Journal of Biogeography (J. Biogeogr.) (2007) 34, 364–375

ORIGINAL ARTICLE

Short-faced bats (Phyllostomidae: Stenodermatina): a Caribbean radiation of strict frugivores Liliana M. Da´valos* 

Department of Ecology, Evolution and Environmental Biology, Columbia University and Division of Vertebrate Zoology, American Museum of Natural History, Central Park West at 79th Street New York, New York 10024-5192, USA

ABSTRACT

Aim To test the hypothesis that Caribbean Short-faced bats descended from a single recent ancestor that originated in the continental Neotropics (Mexico, Central America and/or South America). Location The Neotropics, including the West Indies. Methods New mitochondrial cytochrome b and nuclear Rag2 sequences were combined with published molecular data to estimate phylogenetic relationships and sequence divergence among Short-faced bats. The resulting phylogenies were compared with those compatible with the single-origin hypothesis using two model-based statistical tests. Confidence limits on sequence divergence were estimated using a parametric bootstrap. Results All molecular phylogenies revealed two independent Caribbean lineages and showed that continental Short-faced bats share a recent common ancestor. Morphology-based trees compatible with the single-origin hypothesis were significantly worse at explaining the molecular data than any molecular phylogeny.

*Correspondence: Liliana M. Da´valos. E-mail: [email protected]  Present address: The Institute for Comparative Genomics, American Museum of Natural History, Central Park West at 79th Street New York, New York 10024-5192, USA.

Main conclusions The ancestor of all Short-faced bats reached the Antilles in the Miocene, too recently to have used a proposed Oligocene land bridge, and well before the Pleistocene glaciations that are thought to have facilitated dispersal for many bats. After a long period of isolation, Short-faced bats diversified quickly on the Caribbean islands. A single Short-faced lineage then reached the continent and subsequently expanded its range and diversified into the four extant genera. Among bats, independent lineages of aerial insectivores and nectarivores have also recolonized the continent after evolving in the West Indies. The evidence for an insular origin of the short-faced frugivorous radiation completes a dynamic model of Caribbean biogeography that encompasses an entire biological community. Keywords Caribbean biogeography, Caribbean-origin hypothesis, Central America, evolutionary radiation, island-origin hypothesis, molecular phylogeny, phyllostomid bats.

INTRODUCTION The subtribe Stenodermatina (Chiroptera: Phyllostomidae) includes eight genera restricted to lowland forests and secondary habitats of the Neotropics, including the West Indies (Table 1) (Howe, 1974; Silva-Taboada, 1979; Emmons, 1997). The short rostrum that characterizes members of this 364

www.blackwellpublishing.com/jbi doi:10.1111/j.1365-2699.2006.01610.x

subtribe gives them their common name, the Short-faced bats. All Short-faced bats are strict frugivores and, with the exception of Ardops and Ametrida at some localities, are rarely captured (Goodwin & Greenhall, 1961; Handley, 1976; SilvaTaboada, 1979; Carter et al., 1981; Pedersen et al., 1996). Close evolutionary relationships among the genera have been suggested for several decades (de la Torre, 1961; Greenbaum ª 2006 The Author Journal compilation ª 2006 Blackwell Publishing Ltd

Caribbean radiation of frugivorous bats Table 1 Geographical distribution of all currently recognized Short-faced bats species and taxonomic sampling of this study (Koopman, 1994; Sua´rez & Dı´az-Franco, 2003; Mancina & Garcia-Rivera, 2005)

Taxon

Geographical distribution

Sampled

Centurio senex senex Centurio senex greenhalli Pygoderma bilabiatum bilabiatum

Mexico to Venezuela Trinidad South-east Brazil, Paraguay, north-east Argentina Bolivia, north-central Argentina Northern South America to central Brazil, Trinidad Venezuela to Bolivia Cuba Jamaica Dominica Guadeloupe Martinique St Lucia Montserrat, Nevis, St Eustatius, St Kitts Puerto Rico, Vieques Virgin Islands Cuba Cuba Cuba, Grand Cayman, Isla de Pinos* Hispaniola

Yes Yes Yes

Pygoderma bilabiatum magna Ametrida centurio Sphaeronycteris toxophyllum Cubanycteris silvai* Ariteus flavescens Ardops nichollsi nichollsi Ardops nichollsi annectens Ardops nichollsi koopmani Ardops nichollsi luciae Ardops nichollsi montserratensis Stenoderma rufum rufum Stenoderma rufum darioi Phyllops vetus* Phyllops silvai* Phyllops falcatus falcatus Phyllops falcatus haitiensis

Yes Yes Yes – Yes No No No No Yes Yes No – – Yes Yes

*Indicates extinct taxon.

et al., 1975), and the monophyly of the Short-faced clade has been confirmed by analyses of multiple suites of characters (Lim, 1993; Wetterer, 2003). Discussion of the biogeography of Short-faced bats centres on one question: how did these stocky bats reach the islands of the Caribbean? Only one other strict frugivore, Artibeus jamaicensis, ranges throughout the West Indies, and morphological and molecular evidence suggests it is conspecific with continental populations (Pumo et al., 1988, 1996; Phillips et al., 1989, 1991). Over-water dispersal from the neighbouring

landmasses of Mexico, Central America and/or South America (Baker & Genoways, 1978; Koopman, 1989; Morgan, 2001) has been the prevalent biogeographical explanation for the distribution of Short-faced bats (Fig. 1). More recently, and based on phylogenetic analyses of morphological data, Tavares & Simmons (2000) and Da´valos (2004) concluded that the ancestor of Short-faced bats could have used an ancient land bridge connecting South America and the Greater Antilles and diversified subsequently (Fig. 1). Although differing in the details, all hypotheses are grounded on the assumption that

Figure 1 (a) Map of the Caribbean showing how the ancestor of a monophyletic Caribbean Short-faced clade may have reached the islands. From Middle America to the Greater Antilles: the common ancestor of Antillean Short-faced bats and Artibeus or close relatives (Koopman, 1989), or the common ancestor of a Short-faced clade (Morgan, 2001); from South America through the Lesser Antilles: the common ancestor of a Short-faced clade (Baker & Genoways, 1978). From South America through the ancient Greater Antilles Aves Ridge land bridge (GAARlandia), in grey (Tavares & Simmons, 2000; Da´valos, 2004). (b, c) Two morphology-based phylogenies showing monophyly of a Caribbean Short-faced clade. Part (b) is redrawn from Lim (1993) and (c) is redrawn from Wetterer et al. (2000). Journal of Biogeography 34, 364–375 ª 2006 The Author. Journal compilation ª 2006 Blackwell Publishing Ltd

