American Journal of Botany 98(10): 1716–1726. 2011. MOLECULAR PHYLOGENETICS OF THE PALM SUBTRIBE PTYCHOSPERMATINAE (ARECACEAE)1 Scott Zona2,5, Javier Francisco-Ortega2,3, Brett Jestrow3, William J. Baker4, and Carl E. Lewis3 2Department 3Center of Biological Sciences OE167, Florida International University 11200 SW 8 St., Miami, Florida 33199 USA; for Tropical Plant Conservation, Fairchild Tropical Botanic Garden 11935 Old Cutler Road, Coral Gables, Miami, Florida 33156 USA; and 4Royal Botanic Gardens, Kew, Richmond Surrey TW9 3AB, UK • Premise of the study: We examined the phylogeny and intergeneric relationships among the 12 genera of the palm subtribe Ptychospermatinae. While many of these taxa are familiar, cultivated ornamental palms in warm areas of the world, the monophyly of the subtribe and its component genera required testing. We also examined the biogeographic relationships of this lineage, which has a significant radiation east of Wallace’s Line. • Methods: Phylogenetic analyses were based on maximum parsimony and Bayesian analyses of nucleotide sequences of two low-copy nuclear genes: intron 4 of phosphoribulokinase and intron 23 of RNA polymerase II. Biogeographical reconstructions were explored using S-DIVA. • Key results: The two-gene, combined analysis yielded a monophyletic subtribe with six major clades. The biogeographical analysis suggests that the subtribe originated in New Guinea. • Conclusions: The phylogenetic hypotheses support the monophyly of the subtribe. The genera Drymophloeus, Ponapea, and Veitchia, as presently circumscribed, are not monophyletic. The resurrection and expanded circumscription of the genus Ponapea are supported. A newly discovered species of Adonidia is confirmed as sister species to Adonidia merrillii. Our phylogenetic hypothesis suggests that the Ptychospermatinae diverged into six major clades with repeated radiations into Australia and the western Pacific. The presence of Adonidia to the west of Wallace’s Line is likely to be the result of long-distance dispersal. The following new combinations are made to restore monophyly to Veitchia and Ponapea: Veitchia pachyclada, V. subisticha, V. lepidota, and Ponapea hentyi. Key words: Arecaceae; Areceae; biogeography; Palmae; palms; phylogeny; Ptychospermatinae; systematics; Wallace’s Line. Palms of the subtribe Ptychospermatinae (Arecaceae: Arecoideae: Areceae) were first recognized as a taxonomic group by J. D. Hooker (1883) based on floral and vegetative characters. They formed an easily recognized group based on morphological evidence (Moore, 1973; Uhl and Dransfield, 1987; Zona, 1999a; Dransfield et al., 2008). The distinguishing morphological features of these palms include praemorse leaflet apices; bullet-shaped, staminate flower buds; and numerous stamens. The subtribe comprises 12 genera and ca. 60 species (Dransfield et al., 2008). Many species of Ptychospermatinae are cultivated throughout the warm areas of the world and are among the most popular and familiar cultivated ornamental palms [e.g., Adonidia merrillii (Becc.) Becc., Carpentaria acuminata (H. Wendl. & Drude) Becc., Ptychosperma elegans (R. Br.) Blume, Veitchia arecina Becc. and Wodyetia bifurcata A. K. Irvine]. 1 Manuscript received 13 May 2011; revision accepted 5 August 2011. The authors thank Fairchild Tropical Botanic Garden for a postdoctoral fellowship to C.E.L. and the Tropical Biology Program of Florida International University (TBP-FIU) for the start-up funds to J.F.O. that supported this project. They thank Charlie Heatubun and Universitas Negeri Papua, for sharing DNA samples of the new Adonidia and the late Ray Baker and the Lyon Arboretum for access to their palm collection. They dedicate this paper to Dr. Jack Fisher for his contributions to the understanding of the functional ecology and anatomy of tropical palms. This is contribution number 200 of the TBP-FIU. 5 Author for correspondence (e-mail: zonas@fiu.edu) doi:10.3732/ajb.1100218 Significant progress in delimiting species and producing taxonomic revisions has been made in recent years (see Zona, 1999b, 2005; Zona and Essig, 1999; Zona and Fuller, 1999; Hodel, 2010); only Adonidia Becc. and Ponapea Becc., among the polytypic genera, lack modern taxonomic treatments. Generic boundaries have also received taxonomic scrutiny. Zona (1999a) produced the first generic-level phylogeny using explicit cladistic methods and morphological data. His work was noteworthy in that its results supported the reinstatement of the genera Adonidia and Solfia Rech., which had previously been treated under Veitchia H. Wendl. and Drymophleous Zipp., respectively (Moore, 1973; Uhl and Dransfield, 1987). His results also suggested that an anomalous species of Ptychosperma Labill., Ptychosperma ledermanianum (=Ponapea ledermanniana), belonged in the genus Ptychococcus Becc. His results, however, had low bootstrap support (65% or less). Preliminary results by C. E. Lewis (unpublished data) indicated that Ptychosperma ledermanianum could be accommodated in a reconstituted genus Ponapea, but sampling of this small, poorly known genus had, until this present study, been incomplete. It had been treated as a synonym of Ptychosperma in recent years (Moore, 1973; Uhl and Dransfield, 1987) but was reinstated by Dransfield et al. (2008), based in part on Lewis’ preliminary results. The advent of molecular data gave workers new tools to test the monophyly of the subtribe. The early results, which were familywide analyses based on samples of only a small portion of the Ptychospermatinae, shed limited light on the monophyly of the subtribe. Baker et al. (1999), in their analyses of trnL-trnF American Journal of Botany 98(10): 1716–1726, 2011; http://www.amjbot.org/ © 2011 Botanical Society of America 1716 October 2011] Zona et al.—Phylogeny of the Ptychospermatinae (Arecaceae) data, included two species of Ptychospermatinae, but these species never formed a monophyletic group. Asmussen and Chase (2001) did not find a monophyletic Ptychospermatinae in their separate analyses of rbcL, rps16 intron or trnL-trnF (their combined analysis did, however, identify a monophyletic Ptychospermatinae). Hahn’s (2002) maximum likelihood tree of 18S sequence data recovered a polyphyletic Ptychospermatinae relative to Burretokentia Pichi-Sermoli, a member of subtribe Basseliniinae (Dransfield et al., 2008). Asmussen et al. (2006), sampling eight taxa from different genera in a combined matK, rbcL, rps16 intron, and trnL-trnF data analysis, identified the Ptychospermatinae as a monophyletic group, although with low bootstrap support (62%). They also found four species pairs, although only two were well supported (>90% bootstrap): Carpentaria acuminata and Wodyetia bifucata at 99% and Balaka seemannii and Veitchia arecina at 92%. Norup et al. (2006) sampled 12 taxa from 12 genera and analyzed PRK and RPB2 sequences with maximum parsimony. The subtribe was resolved as monophyletic, but most intergeneric relationships did not have strong bootstrap support. Only the species pair Balaka longirostris and Solfia samoensis had strong support (99%). In a familywide analysis of all palm genera, Baker et al. (2009) found 100% bootstrap support for the Ptychospermatinae in their supermatrix analysis, but generic relationships were not as strongly supported. Only the pair Carpentaria and Wodyetia received more than 90% support. In their parallel supertree analysis, they also recovered a monophyletic Ptychospermatinae. The only difference between the two was the relationship among Drymophloeus, Ptychococcus, and Brassiophoenix, which comprised a weakly supported, monophyletic group in the supertree analysis but not in the supermatrix analysis. The majority of the subtribe Ptychospermatinae is naturally distributed east of Wallace’s Line from the Moluccas through New Guinea, tropical Australia, Solomon Islands, Vanuatu, Fiji, Tonga, and Samoa (Fig. 1). One genus, Adonidia, is found Fig. 1. 