INTRODUCTION The chiropteran family Pteropodidae comprises 42 currently recognized genera and 186 species, of which as many as 65 species are classified in the genus Pteropus Brisson, 1762 (Simmons, 2005). Distributed from islands off the East coast of Afri- ca to Polynesia in the Western Pacific, the majority of Pteropus species are Indian Ocean and Pacific island endemics. Andersen (1912) established the membership of the genus and clearly distinguished Pteropus species from other closely allied genera known at the time ? Acerodon Jourdan, 1837, Pte - ra lopex Thomas, 1888, and Styloctenium Matschie, 1899. Although subsequent authors have generally followed Andersen?s (1912) generic arrangement of Pteropus, previous authors sometimes allocated a few species to separate genera, particularly Mil- ler (1907), who erected the genus Desmalopex to accommodate a single distinctive species, Ptero- pus leu copterus Temminck, 1853, endemic to the Philippines (see description in Miller, 1907 and Andersen, 1909, 1912). Miller (1907: 60) high light - ed affinities between Des ma lopex and Pteropus but listed a number of characters that alternatively ?dis- tinctly suggest Pteralopex.? Andersen (1909, 1912: 294) included leucopterus in Pteropus and associat- ed this form with members of the pselaphon species group, which in his view ?shows decidedly leanings towards the highly specialized genus Pteralopex.? Thus while both Miller (1907) and Andersen (1912) recognized the distinctive morphological at- tributes of leucopterus, their taxonomic arrange- ments differed markedly. Their mutually exclusive systematic hypotheses (leucopterus placed in a mo - no typic genus versus leucopterus included within Pteropus) have remained essentially untested since the beginning of the twentieth century. This fact has gone largely unrecognized because leucopte- rus has been included in Pteropus without com- ment by all recent authors (e.g., Corbet and Hill, 1992; Koopman, 1993, 1994; Heaney et al., 1998; Acta Chiropterologica, 10(1): 11?20, 2008 PL ISSN 1508-1109 ? Museum and Institute of Zoology PAS doi: 10.3161/150811008X331054 The systematic position of Pteropus leucopterus and its bearing on the monophyly and relationships of Pteropus (Chiroptera: Pteropodidae) NORBERTO P. GIANNINI1, 2, 5, FRANCISCA CUNHA ALMEIDA1, 3, NANCY B. SIMMONS1, and KRISTOFER M. HELGEN4 1Department of Mammalogy, American Museum of Natural History, Central Park West at 79th Street, New York, NY 10024-5192, USA 2CONICET, Programa de Investigaciones de Biodiversidad Argentina, Facultad de Ciencias Naturales e Instituto Miguel Lillo, Miguel Lillo 205, Universidad Nacional de Tucum?n, Tucum?n, CP 4000, Argentina 3Division of Invertebrate Zoology, Molecular Systematics Laboratory, American Museum of Natural History, Central Park West at 79th Street, New York, NY 10024, USA 4Division of Mammals, National Museum of Natural History, NHB 390, MRC 108 Smithsonian Institution, P.O. Box 37012, Washington, D.C. 20013-7012 USA 5Corresponding author: E-mail: norberto@amnh.org Pteropus is the most speciose genus in Pteropodidae, currently comprising 65 species in 18 species groups. Here we examine whether Pteropus as currently understood is monophyletic. We sequenced three nuclear genes (RAG-1, RAG-2 and vWF) totalling c. 3.0 kbp from 18 species of Pteropus representing 12 species groups, plus Acerodon celebensis and megachiropteran outgroups representing all other subfamilies and tribes. Separate and combined parsimony and maximum likelihood analyses recovered a clade containing Acerodon as sister to all Pteropus species to the exclusion of the Philippine endemic taxon ?P. leucopterus?, rendering Pteropus paraphyletic. We propose the revalidation of Desmalopex Miller, 1907, an available generic name for leucopterus, adopting the name combination Desmalopex leucopterus (Temminck, 1853). We discuss implications of this result and anticipate further modifications of the classification of Pteropus. Key words: Philippines, Desmalopex leucopterus, Pteropus, Megachiroptera, phylogeny Sim mons, 2005). Here we begin analyzing the com- plex problem of the composition and relationships of Pte ropus by focusing on the relationship of Pteropus leucopterus to other members of this genus from different species groups, as well as to other pteropodine megabats. Based on parsimony and max imum likelihood analyses of DNA sequences from three nuclear coding genes, we provide evi- dence bearing on the phylogenetic position of leu- copterus and on pteropodid systematics in general. MATERIALS AND METHODS Taxa For the purposes of this study we chose Pteropodini sensu Bergmans (1997) as the ingroup. We included as outgroups rep- resentatives of each main megachiropteran clade following Giannini and Simmons (2005) and Giannini et al. (2006). We choose the same taxa as in Giannini et al. (2006), specifically: Nyctimene albiventer and N. vizcaccia (Nyctimeninae), Cyno - pterus sphinx and