365

L. M. Da´valos Caribbean Short-faced bats are one another’s closest relatives. Two features of previous studies stand out: (1) morphology was the only source of data (even for those studies grounded on phylogeny; Tavares & Simmons, 2000; Da´valos, 2004) and (2) whether the phylogeny was compatible with other explanations was not tested. This study presents new sequences of the mitochondrial cytochrome b (1.14 kb) for both continental and Caribbean Short-faced bats. These mitochondrial sequences were added to published fragments of the mitochondrial 12S, tRNAval and 16S genes (the last three hereafter referred to as mitochondrial ribosomal DNA or mtrDNA), and new and published sequences of the recombination-activating gene 2 (Rag2) (Baker et al., 2000). The different loci were analysed separately and in combination to estimate relationships among Shortfaced bats. The phylogenies and molecular divergences obtained were used to test hypotheses on the origin of Caribbean Short-faced bats, and the timing and sequence of diversification events that gave rise to current diversity in the group. MATERIALS AND METHODS Taxon sampling All extant genera and 11 of the 16 extant subspecies (Simmons, 2005) of Short-faced phyllostomid bats were sampled. Individuals from as many localities as available were included to best capture the genetic diversity within species. Table 2 lists the sequences generated for this study. Sequences from the sister taxa to the Short-faced bats (Baker et al., 2000), Artibeus (U26277, AF316432, ACU66519) and Dermanura (AY395810, AF316443, ACU66511), were included for outgroup comparison. Molecular data For most specimens DNA was isolated from tissue that had been frozen or preserved in ethanol or lysis buffer in the field (Table 2), using the Qiagen DNeasy Tissue Extraction Kit (Qiagen, Inc., Valencia, CA, USA) and following the manufacturer’s protocol. DNA was isolated from c. 5 mm ribs that had been stored dry in a museum cabinet for specimens of Sphaeronycteris, Centurio and Pygoderma [American Museum of Natural History (AMNH) nos 194213, 262637, 256330 and 261759], following the protocol of Iudica et al. (2001). Extracted DNA was used as a template in polymerase chain reaction (PCR) amplifications, and sequenced with cytochrome b primers and methods described elsewhere (Da´valos & Jansa, 2004). Sequences of Ardops from Nevis, St Eustatius and St Kitts were acquired in a previous study (Carstens et al., 2004). Amplification of Rag2 from Phyllops was performed following the methods of Baker et al. (2000). DNA extraction and PCR amplification from museum specimens took place in the ancient DNA room of the Molecular Systematics Labor366

atory, a separate facility from the Cullman Laboratory for Molecular Systematics of the American Museum of Natural History, where all other experiments were performed. All mtrDNA and most Rag2 sequences were downloaded from GenBank (Baker et al., 2000, 2003) as follows: Ametrida (AY395802, AF316430), Ardops (AY395803, AF316434), Ariteus (AY395804, AF316435), Centurio (AF263227, AF316438), Pygoderma (AY395826, AF316483), Sphaeronycteris (AY395828, AF316486) and Stenoderma (AY395829, AF316487). Molecular sequences generated as part of this study have been deposited in GenBank under accession numbers AY604431, AY604432, AY604434–AY604453 and DQ211651 (Table 2). Data analysis Protein-coding cytochrome b and Rag2 sequences were aligned by eye using sequencher 4.1 (GeneCodes Corporation, Ann Arbor, MI, USA). paup* 4.0b10 (Swofford, 2002) was used to calculate uncorrected pairwise (‘p’) cytochrome b distances among taxa. Positional homology in the mtrDNA sequences was inferred using mafft version 5.734 (Katoh et al., 2005) with a gap opening : gap extension penalty of 1.53 : 0.123, the setting recommended for ribosomal sequence alignment. A second method to infer sequence homology, BlastAlign version 2.2.9 (Belshaw & Katzourakis, 2005), was applied to evaluate the reliability of the alignment resulting from mafft. Phylogenetic analyses were initially conducted separately on the mtrDNA, cytochrome b and Rag2 data sets. The best-fit maximum likelihood model for analysis of each data set was selected using the Akaike information criterion (AIC) in modeltest 3.06 (Posada & Crandall, 1998). Maximumlikelihood estimates of phylogeny and bootstrap pseudoreplicates were conducted using phyml version 2.4.4 (Guindon & Gascuel, 2003). Non-significant P values (‡ 0.273) using the Shimodaira–Hasegawa test (Shimodaira & Hasegawa, 1999) in consel version 1.19 (Shimodaira & Hasegawa, 2001) suggested that the three data sets were compatible. A single concatenated data set was also included in subsequent phylogenetic analyses. Different maximum likelihood models and parameters were obtained for each data set (Table 3), and MrBayes 3.1.2 (Ronquist & Huelsenbeck, 2003) was used to estimate a phylogeny using all available molecular data while accounting for separate models of sequence evolution. The values for model parameters were allowed to vary between data sets and were not specified a priori but treated as unknown variables to be estimated in each analysis. Bayesian analysis was conducted with random starting trees without constraints. Four simultaneous Markov chains were run for 1,000,000 generations, and trees were sampled every 100 generations. The burn in, or point of stationarity, was determined by plotting observed likelihood scores against number of generations. Analyses were repeated in four separate runs of MrBayes to ensure that trees converged on the same topology.

Journal of Biogeography 34, 364–375 ª 2006 The Author. Journal compilation ª 2006 Blackwell Publishing Ltd

Journal of Biogeography 34, 364–375 ª 2006 The Author. Journal compilation ª 2006 Blackwell Publishing Ltd

AD 493

AMNH 261760 MVZ 185903

MVZ 185904 Sphaeronycteris toxophyllum Sphaeronycteris toxophyllum Stenoderma rufum rufum Stenoderma rufum rufum

Pygoderma bilabiatum bilabiatum

Pygoderma bilabiatum magna Pygoderma bilabiatum bilabiatum

Phyllops falcatus haitiensis

Centurio senex senex Centurio senex senex Centurio senex greenhalli Phyllops falcatus falcatus Phyllops falcatus falcatus Phyllops falcatus falcatus Phyllops falcatus falcatus Phyllops falcatus haitiensis

Ametrida centurio Ariteus flavescens Ariteus flavescens Ariteus flavescens Centurio senex senex Centurio senex senex Centurio senex senex

Taxon Near Sinnamary, Paracou, French Guiana Upper entrance Windsor Great Cave, Trelawney, Jamaica Portland Cave 9, Portland Cottage, Clarendon, Jamaica Monarva Cave, Revival, Westmoreland, Jamaica Rio Uyus, 5 km E San Cristo´bal Acasaguastlan, Guatemala Biotope Cerro Cahui, El Remate, Guatemala 44 km S Constitucio´n, 44 km S and 70 km E Esca´rcega, Campeche, Mexico Ojo de Agua Rio de Atoyac, Veracruz, Mexico El Imposible, El Refugio, El Salvador St Ann’s Ward, Port of Spain, Trinidad Lower Valley Forest, Grand Cayman Lower Valley Forest, Grand Cayman Lower Valley Forest, Grand Cayman Lower Valley Forest, Grand Cayman La Entrada (de Cabrera), Marı´a Trinidad Sa´nchez, Dominican Republic Finca Don Miguel, Platanal (de Cotuı´), Sa´nchez Ramı´rez, Dominican Republic San Rafael de Amboro´, Santa Cruz, Bolivia Fazenda Santa Monica, Municipio Itatiaia, Rio de Janeiro, Brazil Fazenda Guaricana, Grupo Bamerindus, Municipio Guaratuba, Parana, Brazil Finca Aroa, 2 km W Choronı´, Aragua, Venezuela Independencia, Pando, Bolivia Vieques, Puerto Rico Mata de Pla´tano, Arecibo, Puerto Rico