1717 on the island of Palawan and Danjugan Island, off the southwestern coast of Negros Island, the Philippines (Fernando, 2011), and on an island off the northern tip of Sabah, Borneo. It is the only taxon found west of Wallace’s Line. Until recently, Adonidia was presumed monotypic, but a new taxon from Biak Island, in the Indonesia province of Papua, shares some morphological characters with Adonidia and represents a second species in the genus, but one that is disjunct across Wallace’s Line. In this paper, we conduct phylogenetic analyses of nucleotide sequences of two low-copy nuclear genes (phosphoribulokinase (PRK, a Calvin cycle enzyme) and RNA polymerase II (RPB2, one subunit of a transcription enzyme) for the Ptychospermatinae. These two genes have not been reported to be members of large multigene families, and they have been widely used in phylogenetic reconstructions of the palm family (e.g., Roncal et al., 2005; Loo et al., 2006; Thomas et al., 2006). For our study, we sampled all genera of the subtribe including the putative new Adonidia from Biak Island, Papua, Indonesia. Our aim is to test whether the subtribe is monophyletic and whether the proposed generic relationships are supported. Moreover, the analyses will examine current generic boundaries to determine whether the genera are monophyletic. A final aim is to analyze the biogeography of the group, which shows a significant radiation in New Guinea and other areas east of Wallace’s Line. A robust molecular phylogeny is a prerequisite for all objectives. MATERIALS AND METHODS Sampling—We sampled 37 species from 12 genera of the Ptychospermatinae. This sampling represents ca. 60% of the species and 100% of the genera in the subtribe. Sequences for 23 of these species were obtained for the present study (Table 1), the rest of the sequences were retrieved from GenBank. Multiple taxa from all polytypic genera (with the exception of Ptychococcus and Brassiophoenix, genera of two species each) were sampled to assess generic monophyly. We also included taxa from subtribes Archontophoenicinae Map showing the generalized distribution (shaded) of the Ptychospermatinae. 1718 American Journal of Botany [Vol. 98 Table 1. List of species names, voucher information and NCBI accession numbers for taxa used in this study. Herbarium acronyms follow Thiers (2011). An asterisk marks taxa that are not members of the Ptychospermatinae per Dransfield et al. (2008). Sequences with caret (^) were obtained for the current study. Species Adonidia merrillii (Becc.) Becc. Adonidia sp. nov. Biak Balaka seemannii (H. Wendl.) Becc. Balaka tahitensis (H. Wendl.) Becc. (as B. brachychlamys Burret) Brassiophoenix drymophloeoides Burret Brassiophoenix schumannii (Becc.) Essig Carpentaria acuminata (H. Wendl. & Drude) Becc. *Carpoxylon macrospermum H. Wendl. & Drude *Clinosperma bracteale (Brongn.) Becc. *Cyphosperma balansae (Brongn.) H. Wendl. ex Salomon *Dransfieldia micrantha (Becc.) W. J. Baker & Zona Drymophloeus hentyi (Essig) Zona Drymophloeus litigiosus (Becc.) H. E. Moore Drymophloeus oliviformis (Giseke) Mart. Drymophloeus pachycladus (Burret) H. E. Moore Drymophloeus subdistichus (H. E. Moore) H. E. Moore *Dypsis leptocheilos (Hodel) Beentje & J. Dransf. *Kentiopsis oliviformis (Brongn. & Gris.) Brongn. *Nephrosperma vanhoutteanum (H. Wendl. ex Van Houtte) Balf. f. Normanbya normanbyi (W. Hill) L. H. Bailey *Oncosperma tigillarium (Jack) Ridley Ponapea hosinoi Kaneh. Ponapea ledermanniana Becc. Ponapea palauense Kaneh. Ptychococcus paradoxus (Scheff.) Becc. Ptychosperma burretianum Essig Ptychosperma caryotoides Ridl. Ptychosperma cuneatum (Burret) Burret Ptychosperma elegans (R. Br.) Blume Ptychosperma lauterbachii Becc. Ptychosperma lineare (Burret) Burret) Ptychosperma macarthurii (H. Wendl. ex H. J. Veitch) H. Wendl. ex Hook. f. Ptychosperma microcarpum (Burret) Burret Ptychosperma propinquum (Becc.) Becc. ex Martelli Ptychosperma pullenii Essig Ptychosperma salomonense Burret Ptychosperma sanderianum Ridl. Solfia samoensis Rech. Veitchia arecina Becc. Veitchia filifera (H. Wendl.) H. E. Moore Veitchia metiti Becc. Veitchia spiralis H. Wendl. Veitchia vitiense (H. Wendl.) H. E. Moore Veitchia winin H. E. Moore Wodyetia bifurcata A. K. Irvine Area Voucher/Herbarium PRK RPB2 S. Zona & C. E. Lewis 874/FTG AJ831224 AJ830193 W. Baker et al., 1336/K S. Zona & C. E. Lewis 887/FTG S. Zona et al., 713/FTG JF833370^ JF833372^ JF833371^ JF833393^ JF833395^ JF833394^ New Guinea New Guinea Australia Vanuatu New Caledonia New Caledonia New Guinea New Guinea (Bismarck Arch.) New Guinea New Guinea Solomon Islands Solomon Islands Madagascar New Caledonia Seychelles Islands M. P. Coons 1398/FTG C.E. Lewis 99-044/BISH S. Zona & N. Hernandez 827/FTG S. Zona et al., 722/FTG J.-C. Pintaud 349/K R. F. Baker 89-030/BISH W.J. Baker et al., 1066/K S. Zona & C. E. Lewis 1018/FTG AJ831235 JF833373^ AJ831259 AF45337 AJ831261 AF453340 AJ831326 JF833375^ AJ830195 — AJ830196 AJ830055 AJ830057 AY543098 AJ830139 JF833397^ S. Barrow et al., 125/K S. Zona & F. Hausman 605/FTG S. Zona & C. E. Lewis 880/FTG S. Zona & C. E. Lewis 871/FTG W. J. Baker 988/FTG J.