Out of Asia: Natricine Snakes Support the Cenozoic Beringian Dispersal Hypothesis

Home , Adelophis, Atretium, Colubridae, Natricinae, Natriciteres, Natrix maura

Molecular Phylogenetics and Evolution 63 (2012) 825–833

Contents lists available at SciVerse ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier.com/locate/ympev

Out of Asia: Natricine snakes support the Cenozoic Beringian Dispersal Hypothesis

Peng Guo a,b,1, Qin Liu a,1, Yan Xu a,c, Ke Jiang d, Mian Hou e, Li Ding b, R. Alexander Pyron f, ⇑ Frank T. Burbrink g,h, a College of Life Sciences and Food Engineering, Yibin University, Yibin 644007, China b Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu 610041, China c Chengdu University of Technology, Chengdu 610050, China d State Key Laboratory of Genetic Resources and Evolution, Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming 650223, China e Sichuan Normal University, Chengdu 650068, China f Department of Biological Sciences, The George Washington University, 2023 G St. NW, Washington, DC 20052, USA g Department of Biology, The Graduate School and University Center, The City University of New York, 365 5th Ave., New York, NY 10016, USA h Department of Biology, The College of Staten Island, The City University of New York, 2800 Victory Blvd., Staten Island, NY 10314, USA article info abstract

Article history: Based on a combination of six mitochondrial gene fragments (12S RNA, cyt b, ND1, ND2, ND4 and CO1) Received 30 September 2011 and one nuclear gene (c-mos) from 22 genera we infer phylogenetic relationships among natricine snakes Revised 16 February 2012 and examine the date and area of origin of these snakes. Our phylogenetic results indicate: (1) the sub- Accepted 22 February 2012 family Natricinae is strongly supported as monophyletic including a majority of extant genera, and a Available online 12 March 2012 poorly known and previously unassigned species Trachischium monticola, (2) two main clades are inferred within Natricinae, one containing solely taxa from the Old World (OW) and the other comprising taxa Keywords: from a monophyletic New World (NW) group with a small number of OW relatives. Within the ﬁrst clade, Colubridea the genera Xenochrophis and Amphiesma are apparently not monophyletic. Divergence dating and ances- Amphiesma Trachischium tral area estimation indicate that the natricines originated in tropical Asia during the later Eocene or the Biogeography Oligocene. We recover two major dispersals events out of Asia, the ﬁrst to Africa in the Oligocene (28 Ma) Phylogenetics and the second to the Western Palearctic and the New World at 27 Ma. This date is consistent with the Intercontinental dispersal dispersal of numerous other OW groups into the NW. Beringia Ó 2012 Elsevier Inc. All rights reserved.

1. Introduction many organisms, though, a basic understanding of their historical biogeography is unclear, and the commonality of various patterns A major goal of historical biogeography is to understand the and processes that have shaped their modern distribution is largely causes for common spatial and temporal patterns across taxa unknown. (Donoghue, 2001; Wiens and Donoghue, 2004; Lomolino et al., Globally distributed groups that are both diverse and ancient 2006). Using phylogenetic inference, divergence dating, and ances- provide a prime opportunity to examine comparative historical tral area estimation, the various biogeographic factors including biogeographic processes related to area of origin and inter-regional vicariance, dispersal, speciation, and extinction can be investigated dispersal. A common distributional pattern discovered among var- in an evolutionary context. Moreover, understanding the historical ious unrelated groups including insects, mammals, snakes, lizards biogeography of a group provides the structure for testing when and plants, is the tendency to have closely related taxa found and how adaptive radiations occur, how communities assemble throughout the Holarctic (Enghoff, 1995; Sanmartin et al., 2001; with respect to regional taxonomic pools, and how species richness Cook et al., 2005; Smith et al., 2005; Burbrink and Lawson, 2007; is controlled by areas of origin, rates of diversiﬁcation, and niche Brandley et al., 2011). These groups may also include members conservatism (Blackburn and Gaston, 2004; Webb et al., 2006; that have distributions ranging well into the Neotropics or Paleo- Burbrink and Lawson, 2007; Mittelbach et al., 2007; Pyron and tropics. While timing and direction of colonization may be variable Burbrink, 2009a; Wiens et al., 2009; Kozak and Wiens, 2010). For across these groups, particularly given that dispersal across the Holarctic in endotherms was not limited by the onset of glaciers during the Pliocene and Pleistocene (Sanmartin et al., 2001), ⇑ Corresponding author at: Department of Biology, The College of Staten Island, remarkably, a number of taxa exhibit unidirectional dispersal The City University of New York, 2800 Victory Blvd., Staten Island, NY 10314, USA. across the Holarctic within a restricted timeframe. Fax: +1 718 9823961. Speciﬁcally, several squamate groups, including skinks (Plesti- E-mail address: [email protected] (F.T. Burbrink). odon), ratsnakes (NW Lampropeltini and OW Elaphe), and pitvipers 1 Equal contribution for this work.