Locality

AY604452 AY604451 AY604431 AY604432

AY604438

AY604448 DQ914959 AY604439 AY604437

AY604441 AY604443 AY604445 AY604447 AY604453 DQ211651 AY604450 AY604449

AY604446 AY604434 AY604435 AY604436 AY604442 AY604440 AY604444

GenBank accession number

b b b

b b b, 1 kb b

b b b b b b b

Cytochrome Cytochrome Cytochrome Cytochrome

b, 850 bp b, 850 bp b b

Cytochrome b

Cytochrome b Rag2 Cytochrome b Cytochrome b

Cytochrome Cytochrome Cytochrome Cytochrome Rag2 Cytochrome Cytochrome Cytochrome

Cytochrome Cytochrome Cytochrome Cytochrome Cytochrome Cytochrome Cytochrome

Gene, bp

AMNH, American Museum of Natural History; AMCC, Ambrose Monell Cryogenic Collection at the AMNH; ROM, Royal Ontario Museum; TTU, The Museum, Texas Tech University; MVZ, Museum of Vertebrate Zoology, University of California at Berkeley.

AMCC 102376 AMCC 122019

AD 275

AMNH 275512

AMNH 194213 AMNH 262637

AMCC 103067

AMNH 275484

TK13110 F35508

CMNH 55731 ROM 101330 AMNH 256330

101762 101769 101813 101852 103030

AMCC 110324 AMCC 102694 AMCC 102767 AMCC 102761 F34031 FN32253 FN29530

AMNH 267973 AMNH 274601 AMNH 274616 AMNH 274610 ROM 99672 ROM 99584 ROM 95739

AMCC AMCC AMCC AMCC AMCC

Tissue voucher

Museum voucher

Table 2 List of sequences generated for this article, with number of base pairs amplified, when applicable. Tissues without a museum voucher were sampled by wing puncture, identified and released in the field. When no tissue voucher is listed, DNA was extracted from parts of museum specimens.

Caribbean radiation of frugivorous bats

367

368

GTR + I + C GTR + I GTR + I cyt b exemplars Bayes Rag2 ML Rag2 Bayes

ML, parameters used in maximum likelihood analyses; Bayes, parameters obtained in Bayesian analysis of concatenated data; GTR, general time reversible model; R-matrix, rate matrix parameter (with respect to G–T transversion); a, shape parameter; I, proportion of invariant sites; C, gamma distribution for variable substitution rates; )2 log K ¼ 2( log L1 ) log L2), where L1 is likelihood without clock and L2 is likelihood with clock. Bayesian parameters are followed by the 95% confidence interval (in parentheses).

0.6850 8 0.4880 (0.1–0.6) 0.9093 0.9137 (0.9–0.9) 2.0 (0.2–4.1) – –

5.6616

0.5788 0.4531 24 8 22.0066 7.801

0.7981 7 3.8388

0.6231 0.6281 (0.4–0.7) 0.3610 0.5175 0.8524 5.7 (0.3–74.6) 0.428 0.7357 + + + +

I I I I

+ + + + GTR GTR GTR GTR mtrDNA ML mtrDNA Bayes cyt b all individuals cyt b exemplars ML

C C C C

3.1, 9.0, 2.8, 0.07, 53.7 3.4, 9.8, 3.0, 0.4, 56.2 4.4, 39.9, 5.1, 2.2, 70.0 8.8 · 102, 6.0 · 103, 1.1 · 103, 1.4 · 102, 1.6 · 104 4.6, 30.0, 5.5, 1.0, 77.2 0.7, 3.8, 0.0, 1.1, 3.8 0.5, 2.6, 0.1, 0.9, 3.1

I a R-matrix Model Data

Table 3 Models of molecular evolution and parameters selected for each molecular data set. See Table 1 for sequences

)2 log K

d.f.

P

L. M. Da´valos Biogeographical analyses The trees obtained by analysing the data generated in this study were compared with those compatible with the single-origin hypothesis for Caribbean Short-faced bats (Fig. 1) (Baker & Genoways, 1978; Koopman, 1989; Tavares & Simmons, 2000; Morgan, 2001; Da´valos, 2004). Two nonparametric tests, the Shimodaira–Hasegawa test (Shimodaira & Hasegawa, 1999) and the approximately unbiased test (Shimodaira, 2002), were performed using consel. The geographical distribution of Short-faced species was coded as a discrete character and optimized on the resulting trees using MacClade version 4.06 (Maddison & Maddison, 2003). A parametric bootstrap approach (Huelsenbeck & Crandall, 1997) was used to estimate confidence intervals around branch lengths, and evaluate the relative timing of divergence. First, the molecular clock hypothesis was evaluated for each data set using a likelihood ratio test (Goldman, 1993). If the molecular clock was not rejected, the optimal topology obtained for each data set and optimal model parameters (Table 3) were used to simulate 100 data sets in seq-gen version 1.3.2 (Rambaut & Grassly, 1997). paup* was then used to optimize topologies and rate-constant branch lengths for each of the simulated data sets. The resulting branch lengths were then tabulated and used to calculate the 95% confidence interval around the nodes of interest. RESULTS Sequence variation and saturation analysis

mtrDNA Mitochondrial ribosomal sequences were downloaded for all taxa with the exception of Phyllops, which was coded as ‘all missing’ for this partition in concatenated analyses. Alignment of 12S rRNA, tRNAval and 16S rRNA sequences using mafft resulted in 2616 aligned positions. Comparison with the alignment obtained with BlastAlign helped identify 15 sites with potentially ambiguous alignment, but visual inspection clarified the positional homology of the sites in question. Phylogenetic analyses with and without the 15 sites resulted in identical trees, with comparable support measures, so no positions were excluded from combined analyses. The average base composition of sequences was skewed, with deficiency of guanine (17.6%) and overabundance of adenine (35.5%). This bias in base composition did not differ significantly across taxa (v2 test implemented in paup*, P ¼ 1.000).