-C. Pintaud 358/K C. E. Lewis 98-006/BH AJ831267 JF833374^ JF833376^ JF833377^ AF453345 AF453353 AF453362 AJ830197 JF833396^ JF833398^ JF833399^ AY779376 AY543101 AJ830131 Australia Thailand to Java, Borneo Micronesia (Pohnpei) Micronesia (Pohnpei) Micronesia (Palau) New Guinea New Guinea New Guinea New Guinea Australia New Guinea New Guinea New Guinea C. E. Lewis 98-091/BH C. E. Lewis 98-051/BH S. Zona & C. E. Lewis 992/FTG S. Zona & C. E. Lewis 878/FTG C. E. Lewis 99-055/BISH W. J. Baker & T. Utteridge 572/K S. Zona & C. E. Lewis 867/FTG S. Zona & C. E. Lewis 868/FTG S. Zona & J. Francisco-Ortega 1123/FTG Lewis s.n./FTG S. Zona & J. Francisco-Ortega 1125/FTG S. Zona & C. E. Lewis 1019/FTG S. Zona & C. E. Lewis 869/FTG AF453363 AF453364 JF833382^ AJ831323 AJ831328 AJ831324 JF833378^ JF833379^ JF833380^ JF833381^ JF833383^ JF833384^ AJ831325 AJ830198 AJ830134 JF833404^ AJ830199 AJ830203 AJ830200 JF833400^ JF833401^ JF833402^ JF833403^ JF833405^ JF833406^ AJ830201 New Guinea New Guinea New Guinea Solomon Islands New Guinea Samoa Vanuatu Fiji Vanuatu Vanuatu Fiji Vanuatu Australia S. Zona & L. T. Smith 965/FTG Jestrow s.n./FTG S. Zona & C. E. Lewis 888/FTG W. M. Houghton 1300/FTG S. Zona & J. Francisco-Ortega 1124/FTG T. Tipama’a 02/FTG W. J. Baker 1003/FTG J. Roncal 049/FTG S. Zona & J. Francisco-Ortega 1122/FTG S. Zona et al., 724/FTG S. Zona 905/FTG S. Zona & C. E. Lewis 881/FTG S. Zona & C. E. Lewis 906/FTG AJ831327 JF833385^ JF833386^ AF453371 JF833387^ AJ831334 JF833388^ JF833389^ JF833390^ AJ831342 JF833391^ JF833392^ AJ831343 AJ830202 JF833407^ JF833408^ AY543105 JF833409^ AJ830204 JF833410^ JF833411^ JF833412^ AJ830205 JF833413^ JF833414^ AJ830206 Philippines (Palawan), Borneo New Guinea (Biak Island) Fiji Samoa (Kentiopsis oliviformis), Basseliniinae (Cyphosperma balansae), Carpoxylinae (Carpoxylon macrospermum), Clinospermatinae (Clinosperma bracteale), Dypsidinae (Dypsis leptocheilos), Oncospermatinae (Oncosperma tigillarium) and Verschaffeltiinae (Nephrosperma vanhoutteanum), along with the unplaced Dransfieldia micrantha, for a total of eight non-ingroup species. The Archontophoenicinae, Basseliniinae, Carpoxylinae, Clinospermatinae, and Dransfieldia are members of the western Pacific clade of Areceae (Norup et al., 2006), to which the Ptychospermatinae also belong. The Dypsidinae, Oncospermatinae, and Verschaffeltiinae are more phylogenetically distant members of the Areceae (Norup et al., 2006), so the species belonging to those subtribes (N. vanhoutteanum, O. tigillarium, and D. leptocheilos) were designated as outgroups for the phylogenetic analyses. Taxonomic sampling, voucher information, and GenBank sequence accession numbers are shown in Table 1. Primer design—The PRK primers prk717f and prk969r were designed based on sequences from a diverse sample of palms as described in a previous paper, and they amplify intron 4 of this gene (Lewis and Doyle, 2002). We used these primers to amplify the PRK gene from all 23 template DNA samples. The RPB2 primers were designed based on palm sequences derived from a pilot study using the universal forward primer RPB2p10f (Denton et al., 1998) and the reverse primer RPB2-M11R (Roncal et al., 2005) designed against monocot RPB2 sequences. These primers target intron 23 of this gene. The pilot study included Adonidia merrillii, Dypsis leptocheilos, and Ptychosperma burretiana. Primers RPB2-INT23BF and RPB2-INT23R (Roncal et al., 2005) that flank intron 23 were used to amplify a region of the RPB2 gene from 22 template DNA samples. We were unable to amplify this gene for Brassiophoenix schumannii. October 2011] Zona et al.—Phylogeny of the Ptychospermatinae (Arecaceae) Amplification and sequencing—DNA was extracted from fresh, frozen, or silica-dried leaf samples using the DNEasy Plant Mini Kit (Qiagen, Valencia California, USA). We used puRe Taq Ready-To-Go PCR Beads (Amersham Biosciences, Piscataway, New Jersey, USA) following the manufacturer’s instructions. For all amplifications, the following temperature profile was used: 4 min at 94°C; followed by 35 cycles of 1 min at 94°C, 1 min at 55°C and 2 min at 72°C; with a final 7 min extension at 72°C. PCR products were cloned because of the presence of homopolymer regions that yielded unreadable sequences with a significant number of ambiguous base calls. Products were cloned using the TOPO TA Cloning Kit for Sequencing (pCR4-TOPO vector, TOP 10 cell line; Invitrogen, Carlsbad, California), following the manufacturer’s instructions. Five to 10 colonies were screened by PCR to identify successful transformants. Colonies of successfully transformed cells were grown in 5 mL liquid cultures in Luri-Bertani (LB) medium. We sent glycerol stock solutions of the liquid cultures to Amplicon Express (Pullman, Washington, USA) for plasmid purification and DNA sequencing. Sequencing was performed in both directions with the ABI PRISM BigDye Terminator Cycle Sequencing Reaction Kit (Perkin Elmer, Waltham, Massachusetts, USA) following the manufacturer’s instructions and using the same primers from the original PCR gene amplification. After cloning and sequencing, forward and reverse sequences for each sample were assembled and edited into contigs using Sequencher 3.2 (Genecodes Corp., Ann Arbor, Michigan, USA). We found that multiple clones of a single DNA extraction were indistinguishable. Phylogenetic analysis—Three data sets were analyzed: (1) PRK alone, (2) RPB2 alone, and (3) PRK and RPB2 combined. We conducted both maximum parsimony (MP) analyses (equal weights, unordered; Fitch, 1971) and Bayesian analyses (Rannala and Yang, 1996). Sequences were aligned using the program Clustal X (Thompson et al., 1997). The final alignment was adjusted manually using the program Se-Al vers. 1 (A. Rambaut, University of Oxford, Oxford, UK). For parsimony analyses, gaps were coded as binary characters according to the ‘‘simple indel coding’’ methodology of Simmons and Ochoterena (2000). Parsimony analyses for the PRK and PRK-RPB2 combined data sets were computed using PAUP* version 4.0b10 (Swofford, 2002). Heuristic searches for most parsimonious trees were performed with 1000 random entries using the DELTRAN, MULTREES, and TBR options. In contrast, a two-step heuristic search was followed for the RPB2 data set because of computer memory limitation. The initial searches were performed by doing 1000 replicates with RANDOM taxon addition, tree-bisection-reconnection (TBR) branch swapping, and MULTREES, and 10 trees per replicate were saved. These heuristic searches resulted in a pool of trees that were used as starting trees in a second heuristic search in which the trees were swapped to completion. An upper limit of 150 000 trees (MAXTREES) was enforced for all searches. Phylogenetic support for each clade was evaluated through bootstrap analysis (Felsenstein, 1985) of 1000 replicates with one random sequence addition per replicate and the TBR and MULTREES options in effect (DeBry and Olmstead, 2000). The bootstrap protocol was modified to save only 20 trees per replicate due to computer memory limitations. The consistency index excluding uninformative characters (CI; Kluge and Farris, 1969) and the retention index (RI; Farris, 1989) were also calculated. For Bayesian inferences, indel characters were excluded. Each region (PRK exon, PRK intron, RPB2 exon, RPB2 intron) was run separately through the program Modeltest v.3.06 (Posada and Crandall, 1998) to identify the appropriate models and parameters. Models were chosen based on the Akaike information criterion as explained in Posada and Crandall (1998). We used the following models for Bayesian analyses with fixed substitution rates and percentage of invariant sites: TrNef+I for PRK exon, HKY+G for PRK intron, K81 for RPB2 exon, TVM+G for RPB2 intron. Bayesian inferences were conducted using the program MrBayes 3.1.2 (Huelsenbeck and Ronquist, 2001) with two Markov chain Monte Carlo (MCMC) runs of four linked chains for 1 000 000 generations, sampling every 100 generations. The four chains included one cold, and the other three with incremental heating as per the default of MrBayes. Of the 10 000 