Cambridge, Mass. US ISSN 0006-9698 9 May 2018 Number 560 HERPETOFALNA OF CAY SAL BANK, BAHAMAS AND PHYLOGENETIC RELATIONSHIPS OF ANOLIS FAIRCHILDI, ANOLIS SAGREI, AND TROPIDOPHIS CURTUS FROM THE REGION R. Graham Reynolds,''" Alberto R. Puente-Rolon,'* Amy L. Castle,' Martun Van De Schoot,"' and Anthony J. Geneva"^ Abstract. Although the terrestrial herpetofauna of the Bahamas Archipelago is largely well characterized, many smaller or outlying islands and island banks are poorly studied owing to their remoteness. The Cay Sal Bank is an uninhabited island bank lying between the Florida Keys and western Cuba. Politically a part of the Commonwealth of the Bahamas, the Cay Sal islands are disjunct from the rest of the Bahamas Archipelago to the east. The bank supports 117 small emergent islands around its perimeter, on which at least six species of squamates have been recorded. Recent expeditions have added to this species list, although several islands remain unsurveyed. Of these six squamate species, only Anoli.s fciircliildi is considered endemic to the bank. The evolutionary relationships of the other species are speculated to derive from either Bahamian or Cuban progenitors. Here we report on a recent expedition to the Cay Sal Bank, and provide updated and novel island records. We further characterize the two native Anoli.s species and one Tropiclophis species using molecular phylogenetic analyses to estimate the likely origins and divergence times of these species. We find that both .Anolis species are recently (< 2 Mya) derived from, and likely conspecific with, west Cuban ancestors, whereas the Tropidophi.s is likely conspecific with the populations of Tropidopid.s ciirtii.s on the Great Bahamas Bank. Thus the terrestrial squamate herpetofauna on the bank are a product of dispersal from both the Great Bahamas Bank to the east and western Cuba to the south. We pro\ ide an updated understanding of the biogeography and evolutionary history of these squamates on the Cay Sal Bank. Key words: Bahamas; mitochondrial DNA; phylogenetics; systematics ' Department of Biology. University of North Carolina Asheville. One University Heights. Asheville. North Carolina 28804. U.S.A.; e-mail: greynoldm imca.edu. " Museum of Comparative Zoology. Harvard University. 26 Oxford Street. Cambridge. Massachusetts 021.^8. U.S.A.; e-mail; robertreynoldsiu fas.harvard.edu. Departamento de Biologia. Recinto Universitario de Mayagiiez. Call Box 9000. Mayagiiez. Puerto Rico 00681. U.S.A.; e-mail; albertonskiiu gmail.com. Department of Organismic and Evolutionary Biology & Museum of Comparati\e Zoology. Harvard University. 26 Oxford Street. Cambridge. Massachusetts 021.^8. U.S.A.; e-mail: genevam fas.harvard.edu. ® The President and Fellows of Harvard College 2018. BREVIORA No. 560 INTRODUCTION The terrestrial herpetofauna of the Baha¬ mas Archipelago has been well studied over the last century (e.g., Barbour, 1906; Coch¬ ran, 1934; Barbour and Shreve, 1935; Knapp et ah, 2011). A series of compendia has collated island herpetofaunal records into comprehensive resources (MacLean et ah, 1977; Schwartz and Henderson, 1991; Hen¬ derson and Powell, 2009; Buckner et ah, 2012), demonstrating the breadth and depth of our understanding of herpetofaunal di¬ versity and distribution in the archipelago. Nevertheless, some portions of the region remain understudied, largely owing to diffi¬ culty accessing remote islands. Recent stud¬ ies of remote islands, such as the Great Isaac Cays, Ragged Islands, and the Conception Bank, has led to the further characterization of island herpetofaunal distributions (Win- chell et ah, 2015; Geneva et ah, 2016; Reynolds et ah, 2016a; Reynolds and Puente- Rolon, 2016a) and the discovery of new species (Reynolds et ah, 2016b). The Bahamas Islands are a political unit, administered by the Commonwealth of The Bahamas, which encompasses a series of island banks occupying a triangle between the Greater Antilles, peninsular Florida (U.S.A.), and the North Atlantic Ocean. This region, when inclusive of the geologi¬ cally cohesive (yet politically separate) Turks and Caicos Archipelago, is referred to as the Lucayan Archipelago and stretches > 1,360 km from northwest to southeast (19°84- 27°30'N, 68°70'-80°54'W). This archipelago is composed of a complex of large and small carbonate platforms (banks) supporting hundreds of emergent islands. These island banks are generally inundated by water no greater than 20-40 m deep, whereas depths of water between banks can reach thousands of meters (Purkis et al. 2014). During intra¬ glacial periods a much greater portion of these banks was emergent, resulting in substantially larger landmasses than are currently present, although the deep chan¬ nels separating banks have maintained their isolation from one another and from Florida and the Greater Antilles. The terrestrial herpetofauna of the Lucayan Archipelago is composed of lineages derived from over¬ water colonization from Hispaniola and Cuba, with some endemism having likely evolved in situ (Hower and Hedges, 2003; Hedges and Conn, 2012; Knapp et al., 2011; Reynolds et al., 2013, Geneva et al., 2015; Reynolds et al., 2016b). To the west of the Lucayan Archipelago lies the Cay Sal Bank, an isolated and uninhabited carbonate platform adminis¬ tered by the Commonwealth of The Baha- 'y mas. This relatively small (4,000 km"; Goldberg, 1983) bank lies between peninsu¬ lar Florida, Cuba, and the Great Bahamas Bank, and is bounded by the Florida Straits to the north, the Santaren Channel to the east, and the Nicholas Channel to the south (Fig. 1). Previously an exposed landmass 11,000 years ago, the Cay Sal Bank was largely inundated over the subsequent 5,000 years owing to sea-level rise and slight bank subsidence (Purkis et al., 2014). Presently water depths range from 0.5 to 20 m across facies of the bank (Goldberg 1983; Bruckner et al. 2014; Purkis et al., 2014). The edges of the bank are bounded by deep drop-offs, with water depths declining rapidly from 30 to 1000 m. Portions of the bank edges support a maximum of 1 17 low-lying islands (< 10 m above sea level), most of them with little vegetation, representing a total land area of 484 ha (Mackin et al., 2015). Most of these islands are small rocks less than 20 ha in area, whereas six larger islands range from 29 to 1 50 ha in area (Goldberg, 1983; Krysko etal., 2015). The islands of the Cay Sal Bank were first described from a zoological perspective by 2018 CAY SAL BANK ULRPLTOI'AUNA 3 A Florida Bank •• • Cay Sal Bank \ \ \ Depth (m) 0 I - ' 200 ■400 600 800 1000 1200 -1400 -1600 Great Bahama Bank • sampling sites Cuba B Dispersal Hypotheses Cuba AnoUs porcatus 9 AnoHs sagrei sagrei 9 Tropidophis cetiae 9 Trop/dophis melanurus 9 Cubophis canthengerus ^ Florida Bank Ano/is caroHnensis 9 Great Bahama Bank Anolis smaragdinus f AnoHs sagrei ordinatus ? Tropidophis curtus 9 -82 -80 -78 Longitude Figure 1. A. map of study region, including the Cay Sal Bank and the surrounding Florida. Cuba, and Great Bahama banks. Water depth information is indicated by blue colors, and the three currents surrounding the Cay Sal Bank are shown with approximate How rates (from Purkis et al., 2014). Note that Goldberg (1983) incorrectly