theam phiAtlanticspeciesSymp honiaglobu lifera (Clusiaceae) can be dated using ITS sequence divergence to 2852mya,andthesplitbetweenAfricanand Ne otropicalcladesisplacedat1736mya(Dick etal2003) Fossilsofthisspeciesfrom Me soam ericaandSouthAme rica canbedated to theMioceneThedating of theAfrican/ Am ericansplit,theretention of similarmorphologybetween 24 JFSmith etal 123 populationsonthetwocontinentsandthefossilsfoundinthe Am ericasim plythattherehasbeenatleasta15?17my period of mo rphologicalstasis(Dick etal2003) Long periods of stasis ,fol lowed by dive rsi?catio nmay also be seen in mor erecent exam ples of migration such as invasiv especies (Mack et al 2000 ), albe it over amuch shorter time span One curr ent mode lfor inva sive species propos es that aspecies may becom eestabl ished in anew area, but remai ninaquie scent stage for long periods before becomi ng inva sive (Mack etal 2000 )The invasiv eness may either be the resu lt ofrecombi nation amo ng genotypes introdu ced into the new area that were not sympat ric intheir native range (Ellst rand and Schie renbeck 2000 ;Petit et al 2004 ;Willia ms etal 2005 ),or evol ution of new alleles that allow the species to move beyond its origina lintrodu ction Likewi se we may envision the establ ishmen tofasing le species in anew area that takes along perio dbefore a mutation or reco mbinatio nproduces agenotype that allows for range expans ion and subse quent speciatio n This may explain the older stem and more recent crown ages seen for Piperace ae and othe rplant fam ilies A possi ble explanatio nfor along period of stasis in at least neotropi cal Piper is alimit ed amount of suitable habitat for the species to diversi fy into In the Andes of South Amer ica nume rous groups such as Anthur ium , Gesner iaceae ,Miconia ,Campanula ceae, and Piper have high levels of dive rsity betwee n 1,500 and 2,500 m in elevat ion Prio rto10?15 mya, itisunlikel ythat the Andes had an extensi ve belt of elevat ions at this height (Lamb 2004 )and thus, the clou dfore stthat contain shigh diversity in Piper was limit ed in its extent, and thus limited the diversi ty within the genus More recent uplift screate da larger range of suitable habi tats for Piper and correspo nd to the period in time whe re ourphy logenetic and molecu lar dating indicat es the greatest level of dive rsi?catio n occurred in the Neotropic s(Fig 2) Quijan oAbril et al (2006 )have reasoned that Piper is not an Ande an genus and that area sofdiversi ty and endemi sm are cent ered in lower elevat ions with signi?cant areas of diversity in the Caribbean and Amazo nian regions Howeve r, these areas of diversity and endemi sm do not preclude the Andean oroge ny as amechanism for incr eased diversi ?cation within Piper The uplift of the Andes Mount ains in the Mio cenePli ocene also had an enormous effect on speciatio ninman yplan tgroups (Gent ry 1982 , 1989 ;Richa rdson et al 2001 ;Hughes and Eastwoo d2006 ; Ritz et al 2007 )The initial rise of the Ande sasearly as the Eocene produced localize dhabitat change, and isolated the lowland forest swest of the oroge ny from the much larger Am azonian forest (Grego ryWodzi cki 2000 ;Youn g et al 2002 )Th eoroge ny als oproduc ed ahighly dissected and constant ly changi ng landscap e, and generat ed new micro climat ic nich es through valleys of varying slop eand orient ation, newly exposed subst rates, rain shadows, and regular openi ngs of new area sdue to frequent landslides (Gent ry 1982 )Thus, even ifPiper does not ?t the model of an Ande an genus, itisstill likely that the emergence of the Andes had adirect in?uence on the speci ation in this group Paleocl imate The cooc currence of species radiation s among many taxono mically unrelat ed groups and lineage swith differ ent times of origin implies that there is som ething charact er istic of the Tertiary that allowed for extensi ve species radiation s Episo dic glacia tions which result in both a cooler and dri er climat ein tro pical regions have been propos ed as an explanatio n for high levels of tropic al diversi ty as areas of tropic al wet and dry forest swould contrac tand expand with each glacia tion, resp ectively Species surviving in refugia during each glac iation/int er glacia tion woul dresu ltin anew wav eofspeci es radiation during each cycle (Prance 1982 ;Whi tmore and Pranc e 1987 ;Behren smeyer et al 1992 ;Richa rdson et al 2001 ; Dutech et al 2003 )Howeve r,the refug ehypot hesis was based on Qua ternary changes in the Pleist ocene and dates for mos tplant groups plac ecrown ages before the Pleis tocene, at lea stin South Amer ica (some Central Amer ican plant groups have Pleistocen e age crown groups ;Pen nington et al 2004 ) Early Eocene global temper atures reac hed amaximum allowing tropic alforest stoextend into high latitudes Many plant families that are know ntoberestri cted to tropic al regions today were found inEocene and Oligo cene Laura sia (Manches ter 1999 ;Mai 1995 ,Kubi tzki and Krutz sch 1996 ) Diver sity within the Neotrop ics reac hed amax imum atthis point in time correspo nding to either the shif tinclim ate, or the amo unt of land encom passed by asuitabl eclim ate explain ing the increas eindiversity (Jaram illo et al 2006 ) Paleocl imatic anal yses indicate that dispers al to differ ent regions was gener ally not limit ing with regard to man y factor ssuch as light, temperatu