Cytochrome b Complete cytochrome b sequences were obtained for all taxa with the exception of Sphaeronycteris, for which only 850 bp were amplified and sequenced. A few individuals per species were included in the cytochrome b analysis to assess Journal of Biogeography 34, 364–375 ª 2006 The Author. Journal compilation ª 2006 Blackwell Publishing Ltd

Caribbean radiation of frugivorous bats 1

Between Short-Faced bats and Artibeus

2

Between Short-Faced genera Figure 2 Scatter plot of cytochrome b genetic divergence and taxonomic differentiation for the taxa in this study. Numerals indicate cytochrome b distance outliers as follows: 1, between Artibeus and Sphaeronycteris and between Artibeus and Pygoderma from Brazil; 2, between Phyllops and Stenoderma; 3, between subspecies of Pygoderma and between subspecies of Phyllops; 4, between Sphaeronycteris individuals.

3

Between subspecies within species

4

Among individuals within subspecies

intraspecific variation within short-faced species. Two Phyllops falcatus [Ambrose Monell Cryogenic Collection at the AMNH (AMCC) nos 101762 and 101813] individuals had identical sequences. A summary of the uncorrected pairwise divergences among sequenced individuals is shown in Fig. 2. Most substitutions in cytochrome b were synonymous, and translation of sequences to amino acids led to a matrix with only 89 variable sites. The average base composition of sequences was skewed, with little bias at first and second codon positions, a deficiency of adenine (20.0%) and guanine (14.5%) and an overabundance of thymine (40.6%) for second position, and a strong bias in third position: deficiency of guanine (2.3%) and thymine (18.1%), and abundance of adenine (41.8%) and cytosine (37.7%). That bias in base composition did not differ significantly across taxa (v2 test implemented in paup*, P ¼ 1.00). The same result is obtained for the complete cytochrome b gene, and first and second positions separately. Saturation curves indicated that no codon position experienced multiple transition or transversion substitutions.

Rag2 Fragments of Rag2 (1.3 kb) from two Phyllops falcatus individuals (AMCC nos 101769 and 103067, Table 2) were amplified to complete the Rag2 data set published by Baker et al. (2000). The sequences obtained were identical. Cytochrome b and Rag2 sequences aligned without gaps, as expected for protein-coding genes, and no stop codons were found when the matrix was converted to amino acids using MacClade. The average base composition of Rag2 sequences was skewed, with a deficiency of cytosine (17.6%) and thymine (20.5%) at first codon position, a deficiency of guanine (17.9%) for second position, and a deficiency of guanine (19.3%) in third position. The skewed base composition does not differ significantly across taxa (v2 test implemented in paup*, P ¼ 1.000).

0

0.05

0.1

0.15

0.2

Uncorrected pairwise divergence

Phylogenetic analyses Three data sets were analysed separately and concatenated subsequently: (1) mtrDNA, (2) cytochrome b with single and multiple individuals per species, and (3) a fragment of the nuclear Rag2. The models of sequence evolution and parameters obtained for each data set and for the concatenated matrix are shown in Table 3. The phylogeny resulting from analysis of cytochrome b is shown in Fig. 3, and the result of analysis of Rag2 is shown in Fig. 4a. Stationarity in MrBayes runs was reached after 25,000 generations (burn in ¼ 250 trees) and a 50% majority-rule consensus of resulting trees, along with measures of support from analysis of mtrDNA, is presented in Fig. 4b. Separate and combined analyses supported four nodes (Figs 3 & 4): the monophyly of Short-faced bats and of continental Short-faced bats, the sister relationship between Ariteus and Ardops, and the sister relationship between Phyllops and Stenoderma. Previous analyses of morphological and molecular data supported the first three nodes (Lim, 1993; Baker et al., 2000, 2003; Tavares & Simmons, 2000; Wetterer et al., 2000), but the sister-group relationship between Phyllops and Stenoderma is a novel result. This last result had high measures of support across the two loci where Phyllops was represented, and in combined analyses as well (Figs 3 & 4b). Biogeographical analyses Four trees were compared using the Shimodaira–Hasegawa and approximately unbiased tests (Table 4): the two result trees of Fig. 4, and the two trees resulting from previous analyses of morphological data shown in Fig. 1 (Lim, 1993; Wetterer et al., 2000). The approximately unbiased test detected significant incongruence between mitochondrial and nuclear partitions (Table 4). The incongruent nodes, however, have low support values in the Rag2 phylogeny, and the node with Bayesian posterior probability of 0.56 seems to

Journal of Biogeography 34, 364–375 ª 2006 The Author. Journal compilation ª 2006 Blackwell Publishing Ltd

369

L. M. Da´valos 69 100

Pygoderma Brazil, Bolivia

100 Ametrida French Guiana

69 98

Sphaeronycteris Venezuela, Bolivia

68 77

80

93

Centurio Central America, Mexico

100 83

Centurio Trinidad

87

Phyllops Grand Cayman

100 Phyllops Dominican Republic

99 100 99

63 100

Stenoderma Puerto Rico

Ardops St. Eustatius, Nevis, St. Kitts

98 99 100

Ariteus Jamaica Artibeus Dermanura

0.01 substitutions/site

(a)

(b) 0.005 substitutions/site

0.01 substitutions/site Pygoderma 1.00/65 Ametrida

24

1.00/93

57

1.00/67

Sphaeronycteris

67 36

Centurio

0.56/47

Ardops

Stenoderma

Ariteus

Phyllops

1.00/NE

98 100

Stenoderma

Ardops

Phyllops

Ariteus

1.00/75

1.00/100

98

Artibeus Dermanura

have been undermined by this conflict in phylogenetic signal (Fig. 4). The trees obtained from separate or combined analyses imply equivalent biogeographical relationships between the islands and the continent, as shown in Fig. 4. Confidence intervals around the branch lengths of biogeographically important nodes are shown in Fig. 5. 370

Figure 3 Maximum likelihood phylogram () ln L ¼ 4370.38321) of all available cytochrome b sequences obtained using the model and parameters specified in Table 3. Values above branches are percentages of 1000 maximum likelihood bootstrap pseudoreplicates.

Figure 4 Optimization of geographical distribution on resulting trees. Branches optimizing to the Antilles are shown in black, those optimizing to the continent are shown in grey. (a) Maximum likelihood phylogram () ln L ¼ 2430.98215) of Rag2 sequences obtained using the model and parameters specified in Table 3. (b) Majority-rule consensus phylogram of trees obtained from Bayesian analyses of concatenated molecular data. Values above branches are, from left to right: first, Bayesian posterior probability, and second, percentage of 1000 maximum likelihood bootstrap pseudoreplicates of mtrDNA sequences. NE, not estimated. Table 3 shows the models and parameters used for separate and concatenated analyses.

DISCUSSION Short-faced bat species and subspecies Sequence divergence in cytochrome b was proportional to taxonomic rank (Fig. 2), with two exceptions: the divergence between the two Sphaeronycteris individuals and the divergence

Journal of Biogeography 34, 364–375 ª 2006 The Author. Journal compilation ª 2006 Blackwell Publishing Ltd

Caribbean radiation of frugivorous bats Table 4 Significance of fit of each data set to trees obtained from optimization of individual partitions using the models of Table 3, concatenated molecular data, or from previous studies

Data set

Test

Rag2 ML tree

Molecular Bayes tree

Lim (1993) (Fig. 1b)

Wetterer et al. (2000) (Fig. 1c)

mtrDNA

SH AU SH AU SH AU

0.261 0.005 0.369 0.001 – –

0.834 0.427 0.900 0.885 0.341 0.013

< < < < < <

0.009 < 0.001 0.039 < 0.001 0.052 0.002

cyt b Rag2

0.001 0.001 0.001 0.001 0.001 0.001

AU, approximately unbiased test (Shimodaira, 2002); ML, maximum likelihood; SH, Shimodaira–Hasegawa (1999) test.

Mitochondrial sequence divergence (substitutions/site/lineage)

(a)

0.3 0.25 0.2 0.15 0.1 0.05

Rag2 sequence divergence (substitutions/site/lineage)

(b)

Phyllops /Stenoderma

Ardops /Ariteus

Centurio /other cont. Stenodermatina

Ardops-Ariteus / other Stenodermatina

Phyllops-Stenoderma/ other Stenodermatina

Between Artibeus and Stenodermatina

0

0.015

between Stenoderma and Phyllops. Sampling within Sphaeronycteris was insufficient to determine whether there is more than one population in this taxon, and future studies should examine this question. Despite the low sequence divergence between Stenoderma and Phyllops in the mitochondrial cytochrome b gene, divergence in Rag2 was comparable to that found between Ardops and Ariteus (Fig. 5). This suggests the mitochondrial marker has evolved more slowly along the Stenoderma–Phyllops branch than along the Ardops–Ariteus branch, although this difference in rate was not significant (P ‡ 0.453, Table 3). Of three polytypic species studied here – Centurio senex, Pygoderma bilabiatum and Phyllops falcatus – only the subspecies of Centurio had greater cytochrome b divergence than individuals in monotypic species (Fig. 2). By this measure, gene flow has been maintained between geographically disjunct populations within Pygoderma and Phyllops. This last result, and the lack of distinct morphological characters to distinguish subspecies of Phyllops (Timm & Genoways, 2003), supports the hypothesis that the extant population in the Grand Cayman (and by extension Cuba) is conspecific with its Hispaniolan counterpart (Sua´rez & Dı´azFranco, 2003).

0.01

Historical biogeography of the Short-faced bats 0.005

Phyllops /Stenoderma

Ardops /Ariteus

Pygoderma /other cont. Stenodermatina

Ardops-Ariteus/ other Stenodermatina

Phyllops-Stenoderma/ other Stenodermatina

Between Artibeus and Stenodermatina

0

Figure 5 Confidence intervals (95%) around observed sequence divergence resulting from parametric bootstrapping of rate-constant phylogenies of Short-faced bats. (a) Estimates of divergence for mitochondrial ribosomal DNA (black diamonds) and the cytochrome b gene (white diamonds). (b) Estimates of divergence for nuclear Rag2. Cont., continental.

Previous studies of Short-faced bats biogeography assumed the monophyly of a Caribbean clade, and concluded that the common ancestor of this clade reached the islands from the continent through either vicariance or dispersal (Baker & Genoways, 1978; Koopman, 1989; Lim, 1993; Morgan, 2001; Da´valos, 2004). This study investigated two phylogenies consistent with this biogeographical scenario (Fig. 1), and these were both significantly incompatible with every molecular data set available (Table 4). In contrast, all phylogenies derived from the molecular data suggest that the continental Short-faced bats are the ones that share a most recent common ancestor (Figs 3 & 4; see also Baker et al., 2003). The molecular phylogenies support an alternative biogeographical hypothesis: the Short-faced bats differentiated in the Caribbean from a common ancestor of continental origin, and, following diversification into two distinct lineages on the islands, a single lineage colonized the continent and gave rise to the genera that today range from Mexico to central South America (Table 1).

Journal of Biogeography 34, 364–375 ª 2006 The Author. Journal compilation ª 2006 Blackwell Publishing Ltd

371

L. M. Da´valos A single ancient Short-faced fossil found on the continent would overturn the biogeographical scenario proposed here. As it stands, the fossil record is compatible with the Caribbean radiation hypothesis. The earliest fossils clearly corresponding to phyllostomids are relatives of extant nectar-feeding and animal-feeding lineages from the middle Miocene of La Venta in Colombia (Czaplewski et al., 2003b). These are only distantly related to Short-faced bats, and their distribution shows that phyllostomids have been living in north-western South America since at least 12–13 million years ago (Ma). Fossils of Short-faced bats, in contrast, are only known from the Quaternary (maximum 1.8 Ma), on both the islands and the continent. Centurio fossils have been found in Mexico and Central America (reviewed by Czaplewski et al., 2003a), and three extinct Short-faced species are known from Cuba (Table 1). Of the latter, Cubanycteris displays primitive characters when compared with other Short-faced bats, despite sharing the diagnostic features of the group (Mancina & Garcia-Rivera, 2005). Although this fossil is not ancient, if its phylogenetic position were at the base of the Short-faced tree it would represent another basal Caribbean lineage and bolster the biogeographical hypothesis advanced here. This possibility remains to be tested, however, as to date no phylogenetic analysis of morphology has included the recently described Cubanycteris. The molecular analyses also provide some evidence on how the biogeography of the Short-faced bats relates to their differentiation and diversification, and the relative timing of these events. The divergence between Short-faced bats and their sister taxa is significantly greater than those corresponding to subsequent diversification events for all markers (Fig. 5). In contrast, the divergences between the two Caribbean branches – Ardops and Ariteus or Phyllops and Stenoderma – and their sister taxa, are almost indistinguishable, particularly with the mitochondrial markers (Fig. 5). Assuming that these divergences are proportional to time, the results indicate that the common ancestor of Short-faced bats reached the Caribbean long before the rapid diversification of the group into two insular branches, and subsequent colonization of the continent. The rapid diversification events that lead to extant branches in the Short-faced phylogeny might explain the lack of characters and low support for several internal branches in the phylogeny using morphological data (Lim, 1993; Wetterer et al., 2000), as well as the slowly-evolving nuclear marker (Fig. 4a). Simply, there was not enough time in between diversification events for differences between populations to accumulate and be easily discernible today. But how did the Short-faced bats reach the Antilles in the first place? Two recent studies proposed that Short-faced bats reached the Caribbean from northern South America through a land bridge connecting the continent with parts of Cuba, Hispaniola and Puerto Rico (Tavares & Simmons, 2000; Da´valos, 2004). The land bridge is thought to have disappeared in the early Oligocene, some 33 Ma (Iturralde-Vinent & MacPhee, 1999). This provides an approximate date for divergence between populations isolated from each other at 372

opposite ends of the bridge. The 95% confidence interval for the cytochrome b divergence between Artibeus and Shortfaced bats is 0.216–0.254 substitutions per site per lineage (Fig. 5a). This same measure for the divergence between humans and chimpanzees is 0.075–0.08 (Castresana, 2001), corresponding to 4–6 Myr (see Yoder & Yang, 2000). Assuming that rates of molecular evolution for this gene are roughly comparable across different mammal lineages, then Short-faced bats and their closest living relatives were separated 10.8–20.3 Ma. It could be that this gene evolves much faster in primates than in Short-faced bats, skewing the time estimate for bats downward. For the molecular divergence between Artibeus and Short-faced bats to correspond to 33 Ma, one would have to postulate that molecular evolution in bats is two to three times slower than in primates. Such differences in rates are known in mammals (Yoder & Yang, 2000), but cytochrome b actually appears to evolve more rapidly in bats – including phyllostomids – than among other mammals (Ditchfield, 2000). A faster rate would make the node in question more recent rather than older. Further, the divergence dates obtained here are consistent with analyses that use ingroup and outgroup fossil calibrations and do not rely on a single molecular clock across different loci and branches (Teeling et al., 2005). These analyses place the split between Artibeus and the distantly related nectar-feeding Anoura at 15–23 Ma. The evidence from multiple, independent studies shows that the ancestor of all Short-faced bats could not have used the Oligocene land bridge. This leaves over-water dispersal as an alternative mechanism to explain the biogeography of Short-faced bats. Two dispersal routes have been proposed; from Mexico and/or Central America to Cuba and/or Jamaica (Koopman, 1989; Morgan, 2001), or from northern South America north through the Lesser Antilles (Baker & Genoways, 1978) (Fig. 1a). The molecular phylogenies are not decisive on this point; the node separating a basal Short-faced branch from other taxa in the group is the least supported in combined analyses (Fig. 4b). Further, each of the likely basal branches is represented in the western Greater Antilles: Ariteus in Jamaica and Phyllops in Cuba (by one extant and two fossil species; see Sua´rez & Dı´az-Franco, 2003). That is, either phylogenetic hypothesis is compatible with dispersal from Central America. Regardless of the route used, dispersal as a biogeographical mechanism raises questions about what special circumstances might explain a single improbable colonization event, or, conversely, how populations become isolated despite their ability to fly over water. One of the first quantitative studies in Caribbean bat biogeography proposed that drops in sea level in the Pleistocene facilitated dispersal to the islands (Griffiths & Klingener, 1988). The 10.8–20.3 Myr estimate (see above) places the dispersal of a Short-faced bats ancestor well before the Pleistocene. Drops in sea level, however, have not only occurred over the last 2 Myr of Earth history: the transitions from early to middle Miocene (c. 16 Ma) and from middle to late Miocene (c. 11 Ma) were marked by relatively low sea

Journal of Biogeography 34, 364–375 ª 2006 The Author. Journal compilation ª 2006 Blackwell Publishing Ltd

Caribbean radiation of frugivorous bats levels (Haq et al., 1993) and are compatible with the molecular estimate of divergence. Still, facilitated dispersal is not the same as a land bridge, and neither Short-faced bats nor their sister taxa are known for their exceptional dispersal abilities. The home range of Stenoderma at Luquillo forest was found to be below 3 ha on average (Gannon et al., 2005), and Artibeus jamaicensis flew a maximum of 1 km beyond its roost each night on Barro Colorado Island, Panama (Morrison, 1978). Over geological time, however, both Ardops and Artibeus jamaicensis seem capable of maintaining gene flow among islands separated by as much as 100 km (Carstens et al., 2004). These genetic data, along with the geographical distribution of Phyllops (which includes the relatively remote Grand Cayman), suggest these bats are capable of over-water dispersal well beyond their home range. This ability, perhaps in combination with drops in sea level, enabled both the original colonization of the Antilles and the subsequent dispersal to the continent. Traditionally the colonization of a continent by an insular lineage has been considered rare, and subsequent diversification is thought to be even rarer. Several explanations based on faunal saturation on the mainland and competitive disadvantage of island species have been advanced to explain this pattern (reviewed by Nicholson et al., 2005). The model of one-way colonization from continent to islands, and the ecological hypotheses designed to explain it, must be revised as more phylogenies uncover two-way invasions in places as varied as Polynesia and New Guinea (Filardi & Moyle, 2005) or Madagascar and Africa (Raxworthy et al., 2002). One group of Caribbean vertebrates, Anolis lizards, has successfully colonized the Americas (Glor et al., 2005; Nicholson et al., 2005). A vast radiation of several dozen Anolis species is descended from a single Caribbean colonizer to the continental Neotropics (Nicholson et al., 2005), and one species in Florida traces its origin back to Cuba and prior to the Pleistocene (Glor et al., 2005). Among Caribbean bats the molecular data suggest a complex history of two-way invasions between the islands and the continent, starkly at odds with biogeographical convention (Baker & Genoways, 1978; Koopman, 1989). The ancestor of Short-faced bats reached the Antilles long before the Pleistocene through an undetermined route, differentiated over a relatively long period of isolation and then quickly diversified on the islands. Shortly thereafter a single Shortfaced lineage reached the mainland and subsequently expanded its range, eventually giving rise to four distinct continental genera. The pattern of reverse colonization proposed here has been described for two groups of aerial insectivores, natalids (Da´valos, 2005) and probably Mormoops (Da´valos, 2006); and for the lineage comprising the flower-visiting phyllostomids Glossophaga and Leptonycteris (Genoways et al., 2005; Da´valos, in press). The Short-faced radiation of strictly frugivorous bats completes a dynamic model of Caribbean biogeography that encompasses an entire biological community.

ACKNOWLEDGEMENTS This article is based upon work supported by the National Science Foundation under grant no. 0206336. It is a contribution from the Monell Molecular Laboratory and the Cullman Research Facility in the Department of Ornithology, American Museum of Natural History, and has received generous support from the Lewis B. and Dorothy Cullman Program for Molecular Systematics Studies, a joint initiative of The New York Botanical Garden and the American Museum of Natural History. This research is supported by the Department of Mammalogy of the American Museum of Natural History, the NASA grant no. NAG5-8543 to the Center for Biodiversity and Conservation at the American Museum of Natural History, and the Department of Ecology, Evolution and Environmental Biology at Columbia University. Fieldwork was supported by the Ambrose Monell Cryogenic Collection and the Department of Mammalogy, both at the American Museum of Natural History, the Center for Environmental Research and Conservation and the Department of Ecology, Evolution and Environmental Biology, both at Columbia University, Elizabeth Dumont’s NSF grant, and the Explorers’ Club (New York). The author is currently supported by grants to Susan Perkins for the study of malaria parasites. For comments, collecting permits, field or lab assistance, tissue loans and/or other intangible support I thank R. Baker (TTU), F.K. Barker, A.S.P. Corthals, J.L. Cracraft, M. Delarosa, R. DeSalle, A. Donaldson, K. Doyle, E. Dumont, M. Engstrom (ROM), R. Eriksson, J. Feinstein, N. Gyan, R. Harbord (BMNH), S.A. Jansa, S. Koenig, J. Mercedes, J.C. Morales, G.S. Morgan, T. Nicole, J.L. Patton (MVZ), S. Pedersen (SDSU), A.L. Porzecanski, C. Raxworthy, A. Rodrı´guez, R.O. Sa´nchez, P. Schickler, M. Schwartz, N.B. Simmons (AMNH), V. Tavares, A. Tejedor and A. Wright. REFERENCES Baker, R.J. & Genoways, H.H. (1978) Zoogeography of Antillean bats. Zoogeography of the Caribbean. Academy of Natural Sciences of Philadelphia Special Publication 13, pp. 53–97. Baker, R.J., Porter, C.A., Patton, J.C. & Van Den Bussche, R.A. (2000) Systematics of bats of the family Phyllostomidae based on RAG2 DNA sequences. Occasional Papers of the Museum of Texas Tech University 202, pp. 1–16. Baker, R.J., Porter, C.A., Hoofer, S.R. & Van Den Bussche, R.A. (2003) A new higher-level classification of New World leafnosed bats based on nuclear and mitochondrial DNA sequences. Occasional Papers of the Museum of Texas Tech University 230, pp. 1–32. Belshaw, R. & Katzourakis, A. (2005) BlastAlign: a program that uses BLAST to align problematic nucleotide sequences. Bioinformatics, 21, 122–123. Carstens, B.C., Sullivan, J., Da´valos, L.M., Larsen, P.A. & Pedersen, S.C. (2004) Exploring population genetic struc-

Journal of Biogeography 34, 364–375 ª 2006 The Author. Journal compilation ª 2006 Blackwell Publishing Ltd

373

L. M. Da´valos ture in three species of Lesser Antillean bats. Molecular Ecology, 13, 2557–2566. Carter, C.H., Genoways, H.H., Loregnard, R.S. & Baker, R.J. (1981) Observations on bats from Trinidad, with a checklist of species occurring on the island. Occasional Papers of the Museum of Texas Tech University 72, pp. 1–27. Castresana, J. (2001) Cytochrome b phylogeny and the taxonomy of Great Apes and mammals. Molecular Biology and Evolution, 18, 465–471. Czaplewski, N.J., Krejca, J. & Miller, T.E. (2003a) Late Quaternary bats from Cebada Cave, Chiquibul Cave System, Belize. Caribbean Journal of Science, 39, 23–33. Czaplewski, N.J., Masanaru, T., Naeher, T.M., Shigehara, N. & Setoguchi, T. (2003b) Additional bats from the middle Miocene La Venta fauna of Colombia. Revista de la Academia Colombiana de Ciencias Fı´sicas, Exactas y Naturales, 27, 263–282. Da´valos, L.M. (2004) Phylogeny and biogeography of Caribbean mammals. Biological Journal of the Linnean Society, 81, 373–394. Da´valos, L.M. (2005) Molecular phylogeny of funnel-eared bats (Chiroptera: Natalidae), with notes on biogeography and conservation. Molecular Phylogenetics and Evolution, 37, 91–103. Da´valos, L.M. (2006) The geography of diversification in the mormoopids (Chiroptera: Mormoopidae). Biological Journal of the Linnean Society, 88, 101–118. Da´valos, L.M. (in press) Earth history and the evolution of Caribbean bats. Evolution, ecology, and conservation of island bats (ed. by T.H. Fleming and P.A. Racey), University of Chicago Press, Chicago, IL. Da´valos, L.M. & Jansa, S.A. (2004) Phylogeny of the Lonchophyllini (Chiroptera: Phyllostomidae). Journal of Mammalogy, 85, 404–413. Ditchfield, A. (2000) The comparative phylogeography of Neotropical mammals: patterns of intraspecific mitochondrial DNA variation among bats contrasted to nonvolant small mammals. Molecular Ecology, 9, 1307– 1318. Emmons, L.H. (1997) Neotropical rainforest mammals: a field guide, 2nd edn. University of Chicago Press, Chicago, IL. Filardi, C.E. & Moyle, R.G. (2005) Single origin of a panPacific bird group and upstream colonization of Australasia. Nature, 438, 216–219. Gannon, M.R., Kurta, A., Rodriguez Duran, A. & Willig, M.R. (2005) Bats of Puerto Rico: an island focus and a Caribbean perspective. Texas Tech University Press, Lubbock, TX. Genoways, H.H., Baker, R.J., Bickham, J.W. & Phillips, C.J. (2005) Bats of Jamaica. Special Publications of the Museum of Texas Tech University 48, 1–155. Glor, R.E., Losos, J.B. & Larson, A. (2005) Out of Cuba: overwater dispersal and speciation among lizards in the Anolis carolinensis subgroup. Molecular Ecology, 14, 2419– 2432. Goldman, N. (1993) Statistical tests of models of DNA substitution. Journal of Molecular Evolution, 36, 182–198. 374

Goodwin, G.G. & Greenhall, A.M. (1961) A review of the bats of Trinidad and Tobago. Bulletin of the American Museum of Natural History, 122, 187–302. Greenbaum, I.F., Baker, R.J. & Wilson, D.E. (1975) Evolutionary implications of the karyotypes of the stenodermine genera Ardops, Ariteus, Phyllops, and Ectophylla. Bulletin of the Southern California Academy of Science, 74, 156–159. Griffiths, T.A. & Klingener, D. (1988) On the distribution of Greater Antillean bats. Biotropica, 20, 240–251. Guindon, S. & Gascuel, O. (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology, 52, 696–704. Handley, C.O. (1976) Mammals of the Smithsonian Venezuelan Project. Brigham Young University Science Bulletin– Biological Series, 20, 1–89. Haq, B.U., Hardenbol, J. & Vail, P.R. (1993) Chronology of fluctuating sea levels since the Triassic. Science, 235, 1156– 1167. Howe, H.F. (1974) Additional records of Phyllonycteris aphylla and Ariteus flavescens from Jamaica. Journal of Mammalogy, 55, 662–663. Huelsenbeck, J.P. & Crandall, K.A. (1997) Phylogeny estimation and hypothesis testing using maximum likelihood. Annual Review of Ecology and Systematics, 28, 437–466. Iturralde-Vinent, M.A. & MacPhee, R.D.E. (1999) Paleogeography of the Caribbean region: implications for Cenozoic biogeography. Bulletin of the American Museum of Natural History, 238, 1–95. Iudica, C.A., Whitten, W.M. & Williams, N.H. (2001) Small bones from dried mammal museum specimens as a reliable source of DNA. BioTechniques, 30, 733–736. Katoh, K., Kuma, K.-i., Toh, H. & Miyata, T. (2005) MAFFT version 5: improvement in accuracy of multiple sequence alignment. Nucleic Acids Research, 33, 511–518. Koopman, K.F. (1989) A review and analysis of the bats of the West Indies. Biogeography of the West Indies, past, present, and future (ed. by C.A. Woods), pp. 635–644. Sandhill Crane Press, Gainesville, FL. Koopman, K. (1994) Chiroptera: systematics. Handbuch der Zoologie, 8, 1–217. Lim, B.K. (1993) Cladistic reappraisal of neotropical stenodermatine bat phylogeny. Cladistics, 9, 147–165. Maddison, D.R. & Maddison, W.P. (2003) MacClade 4: analysis of phylogeny and character evolution, Ver. 4.06. Sinauer Associates, Sunderland, MA. Mancina, C.A. & Garcia-Rivera, L. (2005) New genus and species of fossil bat (Chiroptera: Phyllostomidae) from Cuba. Caribbean Journal of Science, 41, 22–27. Morgan, G. (2001) Patterns of extinction in West Indian bats. Biogeography of the West Indies (ed. by C.A. Woods and F.E. Sergile), pp. 369–407. CRC Press, Boca Raton, FL. Morrison, D.W. (1978) Foraging ecology and energetics of the frugivorous bat Artibeus jamaicensis. Ecology, 59, 716–723. Nicholson, K.E., Glor, R.E., Kolbe, J.J., Larson, A., Hedges, S.B. & Losos, J.B. (2005) Mainland colonization by island lizards. Journal of Biogeography, 32, 929–938.

Journal of Biogeography 34, 364–375 ª 2006 The Author. Journal compilation ª 2006 Blackwell Publishing Ltd

Caribbean radiation of frugivorous bats Pedersen, S.C., Genoways, H.H. & Freeman, P.W. (1996) Notes on bats from Montserrat (Lesser Antilles) with comments concerning the effects of Hurricane Hugo. Caribbean Journal of Science, 32, 206–213. Phillips, C.J., Pumo, D.E., Genoways, H.H. & Ray, P.E. (1989) Caribbean island zoogeography: a new approach using mitochondrial DNA to study Neotropical bats. Biogeography of the West Indies: past, present, and future (ed. by C.A. Woods), pp. 661–684. Sandhill Crane Press, Gainesville, FL. Phillips, C., Pumo, D., Genoways, H., Ray, P. & Briskey, C. (1991) Mitochondrial DNA evolution and phylogeography in two neotropical fruit bats, Artibeus jamaicensis and Artibeus lituratus. Latin American mammalogy: history, biodiversity, and conservation (ed. by M. Mares and D. Schmidly), pp. 97–123. University of Oklahoma Press, Norman, OK. Posada, D. & Crandall, K.A. (1998) MODELTEST: testing the model of DNA substitution. Bioinformatics, 14, 817–818. Pumo, D.E., Goldin, E.Z., Elliot, B., Phillips, C. & Genoways, H.H. (1988) Mitochondrial DNA polymorphism in three Antillean Island populations of the fruit bat, Artibeus jamaicensis. Molecular Biology and Evolution, 5, 79–89. Pumo, D.E., Iksoo, K., Remsen, J., Phillips, C.J. & Genoways, H.H. (1996) Molecular systematics of the fruit bat, Artibeus jamaicensis: origin of an unusual island population. Journal of Mammalogy, 77, 491–503. Rambaut, A.E. & Grassly, N.C. (1997) SEQ-GEN: an application for the Monte Carlo simulation of DNA sequence evolution along phylogenetic trees. Computer Applications in the Biosciences, 13, 235–238. Raxworthy, C.J., Forstner, M.R.J. & Nussbaum, R.A. (2002) Chameleon radiation by oceanic dispersal. Nature, 415, 784– 787. Ronquist, F. & Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics, 19, 1572–1574. Shimodaira, H. (2002) An approximately unbiased test of phylogenetic tree selection. Systematic Biology, 51, 492–508. Shimodaira, H. & Hasegawa, M. (1999) Multiple comparisons of log-likelihoods with applications to phylogenetic inference. Molecular Biology and Evolution, 16, 1114–1116. Shimodaira, H. & Hasegawa, M. (2001) Consel: for assessing the confidence of phylogenetic tree selection. Bioinformatics, 17, 1246–1247. Silva-Taboada, G. (1979) Los murcie´lagos de Cuba. Editorial Academia, Havana. Simmons, N.B. (2005) Order Chiroptera. Mammal species of the world: a taxonomic and geographic reference (ed. by D.E.

Wilson and D.M. Reeder), pp. 313–529. Johns Hopkins University Press, Baltimore, MD. Sua´rez, W. & Dı´az-Franco, S. (2003) A new fossil bat (Chiroptera: Phyllostomidae) from a Quaternary cave deposit in Cuba. Caribbean Journal of Science, 39, 371–377. Swofford, D.L. (2002) PAUP*. Phylogenetic analysis using parsimony (*and other methods), Version 4.0b10. Sinauer Associates, Sunderland, MA. Tavares, V.d.C. & Simmons, N.B. (2000) Phylogenetic relationships and biogeography of Short-faced stenodermatine bats: preliminary results and hypotheses. Bat Research News, 41, 143. Teeling, E.C., Springer, M.S., Madsen, O., Bates, P., O’Brien, S.J. & Murphy, W.J. (2005) A molecular phylogeny for bats illuminates biogeography and the fossil record. Science, 307, 580–584. Timm, R.M. & Genoways, H.H. (2003) West Indian mammals from the Albert Schwartz collection: biological and historical information. Scientific Papers of the University of Kansas Natural History Museum, 29, 1–47. de la Torre, L. (1961) The evolution, variation, and systematics of the Neotropical bats of the genus Sturnira. PhD Dissertation, University of Illinois, Urbana, IL. Wetterer, A.L. (2003) Evolution and the ecological diversification of phyllostomid bats (Mammalia: Chiroptera). PhD Dissertation, Columbia University, New York. Wetterer, A.L., Rockman, M.V. & Simmons, N.B. (2000) Phylogeny of phyllostomid bats (Mammalia: Chiroptera): data from diverse morphological systems, sex chromosomes, and restriction sites. Bulletin of the American Museum of Natural History, 248, 1–200. Yoder, A.D. & Yang, Z. (2000) Estimation of primate speciation dates using local molecular clocks. Molecular Biology and Evolution, 17, 1081–1090.

BIOSKETCH Liliana M. Da´valos is a post-doctoral researcher studying the ribosomal genes of malaria parasites. Her interest in the bats, forests and people of the Neotropics spans the historical and ecological biogeography of the West Indies, the genomics of model organisms and malaria parasites, and the development of policy conducive to the attrition of cars.

Editor: Lawrence Heaney

Journal of Biogeography 34, 364–375 ª 2006 The Author. Journal compilation ª 2006 Blackwell Publishing Ltd

375