represents the direction of How of the Nicholas Channel. Sampling sites on the Cay Sal Bank are show n as red circles. B. dispersal hypotheses for terrestrial squamate reptiles serving as potential progenitors of Cay Sal Bank species. The study region is shown as a three-dimensional map with bathymetric data on the r-axis. Potential progenitor species are show n adjacent to their respective banks, with green check marks indicating confirmed dispersal and red question marks indicating possible dispersal. Orange arrows indicate likely dispersal trajectories to the Cay Sal Bank from either Cuba or the Great Bahama Bank, the latter of which is currently considered to be the likely pathway for squamate colonization. The gray arrow indicates that dispersal of green anoles {Anolis caroHnensis) from Florida is possible, although unlikely, owing to the strength of the current in the Florida Straits. Agassiz (1894) during a cruise through the region. Since that time some notable expedi¬ tions to the region have increased our knowledge of the biodiversity of the bank (Cory. 1891; Buden and Schwartz, 1968; Goldberg, 1983; Purkis et al., 2014; Krysko et al., 2015; Mackin et al., 2015). Six extant species of squamate reptiles occur on the Cay Sal Bank, and are thought to be derived largely from the Bahamas to the east, as most are considered conspecific with Baha¬ mian subspecies (Table 1; Buden and Schwartz, 1968; Buckner et al., 2012). The squamate diversity of the Cay Sal Bank was first described by P. Bartsch between 1928 and 1930 (in Cochran, 1934), who identified two Anolis lizards (A. sagrei and A. fair- childi) and one Tropidophis snake {T. curtus: Cochran, 1934; Buden and Schwartz, 1968). Thomas later found the gekkonid lizard Sphaerodactylus nigropunctatus and the sco- lecophidian blindsnake Typhlops himiniensis in 1967 on Elbow Cay (Buden and Schwartz. 1968; Thomas, 1968a). Recently the Cuban colubrid snake species Cubophis cantherige- rus was discovered on the Anguilla Cays at the southeastern edge of the Bank (Krysko et al., 2015), suggesting that additional explo¬ ration of the region might continue to resolve species diversity, distribution, evolu¬ tionary relationships, and biogeographic origins of the terrestrial herpetofauna. Of these six squamate species, A. fairchildi is the only presently recognized endemic reptile r.\BLi-; 1 . Islands on tiil Ca^' Sal Bank with known and nlwlv doc umlnthd tlrrlstkial iihrpltofaunal rlxords. No rlxord indicatls that no historic al RIX'ORD LXISTS LOR THAT ISLAND. AND THAT NO SPIX IMLNS WLRL OBSLRVLD ON THAT ISLAND DURINO OUR RLSEARCH CRUISL. RlX'ORDS ARE FROM BUC KNER FT AL. (2012), WHIC H IS A COLLECTION OF PREVIOUSLY DOC UV1ENTED RECORDS FROM A VARIETY OF SOURCES. ASTERISK INDIC ATES ISLANDS THAI WERE VISITED ON OUR RESEARCH C RUISE. BREVIORA No. 560 X < GO < 00 r I < I ? - u o OJ T3 (U -d u u G (75 o o ■G G «- > t: G G G -p r- G > 1-^ CO r>, p ■p P 5 G r“ G G r" O 5 O O G G <— Zj 75 r^. c o G G G G G G G O G o o o G o G -o TD -G G G O "G G > G GO O G G O G G -p O G G O G G G G V . ^ ON Z z o^ z o ri G G G G G G G G G T3 "O *0 "G "G P G G G G G G G G G G G G G G G G G G •— !— G G G G G G r“ G G G G X5 "G ■a -G X r- ■o G G G o G G G o G G G o G G r- o G G G u G G G IJ G G G o G G G G o G r- r“ G •G r-" G G G G X — G G ■G r- r- x ■G ■G XJ G "G G "G "O G G G G > r- G > u © © w © C G X G Urn G tj CJ G G Q f- r- G CO -C G CO X) G G G r- r* r- G G 5 k. k. u G G G G G G it if G G G G G G G G G G ■G "G C e S X X r^l r^l O rj r j Gl o Tj- sO o O — C" • — o c^- ON ON ON O' O' z o^ On On X O' X z Z ri ri ri Z p- r* nC X Gl o 00 tJ* r^i O ko X r J ri n p" rj r^i Tj- o O o O ON 0 O 0 ON o O On O'! o ON o O' ri 0 O' 00 0 ri 00 0 r^> ri 00 o n 0 ri 00 o ri 0 ri r- o r- * X CO e/j u G G X G X * o Q * * 75 C (— Oij t/) IT] o x: uj •w CO G Cay G G G G r- c- (U G G 0^ 5 o r* < U u LU UJ o UJ U 0 < u tNote that this is given incxnTectly as 23°55'47.9994"N. 80°28' 1 1 .9994"W in Buckner et al. (2012). 2018 CAY SAL BANK llLRPETOl'AUNA 5 tound on the Cay Sal Bank, and is thought to be derived from Cuban aneestors (Glor et ah, 2005). Here we define herpetofauna to inelude the terrestrial reptilian and amphibian fauna (the latter of whieh laeks representation on the Cay Sal Bank), and thus excludes marine turtles. We provide a comprehensive review of herpetofaunal records from the Cay Sal Bank on the basis of published accounts and the results of an expedition we conducted to the region in August 2015. We also provide island accounts, including descriptions of terrestrial habitats and herpetofaunal rec¬ ords for each major island on the bank. We infer the phylogenetic relationships of three terrestrial species and their close congeners, which, when combined with other known information on the historical biogeography of other species, provides a clear picture for the origins of this depauperate and isolated herpetofaunal community. MATERIALS AND METHODS Research cruise We (RGR and ARPR) traveled to the Cay Sal Bank aboard a chartered vessel, spending 4 days and 3 nights (1 1 14 August 2015) on the Bank (Reynolds and Puente-Rolon, 2016b). During this cruise, our ability to sample islands was limited by weather and safety considerations; nevertheless, we were able to diurnally survey five islands across the Bank (Figs. 1, 2). Elephant Rocks and Double Headed Shot Cay are technically comprised of two islands each, separated by narrow channels, although we treat them here as single units. Surveys involved spend¬ ing between 1 and 8 hours on each island, depending on the island size, during which time we sampled all available habitats using visual encounter surveys and turning over cover objects. Genetic data and analyses We collected three tissue samples of A. fairchildi from Cay Sal Island, Bahamas (23°41'26.8476"N, 80°23'23.9964"W), the only island on which we observed them. We collected tissue samples of A. sagrei from five islands across the Cay Sal Bank: Great Dog Rock {n = 3; 24°r57.0354"N, 79°5()'().9168"W), Cay Sal (/? = 29; 23Mr26.8476"N, 80°23'24.3564"W), Elbow Cay (77 = 11; 2 3 °5 7 T 7 . 0 0 2 8 "N . 80°26'32.64"W), Double Headed Shot Cay (77 = 2; 23°59'32.82"N, 80°20'14.2794"W), and Elephant Rock (77 = 10; 23°55'44.9796"N, 80°28T0.5312"W). Final¬ ly, we collected one tissue sample from Tropidophis curt us on Elbow Cay (23°57'23.3886"N, 80°26'25.6014"W), Cay Sal Bank. We extracted whole genomic deoxyribo¬ nucleic acid (DNA) from tissue samples of the three terrestrial squamate species using the Wizard SV® kit (Promega, Madison, Wisconsin) and subsequently stored extracts at — 20°C. We used polymerase chain reaction to amplify fragments of the mitochondrial genome for Anolis tissue samples (nicotin¬ amide adenine dinucleotide subunit 2 [ND2]; primers from Macey et ah, 1997; conditions in Revell et ah, 2007) and Tropidophis tissue samples (cytochrome h [CYTB]; primers and conditions in Reynolds et ah, 2013). Cyto¬ chrome b has been shown to be useful in species identification in boas (Campbell, 1997; Burbrink, 2004; Reynolds et ah. 2013); nevertheless, we additionally se¬ quenced three protein-coding nuclear loci for our Tropidophis sample (primers and conditions in Reynolds et ah, 2013. 