(Crotalinae), which are distributed throughout the Holarctic with in conjunction with the other groups likely indicates that entire adjacent tropical ranges, have shown similar patterns of dispersal squamate communities originated in the OW and existed across and diversification (Burbrink and Lawson, 2007; Wüster et al., Beringia and North America during the early Cenozoic. 2008; Pyron and Burbrink, 2009b; Brandley et al., 2011). From these studies, there are two common patterns that may be influ- 2. Methods and materials enced by similar processes: (1) all apparently originated in tropical or subtropical Asia from the late Eocene through the late Oligo- 2.1. Taxon sampling cene, and (2) all display a single unidirectional dispersal through Beringia into the New World (NW) during the late Oligocene or Here we follow Pyron et al.’s (2011) taxonomy for Colubridae. early Miocene, through the mixed mesophytic forests that domi- We obtained sequences for 80 individuals of 72 natricine species nated that area at the time. The collapse of these habitats by the from 21 genera and three outgroup species, for the nuclear pro- mid-Miocene severed connections among the OW and NW com- tein-coding genes oocyte maturation factor Mos (c-mos), the mito- munities, thus yielding disjunct Holarctic patterns seen today (San- chondrial protein-coding genes cytochrome b (cyt b), NADH martin et al., 2001). While distributional patterns and dispersal subunit 4 (ND4), NADH subunit 2 (ND2), cytochrome oxidase sub- within the OW ranges may differ among these groups, the single unit 1 (CO1) and 12S ribosomal RNA (12S RNA). Sequences for 62 area and timing of origin as well as dispersal to the New World individuals of the colubroid species were obtained from GenBank. is consistent in spite of the separate molecular trees each cali- We sequenced c-mos, ND2, ND4 and cyt b for 18 additional indi- brated with different fossils. viduals (15 species, 7 genera). To investigate the monophyly of Another large squamate group with a Holarctic distribution are Natricinae and to clarify the systematic position of Trachischium the watersnakes, Natricinae, which represent a diverse group of monticola, a poorly known putative colubrid, two representatives colubrid snakes composing 210 species in 29 genera found from the other subfamilies of Colubridae (Immantodes cenchoa throughout Asia, Europe, North Africa, Sub-Saharan West Africa, and Coluber constrictor) as well as one homalopsid (Enhydris plum- North America and Central America (The Reptile Database: bea), were chosen as outgroups based on previous studies (Lawson http://www.reptile-database.org). Although many of the species et al., 2005; Pyron et al., 2011). are associated with aquatic habitats and have diets consisting of amphibians or fish, many others are found in dry habitats and may consume invertebrates and mammals (Rossman et al., 1996; 2.2. DNA extraction, amplification and sequencing Gibbons and Dorcas, 2004; Vitt and Caldwell, 2009). Estimates using molecular phylogenies place the origin of the natricines from Total DNA was extracted from 85%-alcohol-preserved livers or other colubrids in the Eocene/Oligocene, indicating they are young muscle tissues by the standard method of proteinase K and phe- enough to have encountered the same environmental conditions as nol/chloroform (Sambrook and Russell, 2002). The entire gene se- the ratsnakes, skinks and crotaline snakes (Pyron and Burbrink, quences for the mitochondrial cytochrome b (cyt b) and NADH 2012). dehydrogenase subunit 2 (ND2), the partial gene sequences of Previous research using morphological and molecular data have NADH dehydrogenase subunit 4 (ND4) and one nuclear gene indicated that the New World natricines form a monophyletic c-mos were amplified by the polymerase chain reaction (PCR) using