Ptenochirus jagori (Cynopterinae), Eonyc - teris spelaea, Rousettus amplexicaudatus, and R. aegyptiacus (Rousettinae), Myonycteris torquata and Megaloglossus woer- manni (Epomophorinae, Myonycterini), Epomops franqueti and Epomophorus wahlbergi (Epomophorinae, Epomophorini), Harpyionycteris whiteheadi, Dobsonia inermis, and D. magna (Harpyionycterinae), Macroglossus minimus (Macro glossinae), Melonycteris (Nesonycteris) fardoulisi [incertae sedis, variably associated with macroglossines (e.g., Ander sen, 1912; Giannini and Simmons, 2005), pteropodines (e.g., Kirsch et al., 1995; Berg mans, 1997), and other groups], and Eido lon helvum [in- certae sedis, allied to Rousettinae (Andersen, 1912; Bergmans, 1997) and Pteropodini (Giannini and Simmons, 2005)]. We included two microchiropteran bats [Rhi no poma hardwickei (Rhinopomatidae) and Artibeus jama icensis (Phyl lo stomidae)] to provide an unambiguous root to our tree. Ingroup taxa included typical pteropodine megabats avail- able to us, including Acerodon celebensis and 18 species of Pteropus from 12 species groups (sensu Andersen, 1912): Pteropus alecto (alecto species group), P. conspicillatus (con- spicillatus group), P. pelewensis, P. tonganus (mariannus group), P. molossinus (molossinus group), P. neohibernicus (neohibernicus group), P. capistratus (temminckii group), P. po- liocephalus (poliocephalus group), P. leucopterus (pselaphon group), P. anetianus, P. samoensis (samoensis group), P. scapu- latus, P. woodfordi (scapulatus group), P. admiralitatum, P. pu - milus, P. hypomelanus (subniger group), P. giganteus, P. lylei, and P. vampyrus (vampyrus group). Museum specimens referenced herein are deposited in the American Museum of Natural History, New York (AMNH); the Natural History Museum, London (BMNH); the Delaware Museum of Natural History, Wilmington (DMNH); the Field Museum of Natural History, Chicago (FMNH); Museum of Vertebrate Zoology, Berkley (MVZ); the Nationaal Natuur - historisch Museum, Leiden, The Netherlands (RMNH); the Senckenberg Museum, Frankfurt (SMF); and the National Mu - seum of Natural History, Smithsonian Institution, Wash ington, D.C. (USNM). Morphological terminology and usage follow Andersen (1912) and Giannini et al. (2006b). Sequences and Phylogenetic Analysis We used sequences of exon 28 of the von Willebrand Factor gene (vWF, 1231 bp) from Giannini et al. (2006), and generat- ed new vWF sequences for Pteropus species for this study. We also generated sequences of two other nuclear coding genes, the Recombination Activating Gene 1 (RAG-1, 1084 bp), and the Recombination Activating Gene 2 (RAG-2, 760 bp), for the to- tal taxonomic sampling. Total DNA was obtained from pre- served tissue samples (see voucher list in Appendix) with the DNeasy tissue kit (QIAGEN). PCR amplification was carried out using previously published primers [RAG-1 and RAG-2 (Teeling et al., 2000); vWF (Porter et al., 1996)]. To obtain both forward and reverse sequences for each gene region, internal primers were used for sequencing in addition to the PCR primers (sequences available upon request). All sequences were obtained with an automated ABI 3730XL sequencer. Sequence editing and prealignment were done with the Sequencher 4.2 software (Gene Codes). GenBank accession numbers and voucher information for megabats included in this study are provided in Appendix. In addition, we used published sequences for microbat outgroup taxa. GenBank accession numbers for A. jamaicensis are AY834655 (RAG-1), AY834663 (RAG-2), and AY834737 (vWF); for R. hardwickei, accession numbers are AF447518 (RAG-1), AF447535 (RAG-2), and AF447551 (vWF). Sequences of our three genes were submitted to parsimony analysis, separately and in combination. Here we report results of our combined analysis based on a tree-search strategy that consisted of 1,000 replicates of random addition sequences of taxa, each followed by tree bisection reconnection branch swap- ping (TBR). An additional round of TBR was done on optimal trees obtained. Clade support was assessed using Bremer or de- cay values (Bremer, 1994) and jackknife character resampling (Goloboff et al., 2003a). We calculated Bremer values via incre- mental sampling of suboptimal trees (see Giannini and Bertelli, 2004). Briefly, we saved up to 1,000 suboptimal trees one step longer than the previous optimum in successive stages. That is, we first searched for suboptimal trees 1 step longer than the op- timal tree length, next saving suboptimals up to 2, 3, 4, 5, 6, 7, 8, 9, and 10 steps longer than the optimal trees (9,968 trees up to 10 steps longer than the optimals were found). Second, jack- knife frequencies were estimated from 5,000 replications using unbiased symmetric resampling (Goloboff et al., 2003a). These analyses were executed in TNT (Goloboff et al., 2003b, In press). In addition, maximum likelihood (ML) analyses were per- formed for individual genes and the combined set using the GTR + ? model. Parameters were estimated from the data for each gene separately, which were treated as partitions in the combined analysis. Starting trees were obtained by maximum parsimony and 100 runs were done on each initial tree to obtain- ing ML trees. Statistical support was obtained with 100 boot- strap replications. All the ML analyses were run with the pro- gram RAxML (Stamatakis, 2006). RESULTS In the combined dataset, 751 sites were variable (RAG1 = 236, RAG2 = 163, vWF = 354) and 395 were parsimony informative (RAG1 = 116, RAG2 = 92, vWF = 187). Most of the substitutions were transitions (ti/tv: RAG2 = 10.3, RAG1 = 4.5, 12 N. P. Giannini, F. Cunha Almeida, N. B. Simmons, and K. M. Helgen vWF = 3.9), but neither transitions nor transversions were saturated in the sample (Fig. 1). Our combined parsimony analysis of all three genes resulted in 92 trees of 1,327 steps (strict con- sensus in Fig. 2). An additional TBR round on those trees did not increase the set of optimal trees. Pte ro - podidae was recovered as monophyletic. Nycti mene was placed as sister to all other megachiropterans, but the backbone of the lower section of the tree was poorly supported. By contrast, clades representing The systematic position of Pteropus leucopterus 13 FIG. 1. Saturation plots of transitions and transversions for each gene analyzed. Proportions of each substitution type were plotted against uncorrected p distances. Circles represent transitions and squares represent transversions. Symbols representing the ingroup are shown in light grey and symbols representing the outgroup are shown in black a number of subfamilies and other recognized sys- tematic groups were highly supported. Clades re- covered included Cynopterinae; a highly-supported group of rousettines and epomophorines; Epo mo - phorinae; a weak association of Macroglos sus with a highly supported Harpyionycterinae clade; and a well-supported pteropodine clade (marked A in Fig. 2) inclusive of the genera Melo nycteris, Ace ro - don, and Pteropus. Within the pteropodine clade, Melonycteris far- doulisi and Pteropus leucopterus formed a trichoto- my with a highly supported Acerodon + Pteropus group (clade B in Fig. 2). All the species of Pte - ropus, to the exclusion of leucopterus, were recov- ered as monophyletic with high support (clade C in Fig. 2). Resolution within Pteropus was poor. Only three clades were supported with resampling fre- quencies > 50%. These were Pteropus anetianus + P. samoensis (i.e., the samoensis species group); Pteropus woodfordi + P. molossinus, which renders the scapulatus species group (here represented by P. scapulatus and P. woodfordi) paraphyletic; and two clades with members from mixed species groups. Partial analyses using individual gene parti- tions (not shown) were less resolved; clades recov- ered in those analyses are marked with symbols in Fig. 1. In no case did P. leucopterus group with oth- er Pteropus species. The combined ML tree is shown in Fig. 3. This tree agrees with the combined parsimony tree (Fig. 2) in all supported groups specifically relevant to this study (marked A, B, and C). Groups A and B were recovered in separate ML analyses of each of the three genes, whereas group C was recovered in separate analyses of RAG-1 and vWF. Also, clades recovered within Pteropus were compatible across analyses (cf. Figs. 2 and 3). SYSTEMATICS Desmalopex Miller, 1907 is a valid pteropodid genus presently considered a junior synonym of Pteropus Brisson, 1762. In our combined analyses, as well as in individual-gene analyses, leucopterus never joined other species of Pteropus, demonstrat- ing the paraphyly of Pteropus under its current generic definition (Andersen, 1912; Simmons, 2005). However, all other species of Pteropus were recovered as a natural group identified by shared changes in all three genes independently and in combination, clearly suggesting that Pteropus (at least as represented by our sampling) is indeed monophyletic to the exclusion of leucopterus (clade C in Figs. 2 and 3) and is sister to the genus Ace - rodon (clade B in Figs. 2 and 3). ?Pteropus? leuco - pterus is also easily diagnosed morphologically, unique amongst megachiropterans in exhibiting a mosaic of features recalling both Pteralopex and Pteropus (cf. Andersen, 1909). As a consequence, we remove leucopterus from Pteropus and resurrect Desmalopex Miller, 1907, originally erected to in- clude this sole species, and adopt the combination Desmalopex leucopterus (Temminck, 1853). We briefly diagnose and discuss the content and distri- bution of Desmalopex below. 