trees produced per both MCMC runs, the first 10% of them were removed as burn-in, resulting in a total of 18 000 trees. The burn-in of 10% was determined to be adequate from the likelihood values, as the values leveled off by 5000 generations for the data sets (0.5% of the generations). All Bayesian analyses produced split frequencies of less than 0.01, showing convergence between the paired runs. We then used PAUP* to compute the majority rule consensus tree for the total data to calculate the posterior probabilities. Biogeographical reconstruction—For ancestral area reconstructions, the members of the Ptychospermatinae were scored according to the following areas: (1) Australia, (2) Fiji, (3) Micronesia (and Palau), (4) New Guinea, 1719 (5) Philippines (6) Samoa, (7) Solomon Islands, and (8) Vanuatu. The remaining taxa were scored for the following areas: (9) Indonesia, (10) Madagascar, (11) New Caledonia, and (12) the Seychelles. The analyses followed the dispersal– vicariance procedures of Ronquist (1997). To incorporate all 18 000 trees produced by the Bayesian analyses of the PRK-RBP2 combined data set, we implemented the program S-DIVA (Yu et al., 2010), which utilizes the program DIVA 1.1 (Ronquist, 1996). The Bayesian posterior tree distribution was favored over the most parsimonious tree set because of the direct relationship between clade frequency and clade support (Nylander et al., 2008). S-DIVA was run with default settings with no more than five areas for each node. RESULTS PRK analyses—The primers yielded an amplicon of ca. 500 bp. The final PRK aligned sequence data matrix contained 669 nucleotide positions and 30 coded gaps (available at website http://www.fairchildgarden.org/aboutfairchild/staffbios/JavierFrancisco-Ortega-PhD2/). There were 69 parsimony-informative characters, 57 of them were nucleotide sites, and 12 were indel positions. The MP analyses yielded 30 trees of 270 steps each (CI = 0.634; RI = 0.862). One of the most parsimonious trees is shown in Fig. 2; branches that collapsed in the strict consensus tree are indicated by solid squares. The PRK trees resolved a monophyletic Ptychospermatinae in both MP and Bayesian analyses (Fig. 2). Under the former, the bootstrap support for the Ptychospermatinae is 76%. Clade credibility after the Bayesian analysis is 99%. Within the subfamily, the major clades are largely identical under both phylogenetic analyses. The greatest conflict between the Bayesian and MP trees is the genus Ptychosperma. In both analyses, Ptychosperma is resolved as monophyletic with strong support. Moreover, in the Bayesian tree, Ptychosperma has within it two well-supported clades, one with unresolved polytomy of six species, the other with an unresolved trichotomy of three species. In contrast, in the MP tree, there is no resolution within Ptychosperma. The clades shown in the MP tree in Fig. 2 collapse in the consensus tree. RPB2 analyses— The primers yielded an amplicon of ca. 700 bp. The final RPB2 aligned data matrix was 971 DNA characters long and included 16 coded gaps (available at http://www. fairchildgarden.org/aboutfairchild/staffbios/Javier-FranciscoOrtega-PhD2/). Six of these gaps and 48 of the nucleotide sites were parsimony informative. The MP search yielded 126 132 trees of 337 steps each (CI = 0.630; RI = 0.848). One of the most parsimonious trees is shown in Fig. 3; branches that collapsed in the strict consensus tree are indicated by solid squares. Note that Brassiophoenix schumannii was unavailable for analysis of RPB2 and is absent from these results. These phylogenetic reconstructions identified a monophyletic Ptychospermatinae in the Bayesian analysis (clade confidence of 86%), but the node collapsed in the MP analysis. The major clades identified by the Bayesian analysis of RPB2 are roughly equivalent to those identified by PRK; areas of conflict will be discussed later. Combined analysis— The congruence among data sets was assessed by examining the phylogenetic results and identifying well-supported (bootstrap > 85%) conflicting relationships. The incongruence length difference test (Farris et al., 1994, 1995) was not used because its results have been shown to be misleading (Dolphin et al., 2000; Reeves et al., 2001; Yoder et al., 2001; Barker and Lutzoni, 2002; Darlu and Lecointre, 2002). 1720 American Journal of Botany [Vol. 98 Fig. 2. Phylogenetic trees of Ptychospermatinae based on nucleotide sequences of PRK. Left: one of the 30 shortest trees (tree length = 270 steps, CI = 0.634, RI = 0.862) resulting from the parsimony analysis. Nodes that collapse in the strict consensus tree are indicated by solid boxes. Numbers above branches are the bootstrap support for the clades. Numbers below branches are branch lengths. Right: Majority rule consensus tree of the Bayesian inference analysis. Bayesian posterior probability support values are shown above branches. Both the PRK analysis and the RBP2 analysis recover trees with similar topologies for a large number of taxa. For 27 terminal taxa (comprising Veitchia and Drymophloeus p.p., Balaka and Solfia, Ptychosperma, and Ponapea) the trees are fully compatible. In the combined analysis, these taxa were grouped into four well-supported clades. For the remaining 10 taxa (nine in RPB2, because data for Brassiophoenix schumannii are missing), the RPB2 and PRK trees have conflicting topologies. Two taxa, both species of Adonidia, formed a well-supported clade in the PRK and combined analyses, but did not form a clade in the RPB2 analysis. The remaining seven taxa were grouped in a single, weakly supported clade in the PRK and combined analyses but were resolved on four separate clades in the RPB2 analysis. The six major clades are named in Fig. 4. Among the major clades, additional areas of conflict between the PRK and RPB2 analyses are in the relationships among the six major clades. In the PRK analysis, the one species of Veitchia (along with two species of Drymophloeus) is sister to a October 2011] Zona et al.—Phylogeny of the Ptychospermatinae (Arecaceae) 1721 Fig. 3. Phylogenetic trees of Ptychospermatinae based on nucleotide sequences of RPB2. Left: one of the 126 132 shortest trees (tree length = 337 steps, CI = 0.630, RI = 0.848) resulting from the parsimony analysis. Nodes that collapse in the strict consensus tree are indicated by solid boxes. Numbers above branches are the bootstrap support for the clades. Numbers below branches are branch lengths. Right: Majority rule consensus tree of the Bayesian inference analysis. Bayesian posterior probability support values are shown above branches. clade of Balaka and Solfia; whereas in the RPB2 analysis, it is sister to the species of Ponapea (and one Drymophloeus). Another area of conflict is in the genus Drymophloeus. In the PRK tree, D. oliviformis and D. litigiosus are sister species, but in the RPB2 tree, D. litigiosus is sister to Normanbya normanbyi. A third area of conflict is the genus Ptychosperma, which is sister to Ponampea, Normanby, and Drymophloeus p.p. in the PRK tree but has no resolved sister