re, moisture (Tiffney and Manc hester 2001 ),or physical cont inuity such as the closing of the Tethys seaw ay (Hall 1998 )and connec tions acro ssthe North Atlanti c(Tiffney 19 85a ,b;Dav iset al 2002 )atthis time Therefor e, itis possibl eboth Piper and Pepero mia diverged in the Cretaceo us, but did not radi ate and disp erse until the Eocene temperat ure max imum Provi ded that the current physi ologica lrequire ment of atropic al climat ehas been unif orm inPiper aceae, subsequent cool ing and epis odic drying inthe late Eocene toMio cene may have disrupt ed the further dispersal of these plants Thus, barrie rs to dispers al and gene ?ow for the lar ger plan tsand fruits of Piper may have isolat ed thes etotheir major bioge ographic region sand only radiation swithi nthe tropical zones of the Amer icas, Biogeography ofPiper 25 123 Paci? cIslands and Asia have occur red sinc e,with dispers al to Af rica and Aus tralia occurri ng more recently Altho ugh som especies of Piper and Pepero mia are found in open area s, often at the edges of forest, the major ity of species in Piperace ae occur in rainfore st under stories (Gri eg 2004 ) Applying physiolo gical uni formita rionis m tothe speci es represe nting the ances tor of these genera, itwould have been esse ntial for tropic al rain forest stoals ohave existe dattheir time of origin in the late Cret aceous Although much fossil evid ence indicates that forest sinthe late Cret aceous were drier and more open than current rain forests (Tiffne y198 4;Upchurc h and Wolfe 1987 ,1993 ,Wolfe and Upchurc h1987 ;Wing and Bou cher 1998 ;Morley 2000 ;Johnson and Ell is2002 ; Ziegler et al 2003 ),molecu lar divergen ce time estimates for lineage sofMalpig hiales (m ajor component sofmod ern tropic al rain forests) indicat ed that 28 major lin eages were derived in tropic al rain forests prima rily betwee n 112 and 94 mya (Davi setal 2005 )These data imply that tropic al rain forests must als ohave been pres ent at this time for the dive rsi?catio nofthese lineages The data for the origins of Piper and Pepero mia presented here pro vide additional suppor tthat tropical forest sexisted during the Cretaceo us Acknowledgments We would like to acknowledge that Chris Fouminyam of the Limbe Botanical Garden, Limbe, Cameroon; George Owusu Afriye ofthe Aburri Botanical Garden, Aburri, Ghana; Jardins etConservatoire Botanique de Nancy, France; Tim Flynn and Dave Lorence of the National Tropical Botanical Garden, Lawai, Hawaii; C W Morden ofthe University ofHawaii; Allan Bornstein of Southeast Missouri State University; Amit Jain, Mindie Funke, Bree Draper, and Wee Seng Wong ofBoise State University; Audrey Mollerup of Whitman College; The Missouri Botanical Garden; Li JiaMei and Wang YinZheng of the Chinese Academy of Sciences all contributed or assisted in obtaining material and data for this project We would also like to thank Stefan Wanke and Marie Stephanie Samain for their comments and for correcting our under standing of Peperomia distribution patterns and two anonymous reviewers who made important suggestions that improved the man uscript The use of the Beowulf Computer Cluster at Boise State University facilitated our data analysis Funding for this project was provided by NSF grant DEB0107763 including RUI support, a Boise State University Faculty Research grant, and agrant from the Marjorie Moore Davidson Foundation, all toJFS Refere nces Alice LA, Campbell CS (1999) Phylogeny ofRubus (Rosaceae) based on nuclear ribosomal DNA internal transcribed spacer region sequences Amer JBot 86:81?97 Alsos IG, Eidesen PB, Ehrich D, Skrede I,Westergaard K, Jacobsen FH, Landvik JY, Taberlet P, Brochmann C (2007) Frequent longdistance plant colonization inthe changing arctic Science 316:1606?1609 ArisBrosou S,Yang Z(2002) Effects ofmodels ofrate evoluti on on estimation of divergence dates with special reference to the metazoan 18S rRNA phylogeny Syst Biol 51:703?714 Australia Plant Names Index http://wwwanbggovau/cpbr/databases/ apnihtml Last visited October 2007 Axelrod DI (1972) Edaphic aridity as afactor in angiosperm evolution Amer Naturalist 106:311?320 Axelrod DI, Raven PH (1978) Late Cretaceous and Tertiary vegetation history ofAfrica In: Werger MJA (ed) Biogeography and ecology ofsouthern Africa Dr W Junk bv Publishers, The Hague, pp 77?130 Baldwin BG, Robichaux RH (1995) Historical biogeography and ecology ofthe Hawaiian silversword alliance (Asteraceae): new molecular phylogenetic perspectives In: Wagner WL, Funk VA (eds) Hawaiian biogeography: evolution on ahot spot archipel ago Smithsonian Institution Press, Washington, DC, pp 259?189 Baldwin BG, Kyhos DW, Dvorak J(1990) Chloroplast DNA variation and adaptive radiation in the Hawaiian silversword alliance (Asteraceae Madiinae) Ann Missouri Bot Gard 77:96?109 Baldwin BG, Kyhos DW, Dvorak J,Carr GD (1991) Chloroplast DNA evidence for aNorth American origin of the Hawaiian silversword alliance (Asteraceae) Proc Natl Acad Sci USA 88:1840?1843 Behrensmeyer AK, Damuth J,DiMichele W, Potts R,Sues HD, Wing S(1992) Terrestrial ecosystems through time University of Chicago Press, Chicago Bell CD (2007) Phylogenetic placement and biogeography of the North Ame rican species of Valerianella (Valerianaceae: Dipsa cales) based on chloroplast and nuclear DNA Molec Phylogenet Evol 44:929?941 Bell CD, Donoghue MJ (2005) Phylogeny and biogeography of Valerianaceae (Dipsacales) with special reference tothe South American valerians Organisms Divers Evol 5:147?159 Borsch T, Hilu KW, Quandt D, Wilde V, Neinhuis C, Barthlott W (2003) Noncoding plastid trnTtrnFsequences reveal awell resolved phylogeny of basal angiosperms JEvol Biol 16:558? 576 Borsch T, Lo?hne C, Mu?ller K, Wanke S, Worberg A, Barthlott W, Neinhus C, Hilu KW, Quandt D(2005) Towards understanding basal angiosperm diversi?cation: recent insights using fast evolving genomic regions Nova Acta Leopoldina 342:85?110 Carine MA, Russell SJ, SantosGuerra A, FranciscoOrtega J(2004) Relationships of the Macaronesian and Mediterranean ?oras: molecular evidence for multiple colonizations into Macaronesia and backcolonization ofthe continent inConvolvulus (Convol vulaceae) Amer JBot 91:1070?1085 Cheng YQ, Xia NH, Gilbe rtMG (1999) Piperaceae In: Wu ZY, Raven PH (eds) Flora of China, Cycadaceae through Fagaceae Missouri Botanical Garden Press St Louis, Missouri, USA, pp 110?129 Chew WL (1972) The genus Piper (Piperaceae) in New Guinea, Solomon Islands, and Australia, 1JArnold Arbor 53:1?25 Conti E, Eriksson T, Scho ?nenberger J,Sytsma KJ, Baum DA (2002) Early Tertiary outofIndia dispersal of Crypteroniaceae: evi dence from phylogeny and molecular dating Evolution 56:1931?1942 Cox CB, Moore PD (2000) Biogeography An ecological and evolutionary approach, 6th edn Blackwell, London Crisp MD, Cook LG (2007) A congruent molecular signature of vicariance across multiple plant lineages Molec Phylogenet Evol 43:1106?1117 Cronk QCB, Kiehn M, Wagner WL, Smith JF (2005) Evolution of Cyrtandra (Gesneriaceae) inthe Paci?c Ocean: the origin of a supertramp clade Amer JBot 92:1017?1024 Darwin C (1859) On the origin of species by means of natural selection, orthe preservation offavoured races inthe struggle for life John Murray, London Davis CC, Bell CD, Fritsch PW, Mathews S(2002) Phylogeny of AcridocarpusBrachylon (Malpighiaceae): implications for 26 JFSmith etal 123 Tertiary tropical ?oras and Afroasian biogeography Evolution 56:2395?2405 Davis CC, Webb CO, Wurdack KJ, Jaramillo CA, Donoghue MJ (2005) Explosive radiation of Malpighiales supports amid Cretaceous origin of modern tropical rain forests Amer Natu ralist 165:E36?E65 de Groot H, Wanke S, Neinhuis C(2006) Revision of the genus Aristolochia (Aristolochiaceae) in Africa, Madagascar and adjacent islands Bot JLinn Soc 151:219?238 DeJoode DR, Wendel JF (1992) Genetic diversity and origin of the Hawaiian islands cotton, Gossypium tomentosum Amer JBot 79:1311?1319 de Queiroz A (2005) The resurrection of oceanic dispersal in historical biogeography Trends Ecol Evol 20:68?73 Des Marais DL, Smith AR, Britton DM, Pryer KM (2003) Phyloge netic relationship sand evolution ofextant horsetails, Equisetum , based on chloroplast DNA sequence data (rbc Land trnLF) Int J PlSci 164:737?751 Dick CW, AbdulSalim K, Bermingham E (2003) Molecular systematic analysis reveals cryptic Tertiary diversi?cation of a widespread tropical rain forest tree Amer Naturalist 162:691? 703 Doyle JA, Endress PK (2000) Morphological phylogene tic analysis of basal angiosperms: comparison and combination with molecular data Int JPlSci 161:S121?S151 Doyle JA, Endress PK (2007) Integrating early Cretaceous fossils into the phylogeny ofrecent angiosperms Plant Biology and Botany 2007, p150 (abstract) Dutech C, Maggia L, Tardy C, Joly HI, Jarne P(2003) Tracking a genetic signal of extinctionrecolonization events inaneotrop ical tree species: Vouacapoua americana Aublet in French Guiana Evolution 57:2753?2764 Drummond A J, Rambaut A (2005) BEAST v13 http://evolve zoooxacuk/beast/ Eklund H, Doyle JA, Herendeen PS (2004) Morphological phyloge netic analysis of living and fossil Chloranthaceae Int JPlSci 165:107?151 Ellstrand NC, Schierenbeck KA (2000) Hybridization as astimulus for the evolution ofinvasiveness inplants? Proc Natl Acad Sci USA 97:7043?7050 Farris SJ, Ka?llersjo ?M, Kluge AG, Bult C(1994) Testing signi?cance ofincongruence Cladistics 10:315?319 Felsenstein J(1985) Con?dence limits on phylogenies: an approach using the bootstrap Evolution 39:783?791 Fleming T(1988) The shorttailed fruit bat: astudy inplantanimal interactions University ofChicago Press, Chicago Fleming T(2004) Dispersal ecology ofNeotropical Piper shrubs and treelets In: Dyer LA, Palmer ADN (eds) Piper :Amodel genus for studies ofPhytochemistry, Ecology, and Evolution Kluwer, New York, pp 58?77 Friis EM, Pedersen KR, Crane PR (1995) Appomattoxia ancistro phora gen et sp nov, anew early Cretaceous plant with similarities toCircaeaster and extant Magnoliidae Amer JBot 82:933?943 Ganders FR, Berbee M, Pirseyedi M (2000) ITS based sequence phylogeny inBidens (Asteraceae): evidence for the continental relatives ofHawaiian and Marquesan Bidens Syst Bot 25:122? 