group (i.e., Thamnophiini) which might indicate that they origi- primers L14910/H16064 (Burbrink et al., 2000), L5238/H5382 (de nated as a single dispersal from OW groups (Rossman and Eberle, Queiroz et al., 2002), ND4/Leu (Arèvalo et al., 1994) and S77/S78 1977; Lawson, 1987; Alfaro and Arnold, 2001; de Queiroz et al., (Lawson et al., 2005), respectively. The cycling parameters were 2002). Malnate (1960) previously proposed that the natricines identical to those described in above studies. Prior to sequencing, originated in Asia and dispersed from there to Europe, Africa, Aus- PCR products were purified using various commercial kits. The tralia and North America, and that Natrix (sensu lato) dispersed to double-stranded product was sequenced using an ABI 3730 North America over the Bering Strait. Given the timing of origin of Genetic Analyzer (Applied Biosystems) following manufacturer’s the global natricines in Eocene, it is also possible that they could protocols. have colonized the New World via the Atlantic through the Thu- lean Land Bridge or though the Pacific via Beringia (Nilsen, 1978; 2.3. Phylogenetic analyses Sanmartin et al., 2001). However, dispersals after the first half of the Cenozoic are unlikely to have occurred via the Thulean Land Alignment of protein coding genes was trivial as there were no Bridge, due to declining environmental conditions suitable for indels and all sequences were in reading frame. The ribosomal ectotherms (Burbrink and Lawson, 2007). Crucially, relationships gene 12S was aligned using the default high-accuracy parameters among OW genera and Thamnophiini are unclear, and divergence in MUSCLE (Edgar, 2004). Phylogenetic analyses were performed dating and ancestral area reconstructions are needed to under- using two different methods including Bayesian Inference (BI) stand basic biogeographic patterns relating to areas of origin, dis- and Maximum Likelihood (ML). For the Bayesian analyses, four persal, and the rise of the several regional assemblages. partitioning strategies were evaluated: all data combined (concat- Here, we examine the temporal and geographic origins of natri- enated analysis), two partitions (mtDNA and nDNA), seven parti- cine snakes in order to provide a comparison to the patterns and tions (12S RNA, cyt b, ND1, ND2, ND4, CO1 and c-mos) and 19 timing of origin of the ratsnakes, skinks, and crotalines, which all partitions (protein coding genes 1st, 2nd and 3rd position sepa- demonstrate similar timing and routes of dispersal through Berin- rately and 12S RNA). The best-fit substitution model was assigned gia to the New World. In addition to sequencing several new spe- to each partition using AIC in MrModeltest 2.3 (Nylander et al., cies, we test hypotheses about the timing and area of origin of 2004; Posada, 2008). The optimal partition model was chosen natricines in the OW. We also examine patterns and timing of dis- using Bayes factors (BF) method as described in Brown and Lem- persal in the OW. Finally, we attempt to understand the causes of mon (2007). We used MrBayes 3.1.2 (Huelsenbeck and Ronquist, their Holartic distribution, which for the other groups of squamates 2001; Ronquist and Huelsenbeck, 2003) to estimate trees for each have suggested a single unidirectional dispersal from Asia through partitioning strategy, for three independent runs, each initiated Beringia to the New World during the Oligocene and Miocene, with random trees. All searches consisted of four Markov chains which we refer to as the Cenozoic Beringial Dispersal Hypothesis (three heated chains and a single cold chain) estimated for 20 (CBDH). We find strong support for the CBDH in Natricinae, which million generations and sampled every 2000 generations with P. Guo et al. / Molecular Phylogenetics and Evolution 63 (2012) 825–833 827