clades in the RBP2 tree. Finally, Wodyetia is sister to Carpentaria on the PRK tree on a clade that is the first branch from the clade that includes Veitchia, Drymophloeus p.p., Balaka, Solfia, Brassiophoenix, and Adonidia. On the RPB2 tree, Wodyetia is part of an unresolved 1722 American Journal of Botany [Vol. 98 Fig. 4. Phylogenetic trees of Ptychospermatinae based on simultaneous analysis of nucleotide sequences of PRK and RPB2. Left: one of the 2448 shortest trees (tree length = 623 steps, CI = 0.591, RI = 0.828) resulting from the parsimony analysis. Nodes that collapse in the strict consensus tree are indicated by solid boxes. Numbers above branches are the bootstrap support for the clades. Numbers below branches are branch lengths. The six major clades recognized within the Ptychospermatinae are indicated. Right: Majority rule consensus tree of the Bayesian inference analysis. Bayesian posterior probability support values are shown above branches. polytomy that includes Drymophloeus p.p., Ptychococcus, Brassiophoenix, Carpentaria, and a clade containing D. litigiosus and Normanbya. In spite of this incongruence, we analyzed the PRK-RPB2 combined data in order to include the largest sample of informative characters and to compare the combined with the separate PRK and RBB2 data sets (Wiens, 1998). Maximum parsimony analysis of combined PRK and RPB2 sequences resulted in 2448 equally parsimonious trees. The trees were 623 steps long (CI = 0.591, RI = 0.828). The resulting trees from the combined PRK and RBP2 analyses are shown in Fig. 4. The subtribe Ptychospermatinae is identified as monophyletic in both MP and Bayesian analyses, with a bootstrap support of 89% and clade confidence of 100%. Both analytical methods produced similar trees with identical major lineages. The major clades are labeled in the MP tree in Fig. 4. They are the Veitchia clade (bootstrap = 100%), the Balaka clade (bootstrap = 98%), the Adonidia clade (bootstrap = 71%), the Drymophleous clade (bootstrap = 57%), the Ptychosperma clade (bootstrap = 97%), and the Ponapea clade (bootstrap = 100%). The corresponding clades in the Bayesian tree each have clade confidences of 100 with the exception of the Adonidia clade, which has a clade confidence of 74. October 2011] Zona et al.—Phylogeny of the Ptychospermatinae (Arecaceae) Biogeographical analysis— For the subtribe Ptychospermatinae, the S-DIVA analysis strongly favored New Guinea (95%) as the ancestral area (AA) (see Table 2). The remaining 5% constituted dozens of combinations of areas, all including New Guinea, each with less than 0.2% support. Within the subtribe, the Adonidia clade had the combined AA of New Guinea+the Philippines (100%) for all topologies where the two species were sister-taxa. The AA of the Balaka clade was statistically split between Samoa (50%) and the combined AA of Fiji+Samoa (50%). The Drymophloeus clade had three AAs of varied support with New Guinea (64%), New Guinea+Australia (33%), and Australia (3%). The Ponapea clade also had three AAs with New Guinea+Micronesia (43%), New Guinea+Australia (29%), and New Guinea+Micronesia+Australia (29%). The Ptychosperma clade had the highest support (100%) with the AA of New Guinea. The primary AA for the Veitchia clade was Fiji (89%) with other AAs of significantly less support with Fiji+Vanuatu (8%), Vanuatu (1%), and multiple combined AAs with the Solomon Islands (all less than 0.1%). DISCUSSION Phylogenetic reconstruction—In the Bayesian and MP analyses, both two-gene and single-gene, the subtribe Ptychospermatinae is resolved as monophyletic with good statistical support, with the exception of the MP analysis of the RBP2 data set. Our results lend additional support to the classification of this distinctive group of related genera in a single subtribe (Dransfield et al., 2008). Our work confirms and validates the conclusions of Asmussen et al. (2006), Norup et al. (2006), Baker et al. (2009) and Baker et al. (in press), based on molecular evidence, that the subtribe Ptychospermatinae is monophyletic. Both MP and Bayesian analyses of the combined data set identify six major clades (Fig. 4): the Veitchia clade (including two species currently residing in Drymophloeus), the Balaka clade, the Adonidia clade, the Drymophloeus clade, the Ptychosperma clade, and the Ponapea clade (which includes one species currently classified in Drymophloeus). With two exceptions, each of these major clades has distinctive morphological characteristics that unite the taxa. The exceptions are the Drymophloeus clade, which has only weak support, and the Ponapea clade, which although strongly supported, is morphologically heterogeneous. The major clades, their component genera and their morphological characteristics will be discussed below, with reference to the combined analysis trees (Fig. 4). The Veitchia clade—Veitchia, a relatively large genus within the subtribe, is resolved as paraphyletic with respect to Drymophloeus subdistichus and D. pachycladus, both formerly of the genus Rehderophoenix Burret. This Veitchia clade has bootstrap support of 100%, a result that, when taken with the similarity in endocarp morphology among the species (Zona, 1999a), supports the inclusion of these species of Drymophloeus within the genus Veitchia. We did not sample a third member of Drymophloeus, D. lepidotus, a serpentine endemic, but its geographic range and morphological characteristics (Zona and Fuller, 1999) indicate that it too should be included in Veitchia. The relationship between these species of Drymophloeus and Veichia was not previously suspected (Uhl and Dransfield, 1987; Zona, 1999a, b; Zona and Fuller, 1999). All these taxa have straw-colored endocarps with a single, flattened ridge. Thus reconstituted, Veitchia becomes a monophyletic genus of 1723 Table 2. Results from the S-Diva ancestral area analysis. Where multiple ancestral areas are proposed, they are ordered by decreasing likelihood (shown in parentheses). Major clade (no. taxa in clade) Clade credibility (%) Ancestral area(s) (statistical likelihood) Adonidia (2) Balaka (3) Drymophloeus (7) 74 100 100 Ponapea (4) 100 Ptychosperma (12) Veitchia (8) 100 100 Subtribe (36) 100 New Guinea+Philippines (100%) Fiji+Samoa (50%) or Samoa (50%) New Guinea (64%); New Guinea+Australia (33%); Australia (3%) New Guinea+Micronesia (43%); New Guinea+Australia (29%) or New Guinea+Micronesia+Australia (29%) New Guinea (100%) Fiji (89%); Fiji+Vanuatu (9%); Vanuatu (1%); combinations with the Solomon Islands all less than 0.1% New Guinea (95%); many combinations with New Guinea, all less than 0.2% medium to large palms from Fiji, Tonga, Vanuatu, and the Solomon Islands. Veitchia filifera of Fiji is resolved as the sister species to the remainder of the clade. The Balaka clade—Until recently (Zona, 1999a), the monotypic genus Solfia had been included within Drymophloeus. In this analysis, Balaka and Solfia form a clade with 98% bootstrap support that is sister to the Veitchia clade. In the RPB2 analysis, Solfia was embedded with Balaka; however, in PRK analysis, Solfia