14 N. P. Giannini, F. Cunha Almeida, N. B. Simmons, and K. M. Helgen FIG. 2. Strict consensus tree resulting from a parsimony analysis of three nuclear coding genes combined (RAG-1, RAG-2, and vWF). Support values are given above (Bremer) and below (Jackknife) branches. Nodes supported by individual-gene analyses are marked with the following symbols: RAG-1 (), RAG-2 (), and vWF () Genus Desmalopex Miller, 1907 Diagnosis Desmalopex uniquely combines distinctive fea- tures of both Pteropus and Pteralopex sensu lato (i.e., incorporating both Pteralopex and Mirimiri). Desmalopex is a pteropodid genus uniquely charac- terized by its relatively large body size (forearm length ca. 100?150); relatively much shortened ex- ternal ears; pale-dark mottled wing membranes (see below); unusually large and moderately spaced up- per incisors, subequal in size, forming an arcuate row; large I2 (wide with one salient lateral cusp) but highly reduced I1, such that I2 is about five times larger than I1; relatively short canines, without a large secondary posterior cusp on C; relative- ly large first premolar that is permanent in both jaws; relatively small cheekteeth; P4 that is larger than M1, which is subsquare rather than rectangular; orbits deflected distinctly upward relative to the cra- nial axis; postorbital processes that are ossified to the zygomatic arches in mature individuals, a com- plete alar canal (an unusual feature in megachiro - pterans, seen elsewhere only in Mirimiri and some Pteralopex ? cf. Giannini et al. 2006b: 113); basi- cranial configuration in which the petrosal is some- what sunk laterally, the posttympanic and para- condylar processes are unusually large, and the post- glenoid foramen is displaced laterally; and relative- ly gracile mandible featuring a low-slung sloping symphysis and a short and slender coronoid process. We plan to explore in greater detail these and other morphological attributes (and their phylogenetic significance) in subsequent contributions. Content and Distribution Desmalopex is endemic to the oceanic Phil- ip pines. The type and only described species of The systematic position of Pteropus leucopterus 15 FIG. 3. ML tree of the combined set based on gene partitions modeled using GTR + ? (likelihood score -11,490.72). Values near nodes represent bootstrap estimates of clade support based on 100 replications. Pteropus scapulatus was excluded from this analysis due to the influence of its missing data on branch lengths Des malopex is Desmalopex leucopterus (Tem - minck, 1853), known only from forested habitats (up to at least 2,300 m) on Luzon and the adjacent land-bridge island of Catanduanes (specimens at AMNH, BMNH, FMNH, RMNH, SMF, and USNM). Syno nyms are Pteropus chinensis Gray, 1870 (misprovenanced from China ? see Ander- sen, 1912) and an incorrect subsequent spelling of leucopterus under the name combination ?Spect- rum leucopterum? by Gray (1870). Currently unde - scribed species referable to Desmalopex have been recorded from the Philippine islands of Mindo ro (in the Mindoro Faunal Region ? see Heaney et al., 1998) and Dinagat (in the Minda nao Faunal Region ? K. M. Helgen, personal observation; specimens at DMNH). DISCUSSION Affinities of Desmalopex remain somewhat un- certain and invite speculation and further research. Desmalopex belongs in Pteropodinae sensu Berg - mans (1997), but exactly where in this group is not yet clear. Within the subfamily Pteropodinae, Bergmans (1997) recognized three tribes: Ptero - podini (including Pteropus, Acerodon, Neopteryx, Styloctenium, Pteralopex, and the more recently recognized genus Mirimiri ? Helgen, 2005), a re- stricted Macroglossini (incorporating only Macro - glos sus and Syconycteris), and Notopterini (Melony - cteris and Notopteris). The problematic African genus Eidolon was suggested as another candidate member of the Pteropodinae based on morphologi- cal evidence and a combined analysis including mi- tochondrial genes (Giannini and Simmons, 2005), but our current analyses using only nuclear genes either rejected this grouping (Fig. 2) or did not pro- vide clear support for it (Fig. 3). Also from the cur- rent analysis (and some of our previous results ? e.g., Giannini and Simmons, 2003, 2005; Giannini et al., 2006), it seems clear that Macroglossus (and thus its apparently closely-related sister lineage, Sy - conycteris ? Andersen, 1912; Kirsch et al., 1995) does not belong in Pteropodinae. However, we re- covered Melonycteris in clade A (Figs. 2 and 3), an arrangement recalling the DNA-DNA hybridization results of Kirsch et al. (1995), suggesting that the Notopterini may be part of Pteropodinae. Based on RAG-1 changes, Desmalopex joined Melonyc te - ris and typical Pteropodini in a trichotomy (Fig. 2), suggesting that, pending further resolution, Desma - lopex may belong in a separate tribe within the sub- family. The single-gene parsimony analyses did not favor any particular association of Desmalopex in this trichotomy. However, the combined ML analy- sis suggested, albeit weakly, that Melonycteris may join a clade including Desmalopex (Fig. 3). Andersen (1912) included leucopterus in Pte ro - pus within the pselaphon species group, which is characterized by many dental traits including heavy teeth, strong ledges (cingula), and relative enlarge- ment of specific incisors (I2) and premolars (P1). To Andersen (1912: 294) ?the pselaphon group shows decidedly leanings towards the highly specialized genus Pteralopex?. We have not had access to tissue samples of Pteralopex, but we hypothesize that, as suggested by Miller (1907), Desmalopex may be part of a lineage including Pteralopex (and its more recently described sister genus, Mirimiri) and per- haps other pteropodine genera. To the craniodental characters, we add several external characters that may support such association. Some photographs of Desmalopex that we have seen seem to show that it has an eye with a red-orange iris, an interesting trait shared also with Pteralopex and Mirimiri, but also with some Pteropus species, in which genus the iris is more usually brown (Flannery, 1995; Helgen, 2005). The external ears of Desmalopex, like those of Pteralopex and Mirimiri, are relatively small in comparison to most species of Pteropus (Corbet and Hill, 1992; Helgen, 2005). Desmalopex also exhibits an uneven distribution of pigment in the patagia (?melanin spotting? of Giannini and Simmons, 2005: character 33). This character state is shared with Pte ralopex, Sty lo ctenium, some Pteropus species (e.g., P. capistratus), and is carried to an extreme in Neopteryx (see Hayman, 1946; Bergmans and Rozendaal, 1988). Perhaps significantly, mel anin spotting is also present in Melonycteris and Ne - sonycteris, though it also occurs in at least one phy- logenetically unrelated lineage, the Nyctime ni- nae, and as an individual feature in other pteropodid genera. Regarding relationships within Pteropus, the nu- clear genes that we sampled lack variation sufficient to provide resolution, suggesting that fast-evolving genes (e.g., mitochondrial markers) should be added to the analysis. However, we anticipate that species groups, as traditionally recognized since Andersen (1912), will require some major rearrangements. While our analyses either recover or did not strong- ly contradict some polytypic species groups (the sa moensis, tonganus, vampyrus groups), monophy- ly of at least two such species groups was rejected. First, P. woodfordi (scapulatus species group) ap- peared as sister to P. molossinus (molossinus species 16 N. P. Giannini, F. Cunha Almeida, N. B. Simmons, and K. M. Helgen group) rather than P. scapulatus. Second, P. pumilus did not group with the other member of the subniger species group included in our study (P. hypome- lanus). The significance of these results will be more fully explored elsewhere in a review of interspecific relationships among the remainder of species classi- fied within Pteropus. Unfortunately, we have not had access to tissues or sequences from Pteropus insularis, P. pselaphon, P. pilosus, or P. tuberculatus, all of which Andersen (1912) allied with Desmalopex leucopterus within his ?Pteropus pselaphon group.? Further study is needed to firmly establish the phylogenetic place- ment of these species. However, we note that none of these species of Pteropus possess some of the more distinctive phenetic anatomical attributes of D. leucopterus, such as its mottled wings, relatively small and subsquare M1, extraordinary discrepancy in size between I2 and I1, co-ossification of the pos- torbital processes and the jugal spine of the zygo- matic arches, and zygomata that are parallel-sided rather than bowed in dorsal view (each of these traits are instead shared with Pteralopex.) Recognition of Desmalopex as a valid genus brings the number of recognized pteropodine genera (sensu Giannini and Simmons, 2005: 24) to seven (Helgen, 2005; Simmons, 2005) or eight (including Eidolon ? Giannini and Simmons, 2005). With the exception of Eidolon, all of these lineages comprise relatively very large-bodied bats with distributions centered on the Indo-Australian region. One is wide- spread throughout Australasia and the Pacific and Indian Ocean regions (Pteropus), one is shared between Wallacea and the oceanic Philippines (Styloctenium), one occurs primarily throughout Wallacea and the oceanic Philippines but also ex- tends marginally to the Sunda Shelf region on Pa la - wan and adjacent islands (Acerodon), one is restrict- ed to Sulawesi (Neopteryx), one is restricted to the oceanic Philippines (Desmalopex), one is restricted to the Solomon Archipelago (Pteralopex), and one is restricted to Fiji (Mirimiri) (Helgen, 2005; Sim - mons, 2005; Esselstyn, 2007). As noted above, our results confirm that one lin- eage formerly classified among ?macroglossines? (Andersen, 1912; Giannini and Simmons, 2005; Giannini et al., 2006), represented by Melonycteris (endemic to the Bismarck Archipelago) and Neso - nycteris (endemic to the Solomon Archipel ago), var - iably recognized as separate genera or congener- ic subgenera (Pulvers and Colgan, 2007), is also referable to the Pteropodinae (Bergmans, 1997). The West Pacific genus Notopteris (occurring in Vanuatu, New Caledonia, Fiji, and the subfossil re- cord of Tonga), unsampled in our study, is potential- ly another member of this clade (Kirsch et al., 1995). Concentration of endemic pteropodine line- ages throughout insular archipelagos from Sulawesi and the Philippines to the Solomons and Fiji indi- cates a probable origin for the group within the Indo-Pacific region?s extensive island arc systems. Finally, our parsimony analyses and our previous studies (Giannini and Simmons, 2005; Giannini et al., 2006) suggest that members of Pteropodinae are nested within Pteropodidae. By contrast, our com- bined ML analysis and a previous study that includ- ed a comparable sample of Pteropus species (Col - gan and Da Costa, 2002: 19 species) place pte ro - podines in a more basal position amongst megachi- ropterans. However, the weakly supported backbone of both MP and ML trees (Figs. 2 and 3) indicate only uncertainty about the placement of pteropo - dines with this limited taxonomic sample. CONCLUSIONS We have provided evidence that leucopterus does not belong within the taxonomic boundaries of a mo no phyletic genus Pteropus, and resurrect Desmalopex Miller, 1907 as a valid generic name to accommodate this species. Morphological evidence points to an association of Desmalopex with Pte - ralopex and Mirimiri, and perhaps to pteropodine genera other than Acerodon and Pteropus. With the exclusion of leucopterus, monophyly of the ge nus Pteropus (as represented by a sample of 18 species from 12 species groups) is supported by shared changes in three nuclear coding genes. At least two of the currently recognized species groups of Pte - ropus may not be monophyletic. ACKNOWLEDGEMENTS We thank Lawrence Heaney (Field Museum of Natural History, Chicago), Jim Patton and Carla Cicero (Museum of Vertebrate Zoology, Berkeley), John Wible and Suzanne McLaren (Carnegie Museum, Pittsburgh), Burton Lim and Judith Eger (Royal Ontario Museum), Denis O?Meally (Australian Museum), Jeremy Jacobs, Louise Emmons, and James Mead (USNM) for access to tissue samples that made possible this contribution. Linda Gordon and Don Wilson (USNM), Lawrence Heaney (FMNH), Paula Jenkins (BMNH), Dieter Kock (SMF), Chris Smeenk (RMNH), Burton Lim and Mark Engstrom (ROM), graciously allowed access to speci- mens in their care. We also thank Natalee Stephens for help in the laboratory, Paul Sweet for additional tissue samples, and especially Lawrence Heaney and Jacob Esselstyn for fruitful exchange of ideas on the problem of Desmalopex. Funding for this report was provided by the National Science Foundation The systematic position of Pteropus leucopterus 17 (research grant DEB-9873663 to N. B. S.), Coleman and Vernay postdoctoral fellowships at the AMNH to N.P.G., a Henry MacCracken doctoral fellowship at New York University to F.C.A., and a postdoctoral fellowship at the Smithsonian Institution and funding from the Bernice P. Bishop Museum to K.M.H. LITERATURE CITED ANDERSEN, K. 1909. On the characters and affinities of ?Des - malopex? and Pteralopex. Annals and Magazine of Natural History, 8, 3: 213?222. ANDERSEN, K. 1912. Catalogue of the Chiroptera in the collec- tion of the British Museum. Volume I: Megachiroptera. Trustees British Museum (Natural History), Lon don, ci + 854 pp. BERGMANS, W. 1997. Taxonomy and biogeography of African fruit bats (Mammalia, Megachiroptera). 5. The genera Lissonycteris Andersen, 1912, Myonycteris Matschie, 1899, and Megaloglossus Pagenstecher, 1885; general remarks and conclusions; annex: key to all species. Beaufortia, 47: 11?90. BERGMANS, W., and F. G. ROZENDAAL. 1988. Notes on collec- tions of fruit bats from Sulawesi and some off-lying islands (Mammalia, Megachiroptera). Zoologische Verhandelingen, 248: 1?74. BREMER, K. 1994. Branch support and tree stability. Cladistics, 10: 295?304. COLGAN, D. J., and P. DA COSTA. 2002. Megachiropteran evolu- tion studied with 12S rDNA and c-mos DNA sequences. Journal of Mammalian Evolution, 9: 3?22. CORBET, G. B., and J. E. HILL. 1992. Mammals of the Indo - malayan region: a systematic review. Oxford University Press, London, 488 pp. ESSELSTYN, J. A. 2007. A new species of stripe-faced fruit bat (Chiroptera: Pteropodidae: Styloctenium) from the Philip - pines. Journal of Mammalogy, 88: 951?958. FLANNERY, T. F. 1995. Mammals of Papua New Guinea. Robert Brown and Association, Sydney, 439 pp. GIANNINI, N. P., and S. BERTELLI. 2004. A phylogeny of extant penguins based on integumentary characters. The Auk, 121: 422?434. GIANNINI, N. P., and N. B. SIMMONS. 2003. A phylogeny of me - ga chiropteran bats (Mammalia: Chiroptera: Pteropo didae) based on direct optimization analysis of one nuclear and four mitochondrial genes. Cladistics, 19: 496?511. GIANNINI, N. P., and N. B. SIMMONS. 2005. Conflict and congru- ence in a combined DNA-morphology analysis of megachi- ropteran bat relationships (Mammalia: Chiroptera: Ptero po - didae). Cladistics, 21: 411?437. GIANNINI, N. P., F. C. ALMEIDA, N. B. SIMMONS, and R. DE - SALLE. 