2014): neutrophin-3 {ntfS). brain-derived neutro- phic factor {hdnf). and bone morphogenetic protein 2 {hmp2). We purified and sequenced products in both directions on an automated sequencer (ABI 3730XL) at the Massachu- 6 B REV 10 R A No. 560 setts General Hospital DNA Core facility, Cambridge, Massachusetts and the Genomic Sciences Laboratory at North Carolina State University, Raleigh, North Carolina. We assembled contigs and manually verified ambiguous base calls using Geneious® 10.2.1 (Biomatters, Auckland, New Zea¬ land). We resolved heterozygous nuclear sequences using PHASE v. 2.1 (Stephens et ah, 2001; Stephens and Donnelly, 2003) implemented in DnaSP vS.lO.l (Librado and Rozas, 2009) using default parameters for 100 iterations with a burn-in of 100, and a cutoff of posterior probabilities (PP) >0.7 for base calling. To determine the phylogenetic relation¬ ship of A. fairchildi relative to other Cuban and Bahamian green anoles, we mined additional ND2 sequences representing A. carolinensis clade green anoles {sensu Nich¬ olson et ah, 2012) from Genbank (data largely from Glor et al., 2004, 2005; Kolbe et al., 2007). We additionally obtained tissues from the Bahamian green anole species A. smaragdinus from South Bimini {n = 2) and Great Ragged {n — 2) islands, representing the northern and southern edges of the western Great Bahamas Bank (Fig. 1) in separate expeditions. These samples were extracted and sequenced as above. To ascertain the origins of Cay Sal Bank A. sagrei, we subsampled (from our collected tissue samples) representatives from four of the five islands (Cay Sal, Elbow Cay, Elephant Rocks, and Great Dog Rock). We then aligned our ND2 sequences as above with a data set containing 295 haplotypes of A. sagrei sampled from the entire range of the species. This includes individuals from across the Bahamas and Cuba, as well as additional Caribbean lineages. To identify phylogenetic affinities of the Tropidophis sample we obtained from Elbow Cay, Cal Sal Bank, we aligned our mito¬ chondrial CYTB and nuclear ntf3, bdnj\ and bmp2 sequences with others from Gen Bank or generated de novo, focusing on species with likely relationships to Cay Sal Bank animals. We included novel sequences from representative samples {n = 2) of T. curtus from Eong Island, Bahamas (Table 1). We also added all available nuclear and mito¬ chondrial sequence data for Tropidophis, the ingroup Trachyboa, and the outgroup Tro¬ pidophis taczanowski (see Reynolds et al., 2014) obtained from GenBank and from the alignments in Reynolds et al. (2014). For all three data sets, and for each locus separately, we aligned our novel and Gen- Bank sequences using the ClustalW 2.1 (Larkin et al., 2007) algorithm implemented in Geneious using reference sequences and default parameters. For the Tropidophis data set, we additionally created a concatenated alignment of both mitochondrial DNA and nuclear DNA sequences. Alignments are publicly available on GitHub (https:// github.com/caribbeanboas/). We selected the best-fit model of molecular evolution for the ND2 locus (TrN + I + G) and CYTB locus (HKY + H- G) using the Bayesian information criterion in jModel- Test2 (Guindon and Gascuel, 2003; Darriba et al., 2012). We conducted separate maxi- mum-likelihood (ME) analysis for each alignment (green anoles, brown anoles, and concatenated Tropidophis) using the RAxME algorithm (Stamatakis, 2006) im¬ plemented in the RAxME plug-in for Gene¬ ious. We used the GTRGAMMA model and the rapid bootstrapping algorithm with 1 ,000 bootstrap (BS) replicates followed by the thorough ME search option with 100 inde¬ pendent searehes. We consider BS values above 70% to indicate relatively well-sup- ported clades (Felsenstein, 2004). To estimate divergence times across the Anolis mitochondrial gene trees, we inferred 2018 CAY SAL BANK I lERPLTOFAUNA 7 CO 00 oo CN Rocks Dog Rocks Damas Cays Cay Sal o lO d 00 Anguilla Cays Nicholas Channel 0.3m/s 80.5 80.17 79.83 79.50 Figure 2. Major islands on the Cay Sal Bank, with currents and current speeds indicated on the bank edges. White circles indicate islands visited durinc our cruise. time-calibrated Bayesian ND2 coalescent trees in the program Beast vl.8 (Drummond et al., 2012) using a relaxed molecular clock model and a rate of molecular evolution of 0.65% divergence per lineage, per million years. This rate has been previously used for the ND2 locus in other lizards (Maccy et al., 1998), including many studies of Caribbean anoles (e.g., Gartner et al., 2013; Geneva et al., 2015). We furthermore note that we arc primarily interested in the relative rather than absolute divergence times, and thus our analyses will be largely insensitive to the specific molecular clock rate used. We ran a separate BEAST analysis for the concate¬ nated Tropidophis multilocus data set. For each of the three separate BEAST analyses, we ran the Markov chain Monte Carlo for 100 million generations using the TrN + 1 + G substitution model, a Yule spcciation prior, and an uncorrelated lognormal re¬ laxed molecular clock model. We repeated each analysis three times with different starting numbers, sampling every 10*^ gener¬ ations and discarding the first 25% of generations as burn-in. We assured adequate mixing of the chains by calculating the effective sample size values for each model 8 B REMORA No. 560 A pon:«ru«. Ay654CMSi A pon^tus Ay»S4tM8 'a poaafw.AveSAtM/ A ponM/«s_Ay65*045 A ponMlM A port«f«»_AY65403« A powafMj^Ay45403< A ^KKcafus.EU 106343 A porcaftrt_Ay$54051 A py6&403a A p(vta/us.AY654037 A poa4fMS_Ay654036 A p(xi:afus.Ay654036 A pon»rus.AY654053 A po'cafua_AY6&4042 A porC8fus_Ay296195 A poreafa».Ay65404i A. p(yi:afuJ_AY664O«0 A po/csr(a_EU^06345 A porcafus oiiT- A porcatus West Cutw A pooictus West Cuba A pcvratus Cenbat East Cuba A p«vc«us Cenbai Weal Cuba A pomaiui Anola cmnlinposit CeniralCubsA poncatus Ar>o4s smvtgOtnus East CuM A portobji AnoAs me/n«r<> Anobi brumous SE Cuba A porcatus AnoAs tonptceps AncAa sagrw 12 5 10 7.5 5 2 5 0 Figure 3. Bayesian phylogeny of Cuban green anole species (carolineusis clade), with major Cuban clades. other recognized species, and outgroups collapsed at the bottom. The clade containing Anolis fairchildi samples from Cay Sal Island is expanded, with A . fairchihli in green, consisting of an A. porcatus lineage from western Cuba. Numbers at nodes are posterior probabilities. The scale bar at the bottom represents coalescent times in Mya. Anolis fairchildi from Cay Sal Island is shown top left (photo by RGR). parameter, with values > 200 indieating adequate sampling of the posterior distribu¬ tion. We assessed convergenee of the inde¬ pendent runs by a eomparison of likelihood scores and model parameter estimates in Tracer vl.5 (Rambaut et ah, 2013). We combined the results from the three analyses using Logcombiner and generated a maxi¬ mum clade credibility tree using TreeAnota- tor. RESULTS Our surveys of five islands across the Cay Sal Bank resulted in some updated distribu¬ tions of the terrestrial herpetofauna (Table 1). We found A. sagrei on additional islands, and failed to find evidence of A. fairchildi outside of Cay Sal Island despite additional records in Buckner et al. (2012). Additional details are provided in the species accounts below. For the A. fairchildi mitochondrial data set, we obtained a 1,172 base-pair (bp) alignment for 224 haplotypes of the eight recognized A. carolinensis clade species, including representative Cuban species (A. porcatus, A. cdlisoni) from across Cuba, and Bahamian species {A. smaragdinus, A. brim- neus) from across the Bahamas. Our three A. fairchildi samples are nearly identical to each other and reciprocally monophyletic with respect to other carolinensis-cVddQ lineages (Fig. 3). We found that A. fairchildi is likely a branch of the West Cuban A. porcatus lineage, with an estimated coalescent time of 1.05 Mya (BS = 37; PP = 0.73; 95% highest posterior density [HPD] = 0.6-1. 