is sister to a monophyletic Balaka. Further analyses of these two genera, incorporating all species of Balaka, is indicated and may resolve this ambiguity. The brown, angular, pointed endocarps of Balaka are strikingly different from the straw-colored, terete, rounded endocarps of Solfia, and this difference was used by Zona (1999a) and subsequent authors to justify a separate generic status. Our findings do not necessitate any change in generic limits for these taxa. The Adonidia clade—Adonidia merrillii had long been included in Veitchia by Moore (1957) and subsequent authors until reinstated by Zona (1999a). Our results support the recognition of Adonidia as a separate genus, placing it as sister to the Balaka and Veitchia clades. The new species from Biak is shown to be sister to A. merrillii (BS = 71%; clade confidence = 74). The vegetative morphology of the Biak palm is reminiscent of Drymophloeus, while the infructescence and fruit morphology suggests Adonidia. Rather than proposing another monotypic genus, we support the inclusion of the Biak palm within an expanded Adonidia. This new species will be formally described in a forthcoming publication (W. J. Baker and C. Heatubun [Universitas Negeri Papua], unpublished manuscript) The Drymophloeus clade—Although this group of taxa is monophyletic (BS = 57%, clade confidence = 100), the branches in the Drymophloeus clade collapse in the consensus tree, yielding an unresolved polytomy. The only surviving clades are the Wodyetia and Carpentaria clade and the Ptychococcus and Brassiophoenix clade. Carpentaria and Wodyetia form a single clade with BS of 76%. A relationship between these two genera had been suggested by Zona (1999a) on the basis of black endocarps, although the other black endocarp taxa, Ponapea ledermanniana and Ptychococcus paradoxus, are not part of the Carpentaria and Wodyetia clade. Prior to Zona’s (1999a) results, Wodyetia had been associated with Normanbya by virtue of 1724 American Journal of Botany their plumose leaves (Irvine, 1983), and Carpentaria was thought to be close to Veitchia (Uhl and Dransfield, 1987; Dowe, 1991), because both genera share an unspecialized morphology. The relationship between Carpentaria and Wodyetia is satisfying in terms of biogeography: both are endemic to northern Australia. The relationship between Brassiophoenix drymophloeoides and Ptychococcus paradoxus is not unexpected. Although Zona (1999a) believed the two genera were not close, based on their different endocarp colors, a relationship based on endocarp morphology had been suggested by Uhl and Dransfield (1987). Previous molecular analyses (Asmussen et al., 2006; Norup et al., 2006) also found a sister relationship between these two genera. Ptychococcus and Brassiophoenix share the morphological characteristic of elaborately sculptured endocarps, and both are from New Guinea. The remaining three taxa of the Drymophloeus clade are Normanbya normanbyi, Drymophloeus oliviformis, and D. litigiosus. Their relationships are unresolved in the MP tree, but the Bayesian tree shows a well-supported relationship between Normanbya and D. litigiosus. This relationship was not previously suspected (Irvine, 1983; Zona, 1999a, b). Normanbya is monotypic, occurs in northern Australia, and has little in common morphologically with Drymophloeus. Drymophloeus litigiosus is native to Indonesian New Guinea and the Moluccas and is thought to be close to Drymophloeus oliviformis (Zona, 1999b). Both Drymophloeus litigiosus and D. oliviformis share the understory habit, stilt roots, and a persistent, marcescent peduncular bract (Zona, 1999b). The two species of Drymophloeus are very similar in overall morphology, and in fact, one can easily be mistaken for the other. They were resolved as sister species in the PRK parsimony analysis. We are unable to explain their disparate positions in both the MP and Bayesian trees and the possible relationship with Normanbya. Our analysis cannot confirm that Drymophleous s.s., of which D. oliviformis is the type, is monophyletic. (For D. hentyi, see the Ponapea clade, below.) The Ptychosperma clade—Ptychosperma, the largest genus of subtribe Ptychospermatinae, is resolved as monophyletic with strong support (BS = 97%; clade confidence = 100). The notable exception is the species of Ponapea that were previously included in Ptychosperma by Essig (1978) and which, based on preliminary analyses of this data set, were now restored to Ponapea by Dransfield et al. (2008). Resolution within the genus Ptychosperma is poor, but some of the infrageneric groups defined by Essig (1978) were recovered. Ptychosperma caryotoides, P. elegans, and P. salomonense, all members of Essig’s subgenus Ptychosperma, are resolved as comprising monophyletic group with moderate support (BS = 74%; clade confidence = 100). These palms have solitary stems, red fruits and rounded lobes on their seeds (as seen in cross section). The remaining palms in our sample belong to Essig’s subgenus Actinophloeus (formerly the genus Actinophloeus Becc.), but they do not form a monophyletic group in our analyses. Ptychosperma pullenii and P. cuneatum belong to section Actinophloeus, according to Essig (1978). In our analysis, they were resolved as sister species with good support (BS = 81; clade confidence = 100). These species share the morphological traits of cuneate leaflets, black fruits, and homogeneous endosperm. However, P. burretianum was also placed in section Actinophloeus by Essig (1978) but is sister to P. macarthurii of section Caespitosa in our analyses. Ptychosperma lineare, P. propinquum, P. lauterbachii, and P. sanderianum are all [Vol. 98 members of section Caespitosa. Ptychosperma microcarpum, also a member of subgenus Actinoploeus section Caespitosa according to Essig, was not placed with the other members of that section in our analyses. The Ponapea clade—This analysis resolved four taxa, Drymophloeus hentyi, Ponapea hosinoi, Ponapea ledermanniana, and Ponapea palauense, as a single clade with 100% bootstrap and clade confidence support. The genus Ponapea Becc. was first recognized by Beccari in 1924 but sunk into Ptychosperma by Moore (1973) and subsequent authors until revived again by Dransfield et al. (2008). Our results confirm that the genus Ponapea should be reinstated and, we propose, expanded to accommodate D. hentyi. This latter species was first described as a Ptychosperma and then transferred to Drymophloeus. The strong evidence from both the PRK and RBP2 data sets indicates that it is phylogentically close to Ponapea rather than these other genera. Our results could be used to argue for recognition of this species as a monotypic genus or as a member of Ponapea. We favor the latter, more inclusive solution to avoid taxonomic inflation. Nevertheless, the expanded Ponapea is a troublesome genus with no obvious morphological distinction or synapomorphy. The pistillodes of P. hosinoi and P. ledermanniana are short and conical. Those of D. hentyi are columnar, and the pistillodes of P. palauensis are lageniform (bottle-shaped), as