133 Gentry AH (1982) Neotropical ?oristic diversity: phytogeographical connections between Central and South America, Pleistocene climatic ?uctuations, oran accident ofthe Andean orogeny? Ann Missouri Bot Gard 69:557?593 Gentry AH (1989) Speciation in tropical forests In: HolmNielsen LB, Nielsen IC, Balslev H (eds) Tropical forests: botanical dynamics, speciation and diversity Academic Press, London, pp 113?134 GivnishTJ,Sytsma KJ,HahnWJ,Sm ithJF(1995)Molecularevolution, adaptive radiation,andgeographicspeciationinCyanea (Cam pa nulaceae,Lobelioideae)In:Wa gnerWL ,FunkVA(eds)Hawaiian biogeography:evolution on ahotspotarchipelagoSm ithsonian InstitutionPress,Washington,DC ,pp288?337 Givnish TJ, Knox E,Patterson TB, Hapeman JR, Palmer JD, Sytsma KJ (1996) The Hawaiian lobelioids are monophyletic and underwent arapid initial radiation roughly 15 million years ago Amer JBot 83:159 (abstract) Givnish TJ, Evans TM, Zjhra ML, Patterson TB, Berry PE, Sytsma KJ (2000) Molecular evolution, adaptive radiation, and geo graphic diversi?cation inthe amphiatlantic family Rapateaceae: evidence from ndh F sequences and morphology Evolution 54:1915?1937 Givnish TJ, Millam KC, Evans TM, Hall JC, Pires JC, Berry PE, Sytsma KJ (2004) Ancient vicariance or recent longdistance dispersal? Inferences about phylogeny and South American African disjunctions inRapateaceae and Bromeliaceae based on ndh Fsequence data Int JPlSci 165:S35?S54 Goldman N, Anderson JP, Rodrigo AG (2000) Likelihoodbased tests inphylogenetics Syst Biol 49:652?670 Gorchov DL, Cornejo F,Ascorra CF, Jaramillo M(1993) The role of seed dispersal in the natural regeneration of rain forest after stripcutting inthe Peruvian Amazon PlEcol 107?108:339?349 Gorchov DL, Cornejo F, Ascorra CF, Jaramillo M (1995) Dietary overlap between frugivorous birds and bats in the Peruvian Amazon Oikos 74:235?250 GregoryWodzicki KM (2000) Uplift history of the central and northern Andes: areview GSA Bull 112:1091?1105 Grieg N(2004) Introduction In: Dyer LA, Palmer ADN (eds) Piper : A model genus for studies of Phytochemistry, Ecology, and Evolution Kluwer, New York, pp 1?4 Hall R(1998) The plate tectonics ofCenozoic Southeast Asia and the distribution of land and sea In: Hall R, Holloway JD (eds) Biogeography and geological evolution of Southeast Asia Backhuys Publishers, Leiden, The Netherlands, pp 99?131 Hasebe M, Ando T, Iwatsuki K(1998) Intrageneric relationships of maple trees based on the chloroplast DNA restriction fragment length polymorphisms JPlRes 111:441?451 Helfgott DM, FranciscoOrtega J, SantosGuerra A, Jansen RK, Simpson BB (2000) Biogeography and breeding system evolu tion ofthe woody Bencomia alliance (Rosaceae) inMacaronesia based on ITS sequence data Syst Bot 25:82?97 Howarth DG, Gardner DE, Morden CW (1997) Phylogeny ofRubus subgenus Ideobatus (Rosaceae) and its implications toward colonization ofthe Hawaiian islands Syst Bot 22:433?441 Howarth DG, Gustafsson MHG, Baum DA, Motley TJ (2003) Phylogenetics of the genus Scaevola (Goodeniaceae): implica tion for dispersal patterns across the Paci?c Basin and colonization ofthe Hawaiian islands Amer JBot 90:915?923 Horn SP, Sanford RL, Dilcher D, Lott TA, Renne PR, Wiemann MC, Cozadd D, Vargas O(2003) Pleistocene plant fossils inand near La Selva Biological Station, Costa Rica Biotropica 35:434?441 Huang J,Price RA (2003) Estimation of the age of extant Ephedra using chloroplast rbc Lsequence data Molec Biol Evol 20:435? 440 Huelsenbeck JP, Ronquist F(2001) MRBAYES: Bayesian inference ofphylogeny Bioinformatics 17:754?755 Hughes C, Eastwood R(2006) Island radiations on acontinental scale: exceptional rates ofplant diversi?cation after uplift ofthe Andes Proc Natl Acad Sci USA 103:10334?10339 Humphries CJ, Parenti LR (1999) Cladistic biogeography Oxford University Press, Oxford, UK IckertBond SM, Wojciechowski MF (2004) Phylogenetic relation ships in Ephedra (Gnetales): evidence from nuclear and chloroplast DNA sequence data Syst Bot 29:834?849 Biogeography ofPiper 27 123 Jaramillo C, Rueda MJ, Mora G(2006) Cenozoic plant diversity in the Neotropics Science 311:1893?1896 JaramilloMA,Ma nosPS(2001)Phylogenyandpatternsof?oral diversityinthegenusPiper(Piperaceae)AmerJBot88:706?716 Jaramillo MA, Marquis R(2004) Future research inPiper biology In: Dyer LA, Palmer ADN (eds) Piper :A model genus for studies of Phytochemistry, Ecology, and Evolution Kluwer, New York, pp 199?203 Jaramillo MA, Kramer EM (2007) Molecular evolution of the petal and stamen identity genes, APETALA3 and PISTILLATA ,after petal loss inthe Piperales Molec Phylogenet Evol 44:598?609 Jaramillo MA, Manos PS, Zimmer EA (2004) Phylogenetic relation ships of the perian thless Piperales: reconstructing the evolution of?oral development Int JPlSci 165:403?416 Jaramillo MA, Callejas R, Davidson C, Smith JF, Stevens AC, Tepe EJ (in press) Aphylogeny of the tropical genus Piper (Piper aceae) using ITS and the chloroplast intron psb Jpet A Syst Bot Johnson KR, Ellis B(2002) Atropical rainforest in Colorado 14 million years after the CretaceousTertiar yboundary Science 296:2379?2383 Kim YD, Jansen RK (1998) Chloroplast DNA restriction site variation and phylogeny of the Berberidaceae Amer JBot 85:1766?1778 Kim YD, Kim SH (1999) Phylogeny of Weigela and Diervilla (Caprifoliaceae) based on nuclear rDNA ITS sequences: bio geographic and taxonomic implications JPlRes 112:331?341 Kishino H, Hasegawa M (1989) Evaluation of the maxim um likelihood estimate of the evolutionary tree topologies from DNA sequence data, and the branching order in Hominoidea JMolec Evol 29:170?179 Kishino H, Thorne JL, Bruno WJ (2001) Performance ofadivergence time estimation method under aprobabilistic model of rate evolution Molec Biol Evol 18:352?361 Knox EB, Downie SR, Palmer JD (1993) Chloroplast genome rearrangements and the evolution of giant lobelias from herba ceous ancestors Molec Biol Evol 10:414?430 Kubitzki K, Krutzsch W (1996) Origins ofEast and South Asian plant diversity In: Auluo Z, Sugong W (eds) Floristic characteristics and diversity ofEast Asian plants Springer, Berlin, pp 65?70 Lamb S(2004) Devil inthe mountain: asearch for the origin ofthe Andes Princeton University Press, Princeton, NJ Lindqvist C,Albert VA (2002) Origin ofthe Hawaiian endemic mints within North American Stachys (Lamiaceae) Amer JBot 89:1709?1724 Lee NS, Sang T,Crawford DJ, Yeau SH, Kim SC (1996) Molecular divergence between disjunct taxa in eastern Asia and eastern North America Amer JBot 83:1373?1378 Mack RN, Simberloff D, Lonsdale WM (2000) Biotic invasions: causes, epidemiology, global consequences, and control Ecol Appl 10:689?710 MacPhail MK, Partridge AD, Truswell EM (1999) Fossil pollen records of the problematical primitive angiosperm family Lactoridaceae inAustralia PlSyst Evol 214:199?210 Magallo ?nS,Sanderson MJ (2001) Absolute diversi?cation rates in angiosperm clades Evolution 55:1762?1780 Mai DH (1995) Tertia ?reVegetationsgeschichte Europas Fischer, Jena Manchester SR (1989) Early history ofthe Juglandaceae PlSyst Evol 162:231?250 Manchester SR (1991) Cruciptera ,anew Juglandace ous winged fruit from the Eocene and Oligocene ofwestern North America Syst Bot 16:715?725 Manchester SR (1999) Biogeogra phical relationships of North American Tertiary ?oras Ann Missouri Bot Garden 86:472?522 Manchester SR, Wheeler EA (1993) Extinct juglandaceous wood from the Eocene of Oregon and its implications for xylem evolution inthe Juglandaceae IAWA J14:103?111 Marquis RJ (2004) Biogeography ofNeotropical Piper In: Dyer LA, Palmer ADN (eds) Piper :A model genus for studies of Phytochemistry, Ecology, and Evolution Kluwer, New York, pp 78?96 McGlone MS (2005) Goodbye Gondwana JBiogeography 32:739? 740 Miquel F A (1843?1844) Systema Piperacearum HAKramer, Rotterdam Mo?ller M, Cronk QCB (2001) Phylogenetic studies inStreptocarpus (Gesneriaceae): reconstruction of biogeographic history and distribution patterns Syst Geogr Plants 71:545?555 Morat P(1993) Our knowledge of the ?ora of New Caledonia: endemism and diversity in relation to vegetation types and substrates Biodivers Lett 1:72?81 Morley RJ (2000) Origin and evolution oftropical rain forests Wiley, New York Morris AB, Bell CD, Clayton JW, Judd WS, Soltis DE, Soltis PS (2007) Phylogeny and divergence time estimation in Illicium with implications for New World biogeography Syst Bot 32:236?249 Nelson G, Platnick NI (1981) Systematics and biogeography: cladistics and vicariance Columbia University Press, New York Neinhuis C,Wanke S,Hilu KW, Mu?ller K,Borsch T(2005) Phylogeny ofAristolochiaceae based on parsimony, likelihood, and Bayesian analyses oftrnLtrnFsequences PlSyst Evol 250:7?26 Nepokroeff M, Sytsma KJ, Wagner WL, Zimmer EA (2003) Reconstructing ancestral patterns of colonization and dispersal inthe Hawaiian understory tree genus Psychotria (Rubiaceae): a comparison of parsimony and likelihood approaches Syst Biol 52:820?838 Nickrent D, L, Blarer A, Qiu YL, Soltis DE, Soltis PS, Zanis M (2002) Molecular data place Hydnoraceae with Aristolochiaceae Amer JBot 89:1809?1817 Olmstead RG, Palmer JD (1994) Chloroplast DNA systematics: a review ofmethods and data analysis Amer JBot 81:1205?1224 Pennington RT, Lavin M, Prado DE, Pendry CA, Pell SK, Butter worth CA (2004) Historical climate change and specia tion: neotropical seasonally dry forest plants show patterns of both Tertiary and Quaternary diversi?cation Phil Trans Royal Soc London Series B359:515?537 Petit RJ, Bialozyt R, GarnierGe ?re?P,Hampe A(2004) Ecology and genetics of tree invasions: from recent introductions toQuater nary migrations Forest Ecol Management 197:117?137 Posada D, Buckley TR (2004) Model selection and model averaging in phylogenetics: Advantages of Akaike information criterion and Bayesian approaches over likelihood ratio tests Syst Biol 53:793?808 Posada D, Crandall KA (1998) Modeltest: testing the model ofDNA substitution Bioinformatics 14:817?818 Poux C, Madsen O, Marquard E,Vieites DR, Jong WWD, Vences M (2005) Asynchronous colonization of Madagascar by the four endemic clades of primates, tenrecs, carnivores, and rodents as inferred from nuclear genes Syst Biol 54:719?730 Prance GT (1982) Biological diversi?cation inthe tropics Columbia University Press, New York Qiu YL, Parks CR, Chase MW (1995) Molecular divergence in eastern Asianeastern North American disjunct section Rytido spermum of Magnolia (Magnoliaceae) Amer JBot 82:1589? 1598 Qiu YL, Lee J,BernasconiQuadroni F,Soltis DE, Soltis PS, Zanis M, Zimmer EA, Chen Z, Savolainen V, Chase MW (1999) The earliest angiosperms: evidence from mitochondrial, plastid, and nuclear genomes Nature 402:404?407 QuijanoAbril MA, CallejasPosada R, MirandaEsquivel DR (2006) Areas of endemism and distribution patterns for Neotropical Piper species (Piperaceae) JBiogeogr 33:1266?1278 28 JFSmith etal 123 Rambaut A, Bromham L(1998) Estimating divergence dates from molecular sequences Molec Biol Evol 15:442?448 Raven PH, Axelrod DI (1974) Angios perm biogeography and past continental movements Ann Missouri Bot Gard 61:539?673 Ree RH, Moore BR, Webb CO, Donoghue MJ (2005) Alikelihood framework for inferring the evolution of geographic range on phylogenetic trees Evolution 59:2299?2311 Reeves G, Chase MW, Goldblatt P, Fay MF, Cox AV, LeJeune B, SouzaChies T (2001) Molecular systematics of Iridaceae: evidence from four plastid DNA regions Amer J Bot 88:2074?2087 Renner SS (2004) Plant dispersal across the tropical Atlantic by wind and sea currents Int JPlSci 165:S23?S33 Renner SS, Clausing G, Meyer K(2001) Historical biogeography of Melastomataceae: the roles of Tertiary migration and long distance dispersal Amer JBot 88:1290?1300 Richardson JE, Pennington RT, Pennington TD, Hollingsworth PM (2001) Rapid diversi?cation ofaspeciesrich genus ofNeotrop ical rain forest trees Science 293:2242?2245 Ridley H (1930) The dispersal of plants througho ut the world Ashford, Reeve Ritz CM, Martins L,Mecklenberg R,Goremykin V,Hellwig FH (2007) The molecular phylogeny of Rebutia (Cactaceae) and its allies demonstrates the in?uence ofpaleogeogr aphy on the evolution of South American mountain cacti Amer JBot 94:1321?1332 Ronquist F(1996) DIVA Ver 11 Computer program and manual Available by anonymous FTP (ftpsysbotuuse) from Uppsala University, Uppsala, Sweden Ronquist F(1997) Dispersalvicariance analysis: anew approach to the quanti?cation ofhistorical biogeography Syst Biol 45:195? 203 Sanderson MJ (1997) A nonparametric approach to estimating divergence times inthe absence of rate constancy Molec Biol Evol 14:1218?1231 Sanderson MJ (2002) Estimating absolute rates of molecular evolu tion and divergence times: apenalized likelihood approach Molec Biol Evol 19:101?109 Sanderson MJ (2003) r8s: inferring absolute rates of molecular evolution and divergence times in the absence of amolecular clock Bioinformatics 19:301?302 Sanmart? ?nI,Ronquist F(2004) Southern hemisphere biogeography inferred by eventbased models: plant versus animal patterns Syst Biol 53:216?243 Sa?rkinen TE, Newman MF, Maas PJM, Maas H, Poulsen AD, Harris DJ, Richardson JE, Clark A, Hollingsworth M, Pennington RT (2007) Recent oceanic longdistance dispersal and divergence in the amphiAtlantic rain forest genus Renealmia L f(Zingiber aceae) Molec Phylogenet Evol 44:968?980 Schnabel A, Wendel JF (1998) Cladistic bioge ography of Gleditsia (Leguminosae) based on ndh F and rpl16 chloroplast gene sequences Amer JBot 85:1753?1765 Seelanen T, Schnabel A, Wendel JF (1997) Congruence and consensus inthe cotton tribe (Malvaceae) Syst Bot 22:259?290 Shimodaira H, Hasegawa M (1999) Multiple comparisons of log likelihoods with applications to phylogenetic inference Molec Biol Evol 16:1114?1116 Simmons MP, Ochoterena H(2000) Gaps ascharacters insequence based phylogenetic analyses Syst Biol 49:369?381 Smith SA (1975) The genus Macropiper (Piperaceae) Biol JLinn Soc 71:1?38 Smith S A (2006) AReA: Ancestral range analysis v 21 http://blackrimorg/programs/areahtml Smith SY, Stockey RA (2007) Establishing afossil record for the perianthless Piperales: Saururus tuckerae sp nov (Saururaceae) from the Middle Eocene Princeton Chert Amer JBot 94:1642? 1657 Smith JF, Funke MM, Woo VL (2006) Aduplication ofgcyc preda tes divergence within tribe Coronanthereae (Gesneriaceae): phylo genetic analysis and evolution PlSyst Evol 261:245?256 Sm ithJF,HilemanLC,PowellM, Baum DA (2004)Evolutionofgcyc,a Gesneriaceaehomo logofCY CL OIDE A,withinsubfamilyGesne rioideae(Gesneriaceae)MolecPhylogenetEvol31:765?779 Soltis DE, Soltis PS, Chase MW, Mort ME, Albach DC, Zanis M, Savolainen V, Hahn WH, Hoot SB, Fay MF, Axtell M, Swensen SM, Prince LM, Kress WJ, Nixon KC, Farris JS (2000) Angiosperm phylogeny inferred from 18S rDNA, rbc L, and atp Bsequences Bot JLinn Soc 133:381?461 Strand AE, LeebensMack J,Milligan BG (1997) Nuclear DNA based markers for plant evolutionary biology Molec Ecol 6:113?118 Stuessy TF, Crawford DJ, Anderson GJ, Jensen RJ (1998a) System atics, biogeography and conservation of Lactoridaceae Pers Pl Ecol Evol Syst 1:267?290 Stuessy TF, Crawford DJ, Marticorena C,Rodriguez R(1998b) Island biogeography ofangiosperms ofthe Juan Fernandez archipelago In: Stuessy TF, Ono M (eds) Evolution and speciation ofisland plants Cambridge University Press, Cambridge UK, pp 121?138 Swenson U, Bremer K(1997) Paci?c biogeography ofthe Asteraceae genus Abrotanella (Senecioneae, Blennospermatinae) Syst Bot 22:493?508 Swenson U, Hill RS, McLoughlin S (2001) Biogeography of Nothofagus supports the sequence of Gondwana breakup Taxon 50:1025?1041 Swofford D L(2002) PAUP* Phylogenetic analysis using parsimony (* and Other Methods) Version 10 Sinauer Associates, Sunderland, Massachusetts Thorne JL, Kishino H(2002) Divergence time estimation and rate evolution with multilocus data sets Syst Biol 51:689?702 Thorne JL, Kishino H, Painter IS (1998) Estimating the rate of evolution of the rate of molecular evolution Molec Biol Evol 15:1647?1657 Tiffney BH (1984) Seed size, dispersal syndromes, and the rise ofthe angiosperms: evidence and hypothesis Ann Missouri Bot Gard 71:551?576 Tiffney BH (1985a) Perspectives on the origin of the ?oristic similarity between eastern Asia and eastern North America J Arnold Arbor 66:73?94 Tiffney BH (1985b) The Eocene North Atlantic land bridge: its importance in Tertiary and modern phytogeography of the Northern hemisphere JArnold Arbor 66:243?273 TiffneyBH,Ma nchester SR (2001) Theuseof geologicaland paleontologicalevidenceinevaluatingplantphylogenetichypoth esesintheNorthernhemisphereTertiaryIntJPlSci162:S3?S17 Upchurch GR, Wolfe JA (1987) MidCretaceous to early Tertiary vegetation and climate: evidence from fossil leaves and wood In: Friis EM, Chaloner WG, Crane PR (eds) The origins of angiosperms and their biological consequences Cambridge University Press, Cambridge, UK, pp 75?105 Upchurch GR, Wolfe JA (1993) Cretaceous vegetation ofthe western interior and adjacent regions of North America In: Kauffman EG, Caldwell WGE (eds) Cretaceous evoluti on of the western interior basin Geological Association ofCanada, Special Paper, vol 39, pp 243?281 Valdebenito HA, Stuessy TF, Crawford DJ (1990) Anew biogeo graphic connection between islands inthe Atlantic and Paci?c oceans Nature 347:549?550 Valdebenito HA, Stuessy TF, Crawford DJ, Silva OM (1992) Evolution of Peperomia (Piperaceae) in the Juan Fernandez Islands, Chile PlSyst Evol 182:107?119 Wagner WL, Funk VA (1995) Hawaiian biogoegraphy: evolution on ahot spot archipelago Smithsonian Institution Press, Washing ton, DC Biogeography ofPiper 29 123 Wagner W L, Herbst D R, Sohmer S H (1990) Manual of the ?owering plants of Hawai?i University of Hawai?i Press, Honolulu, Hawai?i, USA Wanke S,Samain MS, Vanderschaev eL,Mathieu G, Goetghebeur P, Neinhuis C(2006) Phylogeny of the genus Peperomia (Piper aceae) inferr ed from the trnK/mat Kregion (cpDNA) Pl Biol 8:93?102 Wanke S,Jaramillo MA, Borsch T,Samain MS, Quandt D, Neinhuis C(2007a) Evolution of the Piperales matK and trnK intron sequence data reveal lineage speci?c resolution contrast Molec Phylogenet Evol 42:477?497 Wanke S,Vanderschaeve L,Mathieu G, Neinhuis C, Goetghebeur P, Samain MS (2007b) From forgotten taxon to amissing link? The position of the genus Verhuellia (Piperaceae) revealed by molecules Ann Bot 99:1231?1238 Wen J,Stuessy TF (1993) The phylogeny and biogeography ofNyssa (Cornaceae) Syst Bot 18:68?79 Wen J,Jansen RK (1995) Morphological and molecular comparisons of Campsis grandi?ora and C radicans (Bignoniaceae), an eastern Asian and eastern North American vicariad species pair PlSyst Evol 196:173?183 Wen J,Zimmer EA (1996) Phylogeny and biogeography ofPanax L (the ginseng genus, Araliaceae): inferences from ITS seq uences of nuclear ribosomal DNA Molec Phylogenet Evol 6:167?177 Wen J,Janzen RK, Zimmer EA (1996) Phylogenetic relationships and DNA sequence divergence of eastern Asian and eastern North American disjunct plants In: Nei M, Takahata N(eds) Current topics ofmolecular evolution Jointly published by the Pennsyl vania State University and the Graduate University for Advanced Studies, Hayama, Japan, pp 37?44 Wen J, Shi S, Jansen RK, Zimmer EA (1998) Phylogeny and biogeography of Aralia sect Aralia (Araliaceae) Amer JBot 85:866?875 Whitmore TC (1998) An introduction totropical rain forests, 2nd edn Oxford University Press, Oxford Whitmore TC, Prance GT (1987) Biogoegraphy and Quaternary history intropical America Oxford University Press, New York Williams DA, Overholt WA, Cuda JP, Hughes CR (2005) Chloroplast and microsatellite DNA diversities reveal the introduction history of Brazilian peppertree (Schinus terebinthifolius )in Florida Molec Ecol 14:3643?3656 Wing SL, Boucher LD (1998) Ecological aspects of the Cretaceous ?owering plant radiation Annual Rev Earth Planet Sci 26:379? 421 Winkworth RC, Wagstaff SJ, Glenny D, Lockhart PJ (2005) Evolution of the New Zealand mountain ?ora: origins, diversi ?cation and dispersal Organisms Div Evol 5:237?247 Wolfe JA, Upchurch GR (1987) North American nonmarine climates and vegetation during the Late Cretaceous Palaeog Palaeoclim Palaeoecol 61:33?77 Won H, Renner SS (2006) Dating dispersal and radiation in the Gymnosperm Gnetum (Gnetales) Clock calibration when outgroup relationships are uncertain Syst Biol 55:610?622 Wright SD, Yong CG, Dawson JW, Whittaker DJ, Gardner RC (2000) Riding the ice age El Nin?o? Paci?c biogeography and evolution of Metrosideros (Myrtaceae) inferred from nuclear ribosomal DNA Proc Natl Acad Sci USA 97:4118?4123 Wright SD, Yong CG, Dawson JW, Whittaker DJ, Gardner RC (2001) Stepping stones toHawaii: atransequatorial dispersal pathway for Metrosideros (Myrtaceae) inferred from nrDNA (ITS plus ETS) JBiogeogr 28:769?774 Xiang QY, Crawford DJ, Wolfe AD, Tang YC, DePamphilis CE (1998) Origin and biogeography ofAesculus (Hippocastanaceae): amolecular phylogenetic perspective Evolution 52:988?997 Yoder AD, Irwin JA, Payseur BA (2001) Failure of the ILD to determine data combinability for slow loris phylogeny Syst Biol 50:408?424 Yoo MJ, Bell CD, Soltis PS, Soltis DE (2005) Divergence times and historical biogeography ofNymphaeales Syst Bot 30:693?704 Young KR, Ulloa UC, Luteyn JL, Knapp S(2002) Plant evolution and endemism inAndean South America: an introduction Bot Rev 68:4?21 Zanis MJ, Soltis DE, Soltis PS, Mathews S,Donoghue MJ (2002) The root of the angiosperms revisited Proc Natl Acad Sci USA 99:6848?6853 Zavada MS, Benson JM (1987) First fossil evidence for the primitive angiosperm family Lactoridaceae Amer JBot 74:1590?1594 Zhang LB, Renner SS (2003) The deepest splits inChloranthaceae as resolved by chloroplast sequences Int JPlSci 164:S383?S392 Zhang LB, Simmons MP, Renner SS (2007) A phylogeny of Anisophyllaceae based on six nuclear and plastic loci: ancient disjunctions and recent dispersal between South America, Africa, and Asia Molec Phylogenet Evol 44:1057?1067 Zhou S,Renner SS, Wen J(2006) Molecular phylogeny and inter and intracontinental biogeography ofCalycanthaceae Molec Phylo genet Evol 39:1?15 Ziegler AM, Eshel G, Rees PM, Rothfus TA, Rowley DB (2003) Tracing the tropics across land and sea: Permian to present Lethaia 36:227?254 30 JFSmith etal 123

------------- Read More -------------

Download smith-et-al-2008.pdf

Smith et al., 2008 related documents

November 13, 2008 Nanaimo, BC Self-Guided Country Christmas Craft Tour

2 Pages · 2008 · 56 KB · English

November 13, 2008 Nanaimo, BC Self-Guided Country Christmas Craft Tour For immediate release: The Self-Guided Country Christmas Craft Tour, organized by the

ETICA, FILOSOFIA E MEDIAZIONE LINGUISTICA Dall'Etica della Filosofia al Codice Deontologico ...

20 Pages · 2012 · 463 KB · English

As an object of theoretical inquiry (meta-ethics) as well as a guide for human conduct (normative ethics), ethics has been a prime focus of Western moral philosophy since the time of the Ancient un rappresentante di un'istituzione pubblica, o privata, che offre servizi basilari ai cittadini. (ospe

TCF July 2008

6 Pages · 2008 · 7.03 MB · English

with funding from the Pikes Peak Compassionate Friends. Feel free to . Accidental overdose. \iehicle accident. \iiral meningo-encephalitis. Stillborn.

Winter 2007-2008

5 Pages · 2007 · 1.56 MB · English

a taste test between grass fed beef, grain finished beef, local goat will search the internet using the key words: The satisfaction of helping promote and pre- While the Traveler specifically mentions a store . word agritourism is a “hybrid” of two words: agri- Intermediate GIS for Vineyard

Spinoza et le portrait de l'homme libre

14 Pages · 2012 · 104 KB · French

Jacqueline Lagrée : Spinoza et le portrait de l'homme libre " Reproduction et utilisation interdites sans l'accord explicite de l'auteur ou du C.R.U.. -1-. Cercle de . conduit par la raison, comme tout individu forge un modèle idéal d'homme parfait en fonction de son désir et de .. Transmutatio

new development & construction controls city of brentood 2008-2009

29 Pages · 2009 · 10.92 MB · English

will be held on Thursday, September 18, 2008 at Oakley's new City Hall Oakley City Hall Steve Palmer . II AC II cc II CE II cc II DA II EO II INSP II NOV II OA II RB leml RTC II sw II vw II OTH II Total I .. 15 TSM 8311 1DR 05-20.

Securing the Bomb 2008

205 Pages · 2008 · 1.28 MB · English

A companion website to this report is available at .. top priority of the U.S. government—and makes clear .. major effort to keep nuclear weapons and the materials needed to make them out of terrorist hands. In particular, since new opportunities to improve nucl

De la obra al texto

6 Pages · 2008 · 47 KB · Spanish

Por Roland Barthes. Es un hecho comprobado que desde hace algunos años se ha operado (o se opera) un cierto cambio en el interior de la idea que nos hacemos del lenguaje y, en consecuencia, de la obra (literaria) que debe a este mismo lenguaje al menos su existencia fenoménica. Este cambio 

May 2008 LIGHT TRUCKS - Indiana

12 Pages · 2008 · 652 KB · English

In 2006, light trucks (pickup trucks,1 sport utility vehicles, and vans) comprised approxi-mately one-third of registered vehicles and 40 percent of vehicles involved

Vers une politique gouvernementale de lutte contre le racisme et la discrimination

14 Pages · 2006 · 263 KB · French

contre le racisme et la discrimination. Synthèse du document de consultation. Pour la pleine participation des Québécoises et des Québécois des communautés culturelles .. réalité pluraliste du Québec moderne dans sa quotidienneté et qu'elle se donne un cadre d'intervention qui assure