6 Mya; Fig. 3), and thus is not closely related to Bahamian A. smaragdinus . We aligned 1,092 bp of mitochondrial DNA from Cay Sal A. sagrei with 296 sequences of A. sagrei from across the range of the species. We found phylogeographi- cally concordant sequences on the Cay Sal 2018 CAY SAL BANK HERPETOFAUNA 9 A M0nH 2fiOS Cub* A s«0rw 2ve6.Cut« A s«(trw 300:>.Culw A iMgm 2M3 Cut» A 3037. Cub* A s*9r*>_2902_Cjb* A isgm 30b1 Cub* * isgrBt 3063 Cub* A sagm 3060 Cub* A *(r*ra(_22lO.Cub* A 3036 Cub* A s*0r»_3O33_Cjb* A iagrm 305*.Cub* A &*0r*t_283O Cut* A aagm 2988 Cub* * s*0/*< 2M5 Cub* A rngm, ?99B Cub* A »*0/«. 2636. Cub* A 9«9r»_2804,Cut* A «*$K*>_2a40_Cub* A qimJrncalUaf USNM6t6016 Cub* A qu*df>ui.*/ 3066. Cuba * bnmfi USNM&1S899 Cub* A aagfm 3064 Cut* A M0m.'JSNMSl6929..Cub* A s*0f*i_USNMb16928_Cub* A i«gtm^299fJJ\4a A &*gr*> Cay S*i lUmd 1 A aagm Cay Sal 1^*013 2 A 9egn> Etow Cay A iAQfai Ei*pb*n( Rocks t A tagnt Et*pb*nl Rock* 2 A iffem Dog Rock* 1 A Aagrw Oog Rock* 2 A iagm 2260 Cub* A s*gro> 2974.Cut* 4 s*g«9< 2d76 .Cub* A ssg^_2249. Cub* X s*gr*«_USNM3376e6..Cub* ^ s*gw.2673_Cut* A 9*gm.USNM337684 Cub* 4 s*gm. 2972. Cub* A s«gre<. 2975. Cub* A sagm C«no*lC«sl Cub* A s*gr»t Central Cub* A s*^)r«i Cub* and Bahama* A iagm Southwest Cub* A sagfat Cayman Bmc A sagnt unte Caymn A sagnt South Cub* / Mesoamertca Anoks onfTtan, AnoAs sagn' Anoas nomoiacAts 12 5 10 7 5 5 2.5 0 Figure 4. Bayesian phylogeny of the Cuban brown anole, Anolis sagrei, complex, w ith major Cuban chides, sister species, and outgroups collapsed. Blue represents the A. sagrei individuals sampled from islands on Cal Say Bank, and numbers at nodes are posterior probability values. The scale bar at the bottom represents coalescent times in Mya. Anolis sagrei from Cay Sal Island is shown top left (photo by A RPR). Bank, with generally well-supported recipro¬ cally monophyletic groupings from Cay Sal Island, Elbow Cay, Elephant Rocks, and Dog Rocks (Fig. 4). We also found evidence for east-west divergence across the Cay Sal Bank, with the previously unreported Dog Rocks population exhibiting a minimum of 0.35% mitochondrial DNA divergence from the west Cay Sal Bank populations (Elbow Cay). Further, we find that Cay Sal Bank A. sagrei are not sister to Bahamian A. sagrei. as previously suggested (e.g., Budcn and Schwartz, 1 968; Buckner ct al., 20 1 2), but are instead sister to west Cuban A. sagrei lineages (Fig 4). Cay Sal populations share a mean coalescence time of 2. 1 6 Mya with these geographically proximate Cuban pop¬ ulations (BS = 43; PP = 0.73; 95% HPD = 1.71-2.66 Mya). Our Cay Sal Bank Tropidophis specimen exhibited a light rosy coloration (Fig. 5), which has not been reported from other populations of T. curtiis in the Bahamas, although the dorsal patterning is similar (Fig. 6; Schwartz and Henderson. 1991; Tolson and Henderson. 1993). Additional morphological description of the specimen follows below. We aligned 3.976 bp of mitochondrial and nuclear sequence data generated from this specimen with 14 other taxa obtained from Genbank. including three from Bahamian T. curias, one from Caymanian T. sc/nvartzi. one Cuban T. feicki. and six Hispaniolan-derivcd T. greeu- ^\'ayi and T. haetiamis. We find that Cay Sal Tropidophis is likely conspecific with T. curtiis. as it is minimally divergent (0.6%) from Bahamian sequences obtained from the eastern end of the Great Bahamas Bank 10 BREVIORA No. 560 A -• Tropi(iophis__h8eUanus_Samana — TiopidopnisJiaeuanus_haeUanus •>1 100 TropidQphis_tiaet\anus_FJ755 f 0 J %6 TropKiopnis_haetianus_NC012573 • Troptdoph\s_graenwayi_MC n - Tropidophis_greenwayi_NC - Tropidophis_feicki — Tropidophis_scnwaiW_CB _ Tropidophis_curtus_CaySal , Trr>pidophls_curtus_cunus T mpidophis_ cur1us_ 2_LI X - Tropidophis_cuftus_THU69d69 Tropidophistaczanowskyi B - Trachyboa^gulans - Trachyboa__Ooulengen ■ Tfopidophtsjaczanowskyi TropidDphis_hBGtianus_ FJ75516 1 Tropidophis_haelianu5_NC012573 - Tropidophis_haetianus_Samana - Tropidophis_greenwayi — Tmpidophis^greenwayi^MC - Tropidophis_feicki r- Tropidophis_curtus_2_LI Tropidophis_cur1ijs_curius Tropidophis_CiJr1us_ CaySai Trcpidophis_^curtus_ THU69B69 T ropidophis_ sch wBrt2i_ CB Figure 5. Phylogenetic trees for our Tropidophis data sets (top) and photos of the specimen of Tropidophis curtiis encountered on Elbow Cay, Cay Sal Bank. A, maximum-likelihood tree depicting relationships among 1.083 base pairs of mitochondrial cytochrome h sequence data, including Cay Sal T. curtiis (in blue). Nodes indicate bootstrap support. B, Bayesian phylogenetic reconstruction of concatenated mitochondrial DNA and three nuclear genes for Tropidophis. with Cay Sal T. curtiis in blue. Nodes represent posterior probabilities. C. photo of T. curtiis from Cay Sal Bank; D, photo of tail; E, photo of venter. Photos by ARPR and RGR. (Long Island). Indeed, we find weak support (BS = 74; PP = 0.50) for distinctiveness of this specimen from T. curtiis in concatenated ML and Bayesian analyses (Fig. 5). DISCUSSION Species accounts Anolis fairchildi. Anolis lizards in the West Indies are one of the best-studied examples of an adaptive radiation the remarkable evolution of various forms from a single ancestral lineage. Speciation and diversifica¬ tion in the region have led to the present recognition of hundreds of anoline species. Green anoles, specialized to occupying the upper trunks and canopies of trees, have evolved independently on many of these islands, including Cuba. The Cuban green anoles have since colonized other regions, including the Bahamas, the southeastern United States, the Cayman Islands, and coastal Central America (Williams 1969; 2018 CAY SAL BANK 1 iLRPETOl AUNA Figure 6. Two Tropidophis airtiis individuals from Long Island. Bahamas, on the Great Bahama Bank about 520 km to the southeast of Elbow Cay. showing brown and gray morphs present in those populations. Photos by RGR. Schwartz and Henderson, 1991). This dias¬ pora has diversified such that several these populations are recognized as different species. Within this group, eight species are recognized, and evolutionary relationships are known for all but A. fairchildi (Glor et ah, 2005). Morphological features, such as the relatively large size of A. fairchildi (Barbour and Shreve, 1935; Buden and Schwartz, 1968; this work), suggested that it might have evolved from a historical colonization from Cuba. An alternative is that this species colonized Cay Sal from the Bahamas (Cochran, 1934; Buden and Schwartz, 1968), or from the North Amer¬ ican mainland. Determining if A. fairchildi is more closely related to