they are in other Ptychospermatinae. This degree of pistilode diversity is not seen elsewhere in the subtribe. Likewise, they are diverse in endocarp color and shape, ranging from black and angled to straw-colored and terete. Further study may shed light on character state evolution within this intriguing but poorly known genus. Biogeography— Most members of the Ptychospermatinae occur on islands, and most of these islands are volcanic in nature (Neall and Trewick, 2008). With the exceptions of Australia–New Guinea and Palawan–Borneo, no land bridges connected these islands (Hope, 1996; Voris, 2000). Therefore, overwater dispersal appears to be an important biogeographical avenue to account for the distribution patterns of the tribe among the Pacific islands. The biogeographical analysis points to New Guinea as the ancestral area of the subtribe, and indeed, that island is presently the most species-rich land mass and home to a significant radiation of species. New Guinea is also identified as the most likely ancestral area for the Ptychosperma clade (Ptychosperma s.s.) and the Drymophloeus clade of six genera. The Australian genera, Wodyetia, Carpentaria, and Normanbya, along with the Australian species of Ptychosperma, do not form a monophyletic group, so we propose at least three radiations into Australia to account for these taxa. The Ponapea clade has as its ancestral area in New Guinea+Micronesia. If we assume a New Guinea origin for the subtribe, a single dispersal event and subsequent dispersal and radiation within Micronesia could account for the presence of Ponapea in Micronesia. The results are equivocal for the Balaka clade (Balaka and Solfia). It may have its ancestral area in Fiji+Samoa, but the ancestral area of Samoa alone is just as likely. Dispersal between Fiji and Samoa must have occurred, because even during the Pleistocene Last Glacial Maximum, these landmasses were never contiguous (Neall and Trewick, 2008). Fiji is supported as the ancestral area for Veitchia, with subsequent overwater spread to Vanuatu and the Solomon Islands, both volcanic archipelagos. October 2011] Zona et al.—Phylogeny of the Ptychospermatinae (Arecaceae) The presence of Adonidia in Palawan and Borneo, west of Wallace’s Line, has been a biogeographic puzzle to botanists for some time (Dransfield, 1981, 1987; Baker and Couvreur, in press). The discovery of a second species of Adonidia from Biak Island (offshore northern New Guinea) deepens the mystery. Of all the islands of the Philippines, Palawan is westernmost and has strong floristic and faunistic relationships with nearby Borneo (Atkins et al., 2001; Jones and Kennedy, 2008), so the recent discovery of Adonidia in Sabah, Borneo, is not surprising. The relationship between New Guinea and the Philippines has been linked to a possible island stepping-stone migration route via the Philippine-Halmahera arc that provided dispersal opportunities from the late Oligocene through the Pleistocene (Hall, 1998, 2009). Because the two species of Adonidia comprise the Adonidia clade, the S-DIVA analysis forces a conclusion of New Guinea+Philippines as the ancestral area. Given the strong development of the Ptychospermatinae in New Guinea and the lack of additional species in the Philippines, we suggest a New Guinea origin and subsequent dispersal as the explanation for the presence of Adonidia west of Wallace’s Line. New taxonomic combinations— As a result of this study, the following new combinations are proposed: Veitchia lepidota (H. E. Moore) Lewis & Zona, comb. nov. Basionym: Drymophloeus lepidotus H. E. Moore, Principes 13: 75, 76. 1969. Veitchia pachyclada (Burret) Lewis & Zona, comb. nov. Basionym: Rehderophoenix pachyclada Burret, Notizbl. Bot. Gard. Berlin-Dahlem 13: 87. 1936. Synonym: Drymophloeus pachycladus (Burret) H. E. Moore, Principes 13: 76. 1969. Veitchia subdisticha (H. E. Moore) Lewis & Zona, comb. nov. Basionym: Rehderophoenix subdisticha H. E. Moore, Principes 10: 93. 1966. Synonym: Drymophloeus subdistichus (H. E. Moore) H. E. Moore, Principes 13: 76. 1969. Ponapea hentyi (Essig) Lewis & Zona, comb. nov. Basionym: Ptychosperma hentyi Essig, Principes 31: 113. 1987. Synonym: Drymophloeus hentyi (Essig) Zona, Blumea 44: 13. 1999. LITERATURE CITED Asmussen, C. B., and M. W. Chase. 2001. Coding and noncoding plastid DNA in palm systematics. American Journal of Botany 88: 1103–1117. Asmussen, C. B., J. Dransfield, V. Deickmann, A. S. Barfod, J.-C. Pintaud, and W. J. Baker. 2006. A new subfamily classification of the palm family (Arecaceae): Evidence from plastid DNA phylogeny. Botanical Journal of the Linnean Society 151: 15–38. Atkins, H., J. Preston, and Q. C. B. Cronk. 2001. A molecular test of Huxley’s line: Cyrtandra (Gesneriaceae) in Borneo and the Philippines. Biological Journal of the Linnean Society 72: 143–159. Baker, W. J., C. B. Asmussen, S. C. Barrow, J. Dransfield, and T. A. Hedderson. 1999. A phylogenetic study of the palm family (Palmae) based on chloroplast DNA sequences form the trnL-trnF region. Plant Systematics and Evolution 219: 111–126. Baker, W. J., and T. L. P. Couvreur. In press. Biogeography and distribution patterns of Southeast Asian palms. In D. Gower, K. Johnson, J. E. Richardson, B. Rosen, L. Rüber, and S. Williams [eds.], Biotic evolution and environmental change in Southeast Asia. Cambridge University Press, Cambridge, UK. Baker, W. J., M. V. Norup, J. J. Clarkson, T. L. P. Couvreur, J. L. Dowe, C. E. Lewis, J.-C. Pintaud, et al. In press. Phylogenetic relationships among arecoid palms (Arecaceae: Arecoideae). Annals of Botany doi:10.1093/aob/mcr020. 1725 Baker, W. J., V. Savolainen, C. B. Asmussen-Lange, M. W. Chase, J. Dransfield, F. Forest, M. M. Harley, et al. 2009. Complete generic-level phylogenetic analyses of palms (Arecaceae) with comparisons of supertree and supermatrix approaches. Systematic Biology 58: 240–256. Barker, F. K., and F. M. Lutzoni. 2002. The utility of the incongruence length difference test. Systematic Biology 51: 625–637. Darlu, P., and G. Lecointre. 2002. When does the incongruence length difference test fail? Molecular Biology and Evolution 19: 432–437. DeBry, R. W., and R. G. Olmstead. 2000. A simulation study of reduced tree-search effort in bootstrap resampling analysis. Systematic Biology 49: 171–179. Denton, A. L., B. L. McConaughy, and B. D. Hall. 1998. Usefulness of RNA polymerase II coding sequences for estimation of green plant phylogeny. Molecular Biology and Evolution 15: 1082–1085. Dolphin, K., R. Belshaw, C. D. L. Orme, and D. L. J. Quicke. 2000. Noise and incongruence: Interpreting results of the incongruence length difference test. Molecular Phylogenetics and Evolution 17: 401–406. Dransfield, J. 1981. Palms and Wallace’s Line. In T. C. Whitmore [ed.], Wallace’s Line and plate tectonics, 43–56. Clarendon Press, Oxford, UK. Dransfield, J. 1987. Bicentric distribution in Malesia as exemplified by palms. In T. C. Whitmore [ed.], Biogeographical evolution of the Malay Archipelago, 60–72. Oxford, UK. Dransfield, J., N. W. Uhl, C. B. Asmussen, W. J. Baker, M. M. Harley, and C. E. Lewis. 