2006a. Phylogenetic relationships of the enigmatic harpy fruit bat, Harpyionycteris (Mammalia: Chiroptera: Pteropodidae). American Museum Novitates, 3533: 1?12. GIANNINI, N. P., J. R. WIBLE, and N. B. SIMMONS. 2006b. On the cranial osteology of Chiroptera. I. Pteropus (Megachiro - ptera: Pteropodidae). Bulletin of the American Museum of Natural History, 295: 1?134. GOLOBOFF, P. A., J. S. FARRIS, M. K?LLERSJ?, B. OXELMAN, M. RAM?REZ, and C. SZUMIK. 2003a. Improvements to resam- pling measures of group support. Cladistics, 19: 324?332. GOLOBOFF, P. A., J. S. FARRIS, and K. NIXON. 2003b. TNT: Tree analysis using new technology. Version 1.1. Available at www.zmuc.dk/public/phylogeny/TNT. GOLOBOFF, P. A., J. S. FARRIS, and K. NIXON. In press. TNT, a free program for phylogenetic analysis. Cladistics. GRAY, J. E. 1870. Catalogue of monkeys, lemurs and fruit eating bats in the collections of the British Museum. British Muse - um, London, 137 pp. HAYMAN, R. W. 1946. A new genus of fruit-bat and a new squir- rel, from Celebes. The Annals and Magazine of Natural His - tory, 11, 12: 569?578. HEANEY, L. R., D. S. BALETE, M. L. DOLAR, A. C. ALCALA, A. T. L. DANS, P. C. GONZALES, N. R. INGLE, M. V. LEPITEN, W. L. R. OLIVER, P. S. ONG, E. A. RICKART, B. R. TABARANZA, JR., and R. C. B. UTZURRUM. 1998. A synopsis of the mam- malian fauna of the Philippine Islands. Fieldiana: Zoology (N.S.), 88: 1?61. HELGEN, K. M. 2005. Systematics of the Pacific monkey-faced bats (Chiroptera: Pteropodidae), with a new species of Pte - ra lopex and a new Fijian genus. Systematics and Bio diver - sity, 3: 433?453. KIRSCH, J. A., T. F. FLANNERY, M. S. SPRINGER, and F.-J. LA - POINTE. 1995. Phylogeny of the Pteropodidae (Mammalia: Chiroptera) based on DNA hybridization with evidence of bat monophyly. Australian Journal of Zoology, 43: 395?428. KOOPMAN, K. F. 1993. Order Chiroptera. Pp. 137?241, in Mam - mal species of the World: a taxonomic and geographic ref- erence, 2nd edition (D. E. WILSON and D. M. REEDER, eds.). Smithsonian Institution Press, Washington D.C., 1206 pp. KOOPMAN, K. F. 1994. Chiroptera: systematics. Handbook of zoology. Vol. 8, Part 60: Mammalia. Walter de Gruyter, Ber - lin, 217 pp. MILLER, G. S., JR. 1907. The families and genera of bats. Bul - letin of the United States National Museum, 57: 1?282. PORTER, C. A., M. GOODMAN, and M. J. STANHOPE. 1996. Evi - dence on mammalian phylogeny from sequences of exon 28 of the von Willerbrand factor gene. Molecular Phyl o genet - ics and Evolution, 5: 89?101. PULVERS, J. N., and D. J. COLGAN. 2007. Molecular phylogeog- raphy of the fruit bat genus Melonycteris in Northern Mela - nesia. Journal of Biogeography, 34: 713?723. SIMMONS, N. B. 2005. Order Chiroptera. Pp. 312?529, in Mam - mal species of the World: a taxonomic and geographic ref- erence, 3rd edition (D. E. WILSON and D. M. REEDER, eds.). The Johns Hopkins University Press, Baltimore, 2142 pp. STAMATAKIS, A. 2006. RAxML-VI-HPC: Maximum likelihood- based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics, 22: 2688?2690. TEELING, E. C., M. SCALLY, D. KAO, M. ROMAGNOLI, M. S. SPRIN GER, M. J. STANHOPE. 2000. Molecular evidence re- garding the origin of echolocation and flight in bats. Nature, 403: 188?192. 18 N. P. Giannini, F. Cunha Almeida, N. B. Simmons, and K. M. Helgen Received 20 August 2007, accepted 09 February 2008 The systematic position of Pteropus leucopterus 19 A P P E N D IX S pe ci es V ou ch er T is su e ID A cc es si on n um be rs L oc al it y R A G -1 R A G -2 vW F A ce ro do n ce le be ns is A M N H 2 72 87 7 A M C C 1 24 96 6 U E 61 79 46 U E 61 78 96 U E 61 79 28 ?I nd on es ia , S ul aw es i? C yn op te ru s sp hi nx A M N H 2 74 35 4 A M C C 1 01 68 8 U E 61 79 47 U E 61 78 97 D Q 44 56 97 V ie tn am , H a G ia ng P ro vi nc e, V i X uy en D is tr ic t, C ao B o C om m un e, M t T ay C on L in h II D es m al op ex l eu co pt er us F M N H E A R 1 69 7 F M N H E A R 1 69 7 U E 61 79 66 U E 61 79 15 U E 61 79 29 P hi li pp in es , C at an du an es D ob so ni a in er m is A M N H P R S 2 77 1 A M C C 1 24 42 8 U E 61 79 48 U E 61 78 98 D Q 44 56 86 S ol om on I sl an ds , W es te rn P ro vi nc e, N ew G eo rg ia G ro up , V on av on a L ag oo n D . m ag na A M M 2 07 35 E B U 2 57 57 U E 61 79 49 U E 61 78 99 U E 61 79 30 P ap ua N ew G ui ne a, S id ei a M is si on , M il ne B ay P ro vi nc e E id ol on h el vu m C M N H 1 02 02 0 S P 50 79 U E 61 79 50 U E 61 79 00 U E 61 79 31 K en ya , M ba le , K ak am eg a D is tr ic t, W es te rn P ro vi nc e E on yc te ri s sp el ae a M V Z 1 76 48 7 M V Z 1 76 48 7 U E 61 79 51 U E 61 79 01 D Q 44 56 84 C hi na , Y un na n P ro vi nc e E po m op ho ru s w ah lb er gi A M N H 1 17 33 6 JC K 4 82 0 U E 61 79 53 U E 61 79 03 D Q 44 56 91 M oz am bi qu e, Z am be zi a, M t. N am ul i E po m op s fr an qu et i A M N H 2 38 35 6 A M C C 1 09 07 0 U E 61 79 52 U E 61 79 02 D Q 44 56 92 C en tr al A fr ic an R ep ub li c, S an gh a, D za ng a- S an gh a H ar py io ny ct er is w hi te he ad i F M N H 1 46 64 6 L R H 4 81 1 U E 61 79 54 U E 61 79 04 D Q 44 56 90 P hi li pp in es , M in da na o, B uk id no n P ro v. , M ou nt K it an gl ad R an ge M ac ro gl os su s m in im us C E F 8 00 A M C C 1 24 28 3 U E 61 79 55 U E 61 79 05 D Q 44 56 93 S ol om on I sl an ds , W es te rn P ro vi nc e, N ew G eo rg ia G ro up , V el la L av el la I sl an d M eg al og lo ss us w oe rm an ni A M N H 2 68 35 8 A M C C 1 09 06 4 U E 61 79 56 U E 61 79 06 D Q 44 57 02 C en tr al A fr ic an R ep ub li c, S an gh a, D za ng a- S an gh a M el on yc te ri s fa rd ou li si A M N H P R S 2 65 3 A M C C 1 24 27 9 U E 61 79 57 U E 61 79 07 D Q 44 56 99 S ol om on I sl an ds , W es te rn P ro vi nc e, N ew G eo rg ia G ro up , V el la L av el la I sl an d M yo ny ct er is t or qu at a A M N H 2 68 36 2 A M C C 1 09 05 8 U E 61 79 58 U E 61 79 08 D Q 44 57 00 C en tr al A fr ic an R ep ub li c, S an gh a, D za ng a- S an gh a N yc ti m en e al bi ve nt er N o vo uc he r N o vo uc he r A Y 24 98 70 A F 44 75 31 A F 44 75 49 N . v iz ca cc ia A M N H P R S 2 63 6 A M C C 1 24 20 8 U E 61 79 59 U E 61 79 09 D Q 44 56 98 S ol om on I sl an ds , W es te rn P ro vi nc e, N ew G eo rg ia G ro up , V el la L av el la I sl an d P te no ch ir us j ag or i F M N H 1 75 39 5 L R H 6 70 0 U E 61 79 60 U E 61 79 10 D Q 44 56 96 P hi li pp in es , L uz on , K al in ga P ro vi nc e, B al ba la n M un ic ., B al ba la sa ng P te ro pu s al ec to A M M 3 25 64 E B U 9 87 3 U E 61 79 61 U E 61 79 11 U E 61 79 32 A us tr al ia , L is m or e D is tr ic t, N ew S ou th W al es P. a ne ti an us A M N H 2 72 87 4 A M C C 1 24 96 3 U E 61 79 62 U E 61 79 12 U E 61 79 33 V an ua tu P. c ap is tr at us U S N M 5 80 01 8 U S N M 5 80 01 8 U E 61 79 75 U E 61 79 23 U E 61 79 43 ?P ap ua N ew G ui ne a? P. c on sp ic il la tu s M V Z 1 40 20 1 M V Z 1 40 20 1 U E 61 79 63 U E 61 79 34 P ap ua N ew G ui ne a, B ai ta ba g P la nt at io n, M ad an g, M ad an g P ro vi nc e P. g ig an te us C M N H 9 22 05 N K 1 05 24 E U 61 79 64 E U 61 79 13 E U 61 79 35 In di a, A ra ku , A nd hr a P ra de sh P. h yp om el an us U nc at al og ed P 44 47 U E 61 79 65 U E 61 79 14 D Q 44 56 87 C ap ti vi ty L ub ee F ou nd at io n P. l yl ei R O M 1 10 94 3 F 4 42 69 U E 61 79 67 U E 61 79 16 U E 61 79 16 V ie tn am , S oc T ra ng P. m ol os si nu s U S N M 5 66 56 7 U S N M 5 66 56 7 U E 61 79 69 U E 61 79 18 U E 61 79 38 ?C ar ol in e Is la nd s? P. n eo hi be rn ic us A M N H 2 72 87 2 A M C C 1 24 96 1 E U 61 79 70 E U 61 79 19 E U 61 79 39 ?P ap ua N ew G ui ne a? P. p el ew en si s M V Z 1 85 26 3 M V Z 1 85 26 3 E U 61 79 68 E U 61 79 17 E U 61 79 37 W C ar ol in es , Y ap I sl an ds P. p ol io ce ph al us M 3 54 96 E B U 1 37 68 E U 61 79 71 E U 61 79 20 E U 61 74 0 A us tr al ia , M or td al e, N ew S ou th W al es P. p um il us F M N H L R H 4 26 1 F M N H L R H 4 26 1 E U 61 79 72 E U 61 79 21 E U 61 79 41 ?P hi li pp in es ? V ou ch er i nf or m at io n an d G en B an k ac ce si on n um be rs o f m eg ab at s pe ci m en s us ed i n th e pr es en t st ud y, i n al ph ab et ic al o rd er b y sp ec ie s. I m pr ec is e lo ca li ti es a re q uo te d. A bb re vi at io ns o f In st it ut io ns : A M M , A us tr al ia n M us eu m , S yd ne y; A M C C , A m br os e M on el l C ry o C ol le ct io n (A M N H ); A M N H , A m er ic an M us eu m o f N at ur al H is to ry , N ew Y or k; C M N H C ar ne gi e M us eu m of N at ur al H is to ry , P it ts bu rg h; E B U E vo lu ti on ar y B io lo gy U ni t, A us tr al ia n M us eu m , S yd ne y; F M N H , F ie ld M us eu m o f N at ur al H is to ry , C hi ca go ; M V Z , M us eu m o f V er te br at e Z oo lo gy , U ni ve rs it y of C al if or ni a, B er ke le y; R O M R oy al O nt ar io M us eu m , T or on to ; U S N M S m it hs on ia n In st it ut io n, N at io na l M us eu m o f N at ur al H is to ry , W as hi ng to n D .C . O th er a bb re vi at io ns re fe r to c ol le ct or ?s c at al og 20 N. P. Giannini, F. Cunha Almeida, N. B. Simmons, and K. M. Helgen S pe ci es V ou ch er T is su e ID A cc es si on n um be rs L oc al it y R A G -1 R A G -2 vW F P. s am oe ns is A M N H 2 72 87 6 A M C C 1 24 96 5 E U 61 79 73 E U 61 79 22 E U 61 79 42 ?A m er ic an S am oa ? P. s ca pu la tu s A M M 3 24 40 E B U 9 34 1 E U 61 77 94 A us tr al ia , W er ri ng to n D ow ns , P en ri th , N ew S ou th W al es P. t on ga nu s A M N H 2 72 87 3 A M C C 1 24 96 2 E U 61 79 76 E U 61 79 24 D Q 44 56 95 To ng a P. v am py ru s R O M 1 10 94 8 F 4 42 74 E U 61 79 77 E U 61 79 25 E U 61 79 44 V ie tn am , S oc T ra ng P. w oo df or di A M N H 2 72 87 5 A M C C 1 24 96 4 E U 61 79 78 E U 61 79 26 E U 61 79 45 ?S ol om on I sl an ds ? R ou se tt us a eg yp ti ac us A M N H 1 17 38 6 JC K 4 96 0 E U 61 79 79 E U 61 79 27 D Q 44 56 88 M oz am bi qu e, Z am be zi a, M t. N am ul i R . a m pl ex ic au da tu s N o vo uc he r N o vo uc he r A F 44 75 12 A F 44 75 29 A Y 05 78 36 A P P E N D IX . C on ti nu ed