green anoles from Cuba, the Bahamas, or elsewhere using morphological data is complicated by pat¬ terns of morphological convergence in West Indian anoles that can cloud evolutionary relationships (Glor et ah, 2005). and this is clear among carolinensis-d'ddo. Anolis (Buden and Schwartz, 1968). Barbour and Shrcvc (1935) in their description of the species suggested that A. fairchildi is morphologi¬ cally similar to A. smaragdinus and A. porcatus, but that it differs from the two owing to A. fairchildi^ larger dorsal and temporal scales. Oliver (1948) suggested that Bahamian Anolis and A. porcatus were conspecific with A. carolinensis. These au¬ thors (Barbour and Shreve, 1935; Buden and Schwartz, 1968) also indicated differences in coloration, particularly dorsal white spotting in A. fairchildi. although it should be noted that they were working with preserved specimens. Buden and Schwartz (1968) examined meristics and squamation of A. fairchildi in relation to other carolinensis- clade anoles, determining that head squama¬ tion allows diagnosis of this species. Our samples came from three male A. fairchildi of relatively large size (70. 73. and 74 mm snout-vent length). Their dewlaps were relatively small, with a pinkish hue and interspersed white scales. Nevertheless, given the characteristics offered by Barbour and Shreve (1935). Oliver (1948). and Buden and Schwartz (1968). morphological diagnosis of this species without knowing a specimen's provenance would be extremely challenging, with maximum male body size the only character suggesting a relationship to A. 12 B REV 10 R A No. 560 porcatus (Buden and Schwartz, 1968). We observed minimal light dorsal spotting in A. fairchildi on Cay Sal, with this coloration owing to white and light blue scales occa¬ sionally clustering in groups of two. Before the present study, it was unclear how long this species had been on Cay Sal, or even whether these anoles diverged sufficiently to be considered a valid species (under the phylogenetic species concept). Our molecular phylogenetic results can overcome some of the limitations of a strictly morphological data set, and also provide insight into the magnitude and timing of divergence of this species from its closest relatives. Here we are able to determine that A. fairchildi colonized Cay Sal from Cuba, and not from the Bahamas or Florida (hypotheses in Fig. IB); coalescent times indicated that this colonization occurred within the late Pleistocene; and although this species might be considered conspecific with A. porcatus (to avoid paraphyly of the latter species), its unique geographical isola¬ tion and (possibly) diagnostic morphological features might lead to continued recognition as a unique lineage. More generally, a revision of this group might be needed — populations currently recognized as A. por¬ catus are far more divergent from each other than some are from not only A. fairchildi but also from A. allisoni, A. hrunneus, A. carolinensis, A. longiceps, A. maynardi, and A. smaragdinus (Glor et ah, 2005). We refrain from performing any revision here because our sampling of genetic loci and green anole populations was intended to only provide insights into the evolutionary relationships of A. fairchildi and is insuffi¬ cient for a full treatment of the group. Anolis sagrei. Anolis sagrei is a widely distributed species and is thought to have the largest natural range of any West Indian Anolis lizard (Williams, 1969). Populations on the Cay Sal Bank are described as representatives of the Bahamian subspecies A. s. ordinatus, implying a natural dispersal westward or northward across the Santaren Channel. Bahamian populations of A. sagrei were recognized as a separate subspecies, A. s. ordinatus, owing to presumed differences in dewlap color and contacting supraorbital scales (Barbour, 1937; Oliver, 1948). Cay Sal specimens were assigned to A. s. ordinatus on the basis of similar body sizes and some head squamation similarities (Buden and Schwartz, 1968). Contrary to these expectations, we found that the five Cay Sal Bank populations we sampled are derived from Cuban progeni¬ tors. Because our alignment included fine- scale sampling across the native range of A. sagrei, we infer a west-Cuban origin of Cay Sal populations, likely near Havana, Pinar de Rio, or Matanzas provinces. Further, Cay Sal populations from across the bank are monophyletic with respect to their Cuban ancestors, suggesting a single colonization event, or at least the persistence of a single lineage (owing to lineage sorting). The widespread range of A. sagrei appears to be the result of many independent dispersal events from Cuba to elsewhere in the West Indies (Kolbe et ah, 2004). Our time- calibrated phytogeny suggests that A. sagrei colonized Cay Sal around the time when the species also dispersed to the Swan Islands and Mesoamerica, but more recently than out-of-Cuba dispersals to the Lucayan Ar¬ chipelago and the Cayman Islands (Cayman Brae and Little Cayman). Tropidophis curtus. Tropidophis curtus is recorded from the Cay Sal Bank and from across the Great Bahamas Bank (Buckner et al., 2012). Before this study, only three records existed from the Cay Sal Bank, consisting of two specimens from Elbow Cay (KU KUH 269003-04; R. Thomas 1967) and one juvenile female specimen from Double Headed Shot Cay (USNM Amphib- 2018 CAY SAL BANK HERPETOFAUNA 13 ians & Reptiles 81536; listed as “anon. 1930” in aecessioned reeord at the National Muse¬ um of Natural History, but eolleeted by Bartsch in 1930 fide Buden and Sehwartz, 1968). Given the ineonsisteneies in how these two islands are labeled on maps (e.g., Google Earth 201 1 imagery, aeeessed Oetober 2017), it is plausible that all three individuals eame trom Elbow Cay. This population (or populations) was considered to represent T. ciirtus (originally T. carnis ciirtiis from the Bimini Islands and New Providence) after careful morphological analysis of the three known Cay Sal records and additional Bahamian material (Buden and Schwartz, 1968). We find that mitochondrial and nuclear sequence data support the assertion that Elbow Cay Tropidophis are conspecific with T. ciirtus (Fig. 5). Because our sampling of the extensive geographic range of T. curtus is coarse grained, we are not able to estimate which populations on the Great Bahamas Bank are most closely related and thus potential progenitors of the Cay Sal popula¬ tion (e.g.. Fig. 6), although morphological data suggest that the Bimini group or New Providence might be the likely origin (Buden and Schwartz, 1968). That the Cay Sal specimens differ morphologically from An- drosian Tropidophis specimens (Buden and Schwartz, 1968; Hedges, 2002) is interesting if we expect that the Cay Sal population is the product of natural dispersal (see Fig. lA for reference). However, it is a possibility, although not mentioned by previous authors (e.g., Buden and Schwartz, 1968), that the snakes were introduced to Elbow Cay during the construction of the lighthouse and associated buildings in the 19th century. Future studies on the phylogeography of this species could incorporate these sequence data to examine these hypotheses. Of note, this species is considered meso- philic (Schwartz and Henderson, 1991; Hen¬ derson and Powell, 2009), and is frequently found after periods of heavy rain or near semipermanent water sources, such as old wells (Henderson and Powell, 2009). Thus, its persistence on the xeric Cay Sal Bank is somewhat perplexing. Study of the natural history of this population is needed to determine how many individuals exist and whether they might differ behaviorally or physiologically from other Bahamian popu¬ lations. Additional squama te species. We were unable to locate two previously recorded species of terrestrial squamates from the Cay Sal Bank. The blindsnake Typhlops himinien- sis biminiensis (Richmond, 1955; see Hedges et al., 2014; Pyron and Wallach, 2014 for taxonomic discussion) was collected on Elbow Cay (KUH 269649-269653) in 1967 by Thomas (1968b). The dwarf gecko Sphaerodaclylus nigropunctatus flavicauda has been recorded from the Cay Sal Bank (Barbour, 1904; Buden and Schwartz, 1968; Thomas, 1968a; Thomas and Schwartz. 1974), although only from Elbow Cay (Buckner et al., 2012). Finally, a newly discovered population of Ciihophis canther- igeriis from the Anguilla Cays on the southeast edge of the Cay Sal Bank was reported in 2015 (Krysko et al., 2015). We did not visit the Anguilla Islands, although it is worth noting that this population derives from Cuban progenitors (on the basis of multilocus genetic analysis; Krysko et al., 2015). Island accounts Great Dog Rock. Great Dog Rock (24°1'48"N, 79°49'47.9994"W) is the largest of several islands marking the eastern edge of the Cay Sal Bank (Figs. 1, 2), and there were no previous herpetofaunal records from these islands (Buckner et al., 2012). These islands are largely devoid of vegetation and 14 BREVIORA No. 560 low lying, and thus are likely washed over during hurrieanes. Great Dog Rock is < 1.0 km", with a patchy covering of ground vegetation. There is a single relatively large (5 m tall, 10 m wide) pyramid-shaped stand of Cocoloba iivifera near the center of the island (Fig. 7A). Sooty terns {Onychoprion jiisccitiis) and brown noddies {Anous stolidus) nest on the cay. We observed adult A. scigrei within the Cocoloba stand, and juveniles and small females on the ground near the scrub vegetation. We found a single juvenile underneath a discarded queen conch {Strom- bcis gigas) shell. We did not observe any other terrestrial reptiles. Cay Sal Island. Cay Sal (23°57'36"N, 80°26'23.9994"W) is the largest island in the region (~1.7 km long and ~1.0 km wide), and likely formed via the deposition of sediment on the southwestern side of the Cay Sal Bank owing to strong and consistent winds from the northeast, moving sand across the bank (Goldberg, 1983). There is a large salina and small brackish lake in the center of the island. Cay Sal previously housed a Bahamas immigration station, although it is now considered uninhabited. Nevertheless, the island is a major camping ground for illegal fishing boats (poachers). We saw ample signs of human disturbance — large trash accumulations, bottles, shotgun and handgun shell casings, and we observed a least one poaching vessel leaving the east side of the island after an overnight anchor¬ age there. Cay Sal is heavily vegetated with silver palms {Cocothrinax argentata), man¬ groves, grasses, shrubs, and some coconut palms (Fig. 7B; also see Buden and Schwartz [1968] for an island description). The soil is sandy and a 10-17-m-high sand ridge runs northwest to southeast and is covered with patches of silver palms. We observed A. fairchildi between 0730h and 1200h in stands of silver palms lining a dense black {Avicin- nia germinans) and white {Laguncularia race- mosa) mangrove stand. We found A. sagrei between 0930h and 1400h in most vegetation types. We observed abundant loggerhead sea turtle {Caretta caretta) tracks and turtle nesting activity on the beaches, and at least one nest appeared to have been excavated by a poacher (Fig. 8). Elbow Cay. Elbow Cay (23°57'36"N, 80°26'23.9994"W) is a relatively elevated island with the ruins of a 19th-century lighthouse and associated outbuildings. These outbuildings are now used to deposit people being trafficked toward the United States, as we observed during our visit. The island is rocky and is covered with low scrubby vegetation, except for a small Casuarina stand near the peak of the island (Fig. 1C). The island presently supports one of the largest known breeding colonies of Audubon’s shearwaters {Puffimis Iherminieri) and is a significant breeding location for at least three other seabird species (Mackin et ah, 2015; Mackin, 2016). We found a very low abundance (fewer than 25 individuals) of A. sagrei near the Casuarina stand. Although there are records of A. fairchildi from Elbow Cay, we did not observe any individuals, and neither did Thomas in 1967 (Buden and Schwartz, 1968). This apparent absence of A. fairchildi is potentially owing to mislabeling of islands on the Cay Sal Bank. For example, Google Earth (2011 imagery, accessed Octo¬ ber 2017) mislabels the Double Headed Shot Cays as Elbow Cay, whereas other references (e.g., Purkis et ah, 2014; Krysko et ah, 2015) do not explicitly label the cay itself. We observed a single Tropidophis ciirtus climbing from beneath a rock into a small shrub at 1025h. We did not observe Typhlops or Sphaerodactylus, both of which have been recorded from the island (Buckner et ah, 2012). East Double Headed Shot Cay. East Double Headed Shot Cay (23°59'23.9994"N, 80°20'24"W), located 9 2018 CAY SAL BANK HERPETOFAUNA 15 Figure 7. Islands surveyed on the Cay Sal Bank. A. Cocoloha stand on Great Dog Rock, the only significant vegetation on this tiny island. B, the southwestern side of Cay Sal Island, with the beach on the left and the ridge lined with palms and mangroves dropping to the salina on the right. C. the Casucirina stand on Elbow Cay. with nesting seabirds in the foreground. The lighthouse ruins on this island are freciuently used to house people being trafficked across the Straits of Florida. D. Elephant Rocks, looking east toward Elbow Cay in the background. Note the steep slopes and relative lack of vegetation; this island is home to a sa.xicolous population of Anolis .sv/g/r/. ARPR is in the foreground for scale. E. Double Headed Shot Cay. with a view of the thick \'egetation. rock piles, and sandy substrate. Photos by RGR and ARPR. km northeast of Elbow Cay, is a component of the necklace of narrow islands forming the northern border of the Cay Sal Bank. This island (actually two islands separated by a very small channel) is long and narrow (~2.8 km long and 15-25 m wide), with signifi¬ cantly more vegetation than other islands on the northern end of the bank. Parts of the 16 BREVIORA No. 560 ♦ Figure 8. Loggerhead sea turtle (Caret to caret tu) tracks on the southwest beach of Cay Sal Island. Note that the nest has been excavated by a poacher. island are quite lush with low vegetation thick enough that it is difficult to penetrate (Fig. 7D). Some larger mangroves (L. race- mosa) are also present. A single herpetofau- nal record exists for this island: Tropidophis curtus. We did not observe this species, although we found a generally low abun¬ dance of A. sagrei, and did not observe any other species of reptiles. Elephant Rocks. Elephant Rocks (23°55A7.9994"N, 80°28'1 1.9994"W) are a series of cays west of Elbow Cay defining the northwestern edge of the Cay Sal Bank (Fig. 2). The islands are relatively tall and very narrow and provide habitat for large densi¬ ties of brown boobies {Sula leucogaster), sooty terns, and brown noddies. Vegetation is scant and what scrub does exist is very low to the ground or grows in crevices. The terrain is steep and requires scrambling up and down jagged cliffs (Fig. 7E). No herpetofaunal records exist for these islands as far as we are aware (Buckner et al., 2012), an unsurprising situation given the apparent lack of suitable habitat and hazardous landing conditions. Nevertheless, we discov¬ ered A. sagrei on two of these cays, although in low abundance. All the individuals we observed were using rocks and small caves as cover and refuge, and were very shy when approached. This is apparently a saxicolous population of A. sagrei and warrants further study. Conclusions A great deal is known about the impact of dispersal on intra- and interspecific diversi- T.\ble 2. T.xxonomy of C.\y Sal Bank tf:rrestrial herpetofauna and likely biogeograpiiic origins of each LINEAGE. Species Origins New Taxon Old Taxon A not is sagrei Western Cuba Anolis sagrei sagrei Anolis sagrei ordinatus Anolis fairchihli Western Cuba — Anolis faircliildi T rop idoph is curtus Great Bahamas Bank — Tropidophis curtus Cuhophis cantherigerus Cuba — Cuhophis cantherigerus Sphaerodactylus Great Bahamas Bank — Sphaerodactylus nigropuuctatus flavicauda (presumed) nigropunctat us flavicauda T vp/dops hitiiinietisis hiudfiiensis Great Bahamas Bank (presumed) Typhlops hinnniensis hiininiensis 2018 CAY SAL BANK 1 ILRPHTOFAUNA 17 fication in West Indian Aiwlis lizards (Losos, 2009), yet many species and populations remain understudied. On the remote Cay Sal Bank, we find that the two species of anoles occurring there are a product of dispersal trom western Cuba, contrary to the previ¬ ously assumed Bahamian origin for A. scigrei. Although we refrain from sinking the specific epithet A. faircliildi into A. porcatus. largely owing to the necessity for further systematic study of the porcatus complex (Glor et ah, 2005), we do recognize the brown anole populations as A. sagrei scigrei instead of A. scigrei ordincitiis (Table 2). We further find that the Tropidophis snake from Elbow Cay is likely conspecific with Great Bahamas Bank T. ciirtiis. Finally, we have added new island herpetofaunal records for the region, although we note that invasive predators (Rcittiis sp.) are recorded from most islands in the region and could be affecting terrestrial and marine reptiles (Mackin et ah, 2015). CONFLICT OF INTEREST The authors declare that they have no conflict of interest. ACKNOWLEDGMENTS We thank Captain Wasson and the crew of the M/V Spree for transit to and from the Cay Sal Bank, zodiac runabout services, and safety oversight. We especially thank .lona- than Losos for advice and support for travel to the region, as well as the support of the Museum of Comparative Zoology. We are grateful for funding from the Putnam Fund for Research and Exploration from the Museum of Comparative Zoology (to RGR), as well as the John Templeton Foundation (to Jonathan B. Losos). We thank the Bahamas Department of Agricul¬ ture, the Bahamas Environment, Science and Technology Commission, Ministry of the Environment, and the Bahamas National Trust for research and export permits. We also thank Sandra Buckner for discussions and advice related to this work. LITERATURE CITED Agassiz, A. 1894. Reconnaissance of ihe Bahamas and the elevated reefs of Cuba in the steam yacht "Wild Duck" January to April. 1893. Bullclin of the Miiseiini of C'oDi/Hirative Zoolof^y 26: 1-204. Barbour. T. 1904. Batrachia and Reptilia from the Bahamas. Bulletin of the Museum of Comparative Zoology 46; 55-61. Barbour. T. 1906. Additional notes on Bahama snakes. Ameriean Naturalist 40: 229-232. Barbour. T. 1937. Third list of Antillean reptiles and amphibians. Bulletin of the Museum of Comparative Zoology 82: 77-166. Barbour, T., and B. Shreve. 1935. Concerning some Bahamian reptiles, with notes on the fauna. Proeeedings of the Boston Soeiety of Natural History 5: 347-365. Bruckner A.. J. Kerr. G. Rowlands. A. Dempsey. S. Purkis. and P. Renaud. 2014. Atlas of shallow marine habitats of Cay Sal Bank. Great Inagua. Little Inagua and Hogsty Reef Bahamas. Phoeni.x. Arizona: Panoramic Press. Buckner, S. D.. R. Franz, and R. G. Reynolds. Bahama Islands and Turks & Caicos Islands. Bulletin of the Florida Museum of Natural History 51: 85-166. Buden. D. W.. and A. Schwartz. 1968. Reptiles and birds of the Cay Sal Bank. Bahama Islands. Quarterly Journal of the Florida Academy of Sciences 3 1 : 290-320. Burbrink. F. T. 2004. Inferring the phylogenetic position of Boa constrictor among the Boinae. Molecular Phylogenetics and Evolution 34: 167-180. Campbell, B. N. 1997. Hie sunt serpentes — molecular phylogenetics and the Boidae (Serpentes; Booidae). Ph.D. Dissertation. Kingston. Ontario. Canada. Queen's University. Cochran. D. M. 1934. Herpetological collections from the West Indies made by Dr. Paul Bartsch under the Walter Rathbone Bacon Scholarship. 1928-1930. Smithsonian .Miscellaneous Collection 92: 1-48. Cory. C. B. 1891. On a collection of birds made on the islands of Anguilla and Cay Sal or Salt Cav. Bahama Islands, by Mr. Cyrus S. Winch, during May. 1891. .J/A 8: 352. Darriba. D.. G. L. Taboada. R. Doallo. and D. Posada. 2012. jModelTest 2: more models, new heuristics and parallel computing. Nature Methods 9: 772. 18 B REV 10 R A No. 560 Drummond, A. J., M. A. Suchard, D. Xie, and A. Rambaut. 2012. Bayesian phylogenetics with BEAUTi and the BEAST 1.7. Molecular Biology and Evolution 29: 1969-1973. Eelsenstein. J. 2004. Inferring Phylogenies. Sunderland, Massachusetts: Sinauer Associates. Gartner, G. E. A., T. Gamble, A. Jaffe, A. Harrison, and J. B. Losos. 2013. Left-right dewlap asymmetry and phylogeography of Anolis lineatus on Aruba and Cura9ao. Biological Journal of the Linnean 5o(/>/v 110: 409^26. Geneva, A. J., J. Hilton, S. Noll, and R. E. Glor. 2015. Multilocus phylogenetic analyses of Hispaniolan and Bahamian trunk anoles {disticlnis species group). Molecular Phylogenetics and Evolution 87: 105-117. Geneva, A. J., R. G. Reynolds, N. C. Herrmann, A. R. Puente-Rolon, and K. .1. Aviles-Rodriguez. 2016. Henudactylus mabouia (Woodslave). Distribution. Herpetological Review 47: 423-424. Glor, R. E., M. E. Gifford, A. Larson, J. B. Losos, L. R. Schettino, A. R. C. Lara, and T. R. Jackman. 2004. Partial island submergence and speciation in an adaptive radiation: a multilocus analysis of the Cuban green anoles. Proceedings of the Royal Society BllV. 2257-2265. Glor, R. E., J. B. Losos, and A. Larson. 2005. Out of Cuba: overwater dispersal and speciation among lizards in the Anolis carolinensis subgroup. Molec¬ ular Ecology 14: 2419-2432. Goldberg, W. M. 1983. Cay Sal Bank, Bahamas: a biologically impoverished, physically controlled environment. Atoll Research Bulletin 271: 1-23. Guindon, S., and O. Gascuel. 2003. A simple, fast and accurate method to estimate large phylogenies by maximum-likelihood. Systematic Biology 52: 696- 704. Hedges, S. B. 2002. Morphological variation and the definition of species in the snake genus Tropidophis (Serpentes, Tropidophiidae). Bulletin of the Natural History Museum of London 68: 83-90. Hedges, S. B., and C. E. Conn. 2012. A new skink fauna from Caribbean islands (Squamata, Mabuyidae, Mabuyinae). Zootaxa 3288: 1-244. Hedges, S. B., A. B. Marion, K. M. Lipp, J. Marin, and N. Vidal. 2014. A taxonomic framework for typhlopid snakes from the Caribbean and other regions (Reptilia, Squamata). Caribbean Herpetol¬ ogy A9: 1-61. Henderson, R. W., and R. Powell. 2009. Natural History of Vest Indian Reptiles and Amphibians. Gaines¬ ville: University Press of Elorida. Hower, L. M., and S. B. Hedges. 2003. Molecular phylogeny and biogeography of West Indian Teiid lizards of the genus Ameiva. Caribbean Journal oj Science 29: 298-306. Knapp, C. R., J. B. Iverson, S. D. Buckner, and S. V. Cant. 2011. Conservation of amphibians and reptiles in the Bahamas. Pp. 53-87 in: A. Hailey, B. S. Wilson, and J. A. Horrocks, editors. Conservation of Caribbean Island Herpetofamtas, vol. 2. Leiden, the Netherlands: Brill. Kolbe, J. J., R. E. Glor, L. R. Schettino, A. C. Lara, A. Larson, and J. B. Losos. 2004. Genetic variation increases during biological invasion by a Cuban lizard. Nature 431: 177-181. Kolbe, J. J., R. E. Glor, L. R. Schettino, A. C. Lara, A. Larson, and J. B. Losos. 2007. Multiple sources, admixture, and genetic variation in introduced Anolis lizard populations. Conservation Biology 21: 1612-1625. Krysko, K. L., D. W. Steadman, L. P. Nuiiez, and D. S. Lee. 2015. Molecular analyses of Caribbean dipsa- did (Xenodontinae: Alsophiini) snakes, including identification of the first record from the Cay Sal Bank, The Bahamas. Zootaxa 4028: 441-450. Larkin, M. A., G. Blackshields, N. P. Brown, R. Chenna, P. A. McGettigan, H. McWilliam, F. Valentin, 1. M. Wallace, A. Wilm, R. Lopez, J. D. Thompson, T. J. Gibson, and D. G. Higgins. 2007. Clustal W and Clustal X version 2.0. Bioinformatics 23: 2947-2948. Librado, P., and J. Rozas. 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25: 1451-1452. Losos, J. B. 2009. Lizards in an Evolutionary Tree. Berkeley: University of California Press. Macey, J. R., A. Larson, N. B. Ananjeva, and T. J. Papenfuss. 1997. Evolutionary shifts in three major structural features of the mitochondrial genome among iguanian lizards. Journal of Molecular Evolution 44: 660-674. Macey, J. R., J. A. Schulte 11, N. B. Ananjeva, A. Larson, N. Rastegar-Pouyani, S. M. Shammakov, and T. .1. Papenfuss. 1998. Phylogenetic relation¬ ships among agamid lizards of the Laudakia caucasia species group: testing hypotheses of biogeographic fragmentation and an area dado- gram for the Iranian Plateau. Molecular Phyloge¬ netics and Evolution 10: 118-131. Mackin. W. A. 2016. Current and former populations of Audubon's Shearwater (Puffinus Iherminieri) in the Caribbean region. Condor 1 18: 655-673. Mackin, W. A., P. Moore, D. S. Lee, and L. M. Ferguson. 2015. Seabirds of the Cay Sal Bank, The Bahamas. IVaterbirds 38: 407-414. MacLean, W. P., R. Kellner, and H. Dennis. 1977. Island lists of West Indian amphibians and reptiles. 2018 CAY SAL BANK 1 ILRPHTOFAUNA 19 Sniith.sonidn Hcrpctologiccil Information Services 40: 1-47. Nicholson, K. E.. B. 1. Crother, C. Guyer. and .1. M. Savage. 2012. It is time for a new classification of anoles (Squamata: Dactyloidae). Zoolaxa 3477: I- 108. Oliver. J. A. 1948. The anoline lizards of Bimini. Bahamas. American Museiini Novitates 1383: 1-36. Purkis. S.. J. Kerr. A. Dempsey. A. Calhoun. L. Metsamaa. B. Riegl. V. Kourafalou. A. Buckner, and P. Renaud. Large-scale carbonate platform development of Cay Sal Bank, Bahamas, and implications for associated reef geomorphology. Geomorphology 222: 25-38. Pyron, R.. and A. V. Wallach. 2014. Systematics of the blindsnakes (Serpentes: Scolecophidia: Typhlopoi- dea) based on molecular and morphological evi¬ dence. Zootaxa 3829: 001-081. Rambaut, A.. M. A. Suchard, D. Xie, and A. J. Drummond. 2013. Tracer vL5. Available from: http://beast.bio.ed.ac.uk Tracer Revell. L. J.. L. J. Harmon, R. B. Langerhans. and J. J. Kolbe. 2007. A phylogenetic approach to deter¬ mining the importance of constraint on phenotypic evolution in the Neotropical lizard Anolis eristatel- liis. Evolutionary Ecology Research 9: 261-282. Reynolds, R. G., N. C. Herrmann. A. R. Puente-Rolon, A. J. Geneva, and K. J. Aviles- Rodriguez. 2016a. Osteopihis septetitrionalis (Cuban Treefrog). Distri¬ bution. Herpetological Review 47: 420. Reynolds. R. G.. M. L. Niemiller. S. B. Hedges. A. Dornburg. A. R. Puente-Rolon, and L. J. Revell. 2013. Molecular phylogeny and historical biogeog¬ raphy of West Indian boid snakes {Chilahothnis). Molecular Phylogenetics anti Evolution 68: 461-470. Reynolds, R. G.. M. L. Niemiller, and L. ,1. Revell. 2014. Toward a Tree-of-Life for the boas and pythons: multilocus species-level phylogeny with unprece¬ dented taxon sampling. Molecular Phylogenetics ant! Evolution 71: 201-213. Reynolds. R. G. and A. R. Puente-Rolon. 2016a. Chilahothnis strigilatus (Bahamas Boa). Distribu¬ tion. Herpetological Review 47: 425. Reynolds. R. G. and A. R. Puente-Rolon. 2016b. Anole surveys on the Cay Sal Bank. Bahamas. Blog Post 08-03-2016 [http://ww\v. anoleannals.org 2016 08 03/anole-surveys-on-the-cay-sal-bank-bahamas]. Reynolds, R. G.. A. R. Puente-Rolon. K. J. Aviles- Rodriguez. A. .1. Geneva, and N. C. Herrmann. 2016b. Discovery of a remarkable new boa Irom the Conception Island Bank. Bahamas. Breviora 549: 119. Richmond, N. D. 1955. fhe blind snakes (Typhlops) of Bimini. Bahama Islands. British West Indies, with description of a new- species. American Museum Novitates 1734: 1-7. Schw'artz, A., and R. W. Henderson. 1991. Amphibians and Reptiles of the li’est Indies: Descriptions. Distributions, and Natural History. Gainesville: University Press of Florida. Stamatakis. A. 2006. RaxML-VI-HPC: maximum like¬ lihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22: 2688- 2690. Stephens. M. and P. Donnelly. 2003. A comparison of Bayesian methods for haplotype reconstruction from population genotype data. American Journal of Human Genetics 73: 1 162-1 169. Stephens, M.. N. .1. Smith, and P. Donnelly. 2001. A new statistical method for haplotype reconstruction from population data. American Journal of Human Genetics 68: 978-989. Thomas. R. 1968a. Notes on Antillean geckos (Sphaer- odactyhis). Herpetologica 24: 46-60. Thomas, R. 1968b. The Typhlops biminiensis group of Antillean blind snakes. Copeia 1968: 713-722. Thomas, R.. and A. Schwartz. 1974. The status of Sphaerodaetylus gilvitorques Cope and of Sphaer- odaetylus nigropunetatus Gray (Sauria: Gekkoni- dae). Journal of Herpetology 8: 353-358. Tolson. P. J., and R. W. Henderson. 1993. The Natural History of fFc.v/ Indian Boas. Taunton. Somerset. England: R&A Publishing Ltd. Williams, E. E. 1969. The ecology of colonization as seen in the zoogeography of anoline lizards on small islands. Quarterly Review of Biology 44: 345-389. Winchell. K. M.. P. Muralidhar. J. R. Fredette. and R. G. Reynolds. 2015. New herpetofaunal records for Great Isaac Cay. Bimini Group. Bahamas. IRCE Reptiles Amphibians: Conservation and Natural History 22: 106-1 10. •i“, ; iff-' f. ■ ■ ‘ =v^Ay.?