2008. Genera palmarum: The evolution and classification of palms. Royal Botanical Gardens, Kew, UK. Dowe, J. 1991. The palms of Vanuatu and Fiji. Notes on distribution, classification, and taxonomy. Mooreana 1: 13–20. Essig, F. B. 1978. A revision of the genus Ptychosperma Labill. (Arecaceae). Allertonia 1: 415–478. Farris, J. S. 1989. The retention index and homoplasy excess. Systematic Zoology 38: 406, 407. Farris, J. S., M. Källersjö, A. G. Kluge, and C. Bult. 1994. Testing significance of incongruence. Cladistics 10: 315–319. Farris, J. S., M. Källersjö, A. G. Kluge, and C. Bult. 1995. Constructing a significance test for incongruence. Systematic Biology 44: 570–572. Felsenstein, J. 1985. Confidence limits on phylogenies: An approach using bootstrap. Evolution 39: 783–791. Fernando, E. S. 2011. Adonidia merrillii—A new locality record for a wild population in the Philippines. Palms 55: 57–61. Fitch, W. M. 1971. Toward defining the course of evolution: Minimum change for a specific topology. Systematic Zoology 20: 406–416. Hahn, W. J. 2002. A molecular phylogenetic study of the Palmae (Arecaceae) based on atpB, rbcL, and 18S nrDNA sequences. Systematic Biology 51: 92–112. Hall, R. 1998. The plate tectonics of Cenozoic SE Asia and the distribution of land and sea. In R. Hall and J. D. Holloway, Biogeography and geological evolution of SE Asia, 99–131. Backbuys Publishers, Leiden, Netherlands. Hall, R. 2009. Southeast Asia’s changing palaeogeography. Blumea 54: 148–161. Hodel, D. R. 2010. Synopsis of Balaka. Palms 54: 161–188. Hooker, J. D. 1883. Palmae. In G. Bentham and J. D. Hooker. Genera plantarum, vol. 3, 870–948. L. Reeve & Co., London, UK. Hope, G. 1996. Quaternary change and the historical biogeography of Pacific Islands. In A. Keast and S. E. Miller [eds.], The origin and evolution of Pacific island biotas, New Guinea to eastern Polynesia: patterns and processes, 165–190. SPB Academic Publishing, Amsterdam, Netherlands Huelsenbeck, J. P., and F. Ronquist. 2001. MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics17: 754–755. Irvine, A. 1983. Wodyetia, a new Arecoid genus from Australia. Principes 27: 158–167. Jones, A. W., and R. S. Kennedy. 2008. Evolution in a tropical archipelago: Comparative phylogeography of Philippine fauna and flora reveals complex patterns of colonization and diversification. Biological Journal of the Linnean Society 95: 620–639. 1726 American Journal of Botany Kluge, A. G., and J. S. Farris. 1969. Quantitative phyletics and the evolution of anurans. Systematic Zoology 18: 1–32. Lewis, C. E., and J. J. Doyle. 2002. A phylogenetic analysis of tribe Areceae (Arecaceae) using two low-copy nuclear genes. Plant Systematics and Evolution 236: 1–17. Loo, A. H. B., J. Dransfield, M. W. Chase, and W. J. Baker. 2006. Lowcopy nuclear DNA, phylogeny and the evolution of dichogamy in the betel nut palms and their relatives (Arecinae; Arecaceae). Molecular Phylogenetics and Evolution 39: 598–618. Moore, H. E. Jr. 1957. Veitchia. Gentes Herbarum 8: 483–536. Moore, H. E. Jr. 1973. The major groups of palms and their distribution. Gentes Herbarum 11: 27–141. Neall, V. E., and S. A. Trewick. 2008. The age and origin of the Pacific islands: A geological overview. Philosophical Transactions of the Royal Society, B, Biological Scoences 363: 3293–3308. Norup, M. V., J. Dransfield, M. W. Chase, A. S. Barfod, E. S. Fernando, and W. J. Baker. 2006. Homoplasious character combinations and generic delimitation: A case study from the Indo-Pacific Arecoid palms (Arecaceae: Areceae). American Journal of Botany 93: 1065–1080. Nylander, J. A. A., U. Olsson, P. Alström, and I. Sanmartín. 2008. Accounting for phylogenetic uncertainty in biogeography: A Bayesian approach to dispersal–vicariance analysis of the thrushes (Aves: Turdus). Systematic Biology 57: 257–268. Posada, D., and K. A. Crandall. 1998. Modeltest: Testing the model of DNA substitution. Bioinformatics 14: 817–818. Rannala, B., and Z. Yang. 1996. Probability distribution of molecular evolutionary trees: A new method of phylogenetic inference. Journal of Molecular Evolution 43: 304–311. Reeves, G., M. W. Chase, P. Goldblatt, P. Rudall, M. W. Fay, A. V. Cox, B. Lejeune, and T. Souza-Chies. 2001. Molecular systematics of Iridaceae: Evidence from four plastid DNA regions. American Journal of Botany 88: 2074–2087. Roncal, J., J. Francisco-Ortega, C. B. Asmussen, and C. E. Lewis. 2005. Molecular phylogenetics of tribe Geonomeae (Arecaceae) using nuclear DNA sequences of phosphoribulokinase and RNA polymerase II. Systematic Botany 30: 275–283. Ronquist, F. 1996. DIVA version 1.1. Computer program and manual available by anonymous FTP from Uppsala University. Uppsala, Sweden [ftp. uu.se or ftp.systbot.uu.se]. Ronquist, F. 1997. Dispersal–vicariance analysis: A new approach to the quantification of historical biogeography. Systematic Biology 46: 195–203. Simmons, M. P., and H. Ochoterena. 2000. Gaps as characters in sequence-based phylogenetic analyses. Systematic Biology 49: 369–381. Swofford, D. L. 2002. PAUP*: Phylogenetic analysis using parsimony (*and other methods), version 4.0b10. Sinauer, Sunderland, Massachusetts, USA. Thiers, B. 2011. Index herbariorum: A global directory of public herbaria and associated staff. New York Botanical Garden’s Virtual Herbarium. Website http://sweetgum.nybg.org/ih/. Thomas, M. M., N. C. Garwood, W. J. Baker, S. A. Henderson, S. J. Russell, D. R. Hodel, and R. M. Bateman. 2006. Molecular phylogeny of the palm genus Chamaedorea, based on the low-copy nuclear genes PRK and RPB2. Molecular Phylogenetics and Evolution 38: 398–415. Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The CLUSTAL_X windows interface: Flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 25: 4876–4882. Uhl, N. W., and J. Dransfield. 1987. Genera palmarum, a classification of palms based on the work of H. E. Moore Jr. Lawrence, Kansas, USA. Voris, H. K. 2000. Maps of Pleistocene sea levels in Southeast Asia: Shorelines, river systems and time durations. Journal of Biogeography 27: 1153–1167. Wiens, J. J. 1998. Combining data sets with different phylogenetic histories. Systematic Biology 47: 315–324. Yoder, A. D., J. A. Irwin, and B. A. Payseur. 2001. Failure of the ILD to determine data combinability for slow loris phylogeny. Systematic Biology 50: 408–424. Yu, Y., A. J. Harris, and X. J. He. 2010. S-DIVA (Statistical Dispersal– Vicariance Analysis): A tool for inferring biogeographic histories. Molecular Phylogenetics and Evolution 56: 848–850. Zona, S. 1999a. New perspectives on generic limits and relationships in the Ptychospermatinae (Palmae: Arecoideae). Memoirs of the New York Botanical Garden 83: 255–263. Zona, S. 1999b. Revision of Drymophloeus (Arecaceae: Arecoideae). Blumea 44: 1–24. Zona, S. 2005. A revision of Ptychococcus (Arecaceae). Systematic Botany 30: 520–529. Zona, S., and F. B. Essig. 1999. How many species of Brassiophoenix? Palms 43: 45–48. Zona, S., and D. Fuller. 1999. A revision of Veitchia (Arecaceae– Arecoideae). Harvard Papers in Botany 4: 543–560.