25% initial samples discarded as burn-in. Substitution parameters and Dorcas, 2004). Dispersal probabilities between areas were were unlinked and rates were allowed to vary across partitions. not constrained to avoid excessive parameterization, as the Stationarity was conﬁrmed by plotting the likelihood against geographical formation of the targeted areas remained constant generation in the program Tracer v1.4 (Drummond and Rambaut, over the time period covered by the natricines. Similarly, dispersal 2007). After conﬁrming that three analyses reached stationarity among regions was not weighted by direction or by time, for at a similar likelihood score and the topologies were similar, the same reason. Dispersal and extinction rates were optimized the remaining trees were combined to calculate posterior by the program, and no ancestral area combinations were probabilities (PP) for each node in a 50% majority-rule consensus excluded. tree. Additionally, BI was also run in BEAST (see below). The ML tree and bootstrap support were obtained under the optimal 3. Results partitioning strategy in RAxML, using the GTRGAMMA model with 25 rate categories (Stamatakis et al., 2008). 3.1. Sequence data and alignment

2.4. Divergence time estimation The final alignment of seven gene fragments consisted of 6243 aligned base pairs. Of them, 583 bp were from c-mos (38 taxa), The program BEAST v 1.6.2 (Drummond and Rambaut, 2007) 1094 bp from cyt b (80 taxa), 696 bp from ND4 (59 taxa), was used to estimate divergence dates in the natricines, using 1032 bp from ND2 (69 taxa), 877 bp from CO1 (17 taxa) and the relaxed phylogenetics method (Drummond et al., 2006), where 997 bp from 12S RNA (40 taxa). The concatenated dataset con- rates of evolution are independently drawn from a lognormal dis- tained 3105 variable sites (50%, including outgroups), of which tribution while accounting for phylogenetic uncertainty. Each gene 2366 (38%) were informative under MP criteria. No deletions, was provided with its own partition and model (GTR + C + I) and insertions or stop codons were detected, indicating that uninten- rates and times were estimated under the uncorrelated lognormal tional amplification of pseudogenes was unlikely. New sequences tree-prior, with a birth–death prior on speciation rates and Jeffrey’s generated here were deposited in GenBank (Table 1, Accession priors on the substitution model parameters. Nos. JQ687411-JQ687524; GQ281776-GQ281787). Three calibrations placed at the stem of defined groups were obtained from the paleoherpetolgical record of natricines. The oldest fossil of a stem natricine, Natrix mlynarskii, which also coincides 3.2. Phylogenetic analysis with the oldest colubrine and the potential sister subfamily natricines, is from 30 to 32 Ma (Rage, 1988). While the fossil was The best-fit model was determined for each partition (Table 2), placed in the genus Natrix, an extant genus, it is more likely a rep- and Bayes factors indicated that the preferred partitioning model resentative of a stem natricine (J. Head, pers. comm.). Therefore, we included the 19-partition model, used for the final BI and ML anal- placed this fossil at the MRCA of all Natricinae with a lognormal yses. All analyses showed strong support (100% PP) for a monophy- mean of 30 Ma with a standard deviation (SD) of 0.115, yielding letic Natricinae, including the newly sequenced Trachischium a 95% CI from 23 to 36 Ma, which does not exceed the earliest col- monticola (Fig. 1). Within Natricinae, the earliest division yields ubrid fossil. The oldest member of the extant genus Natrix is N. san- two highly supported clades, one consisting entirely of OW taxa saniensis or N. merkurensis, both dated at 22 Ma (Szyndlar, 2009). (clade B) including Afronatrix, Amphiesma, Atretium, Macropisth- This fossil was placed at the mrca of Natrix on the tree and we pro- odon, Natriciteres, Rhabdophis, Trachischium and Xenochrophis, and vided a lognormal mean of 22 Ma and a SD of 0.15, yielding a 95% another (Clade A) consisting of the remainder of the OW and all CI for that node of 16–28 Ma, which does not exceed the age of the NW species. Within clade B, support for most nodes was high. Sev- oldest natricine. Finally, the oldest member of the New World eral genera, however, within this group were not monophyletic, genus, Thamnophis is known to up 16 Ma (Holman, 2000). There- including Xenochrophis which is paraphyletic, and Amphiesma fore, for the MRCA of Thamnophis, we provided a mean lognormal which is polyphyletic. Additionally, a clade composed of Amphi- date of 16 Ma, with a SD of 0.15, yielding a 95% CI range of 12– esma sauteri and A. craspedogaster (GP 139) were sister to Trac- 20 Ma, not exceeding the date of the oldest combined New and hischium monticola to the exclusion of another sample of A. Old World genus Neonatrix in North America (Holman, 2000). We craspedogaster (GP 600). The other major clade (clade A) is com- ran these analyses two times for 50 million generations, each proposed of a strongly supported monophyletic NW Thamnophiini duced similar results and yielded effective sample size (ESS) of the (A1), which is sister to the monophyletic OW genus Natrix (A2). posterior probability distribution of all parameters >200 (Drum- Two Asian genera, Sinonatrix and Opisthotropis, are each strongly mond et al., 2006), as calculated in Tracer v1.4 (Drummond and supported as monophyletic, and together form the sister group to Rambaut, 2007). the clade composed of Natrix and Thamnophiini. Similar to Pyron et al. (2011), several thamnophine genera are paraphyletic, includ- 2.5. Ancestral-area estimation ing Thamnophis, Nerodia, and Regina.

We used Lagrange 2.01 (Ree et al., 2005; Ree and Smith, 2008) 3.3. Divergence dating and ancestral-area estimation to examine the area of origin and subsequent dispersal of natricines throughout their modern range in a likelihood context, where The two BEAST runs converged after 5 Â 107 generations, yield- the impact of extinction and dispersal rates is considered along ing ESS values > 200 for all parameters, and producing highly con- with probabilistic estimates of within-region diversiﬁcation. Spe- gruent trees and dates. Additionally, Lagrange yielded a global lnL cies were coded to one of 6 standard zoogeographic regions recog- at the root node for the natricine ingroup of À167.8 with dispersal nized by Wallace (1876), Barry (2001) and Kreft and Jetz (2010), and extinction rates estimated at 0.008832 and 4.285 Â 10À9 which have been shown to be useful at estimating ancestral areas respectively. Results indicate that the natricines likely originated in other snake groups (Burbrink and Lawson, 2007): Neotropic, in Asia (lnL = À169.3 with a relative probability [RP] = 0.21; Node Nearctic, Western Palearctic, Ethiopian (African), Eastern Palearc- 1, Fig. 2; Table 3) during the late Eocene or the Oligocene tic, and Oriental. These ranges were determined using the Reptile (33 Ma; 95% CI 28–40 Ma; Fig. 2). Two major dispersals out of Asia Database and associated literature included with each taxon occurred during the Oligocene, with one dispersal event to the (http://www.reptile-database.org Rossman et al., 1996; Gibbons Western Palearctic and the New World (Node 2, Fig. 2; 828 P. Guo et al. / Molecular Phylogenetics and Evolution 63 (2012) 825–833

Table 1 Samples and sequence sources for the analyses used in this paper.

Species Family Subfamily Voucher 12S cmos COI cytb ND1 ND2 ND4 Coluber constrictor Colubridae Colubrinae AY122667 AY486937 AY122649 AY486913 AY486962 AY487001 AY487040 Imantodes cenchoa Colubridae Dipsadinae EU728586 GQ457865 EU728586 EU728586 EU728586 EU728586 EU728586 Adelophis foxi Colubridae Natricinae AF420069 AF420070 AF420071 AF420072 Afronatrix anoscopus Colubridae Natricinae AF471123 AF420073 AF420074 AF420075 AF420076 Amphiesma craspedogaster 1 Colubridae Natricinae GP139 JQ687437 JQ687429 JQ687412 Amphiesma craspedogaster 2 Colubridae Natricinae GP600 JQ687443 GQ281781 JQ687459 JQ687418 Amphiesma sauteri Colubridae Natricinae AF402622 AF402905 AF384824 Amphiesma stolatum Colubridae Natricinae AF471097 AF471030 Amphiesma stolatum 1 Colubridae Natricinae GP901 JQ687450 JQ687432 JQ687464 JQ687425 Atretium yunnanensis 1 Colubridae Natricinae GP842 JQ687448 GQ281787 JQ687463 JQ687423 Clonophis kirtlandii Colubridae Natricinae AF402625 AF402908 AF384827 Lycognathophis Colubridae Natricinae FJ387220 seychellensis Macropisthodon rudis 1 Colubridae Natricinae GP384 JQ687442 GQ281780 JQ687458 JQ687417 Macropisthodon rudis 2 Colubridae Natricinae GP1266 JQ687452 JQ687434 JQ687427 Natriciteres olivacea Colubridae Natricinae AF544772 AF471146 AF471058 Natrix maura Colubridae Natricinae AF402623 AY866530 AY873742 AY870616 EU437572 Natrix natrix Colubridae Natricinae AY122682 AF471121 AY122664 AF471059 AY873760 AY870640 AY873736 Natrix tessellata Colubridae Natricinae EU119168 AY873772 AY870641 AY873735 Nerodia cyclopion Colubridae Natricinae AF402626 GQ279084 AF402909 AF384828 Nerodia erythrogaster Colubridae Natricinae AF402629 GQ279023 GQ285504 HQ121595 GQ285402 AF420084 Nerodia fasciata Colubridae Natricinae AF402627 GQ279080 AY866529 AY873738 AY870612 AY873705 Nerodia ﬂoridana Colubridae Natricinae AF402628 AF402911 AF384830 Nerodia harteri Colubridae Natricinae AF402652 AF402935 AF384854 Nerodia rhombifer Colubridae Natricinae GQ279077 GQ285446 HQ121995 GQ285289 Nerodia sipedon Colubridae Natricinae AF402630 GQ278935 AF402913 HQ121602 DQ915161 Nerodia taxispilota Colubridae Natricinae AF402631 GQ279082 AF402914 HQ121606 AF384833 U49322 Opisthotropis cheni 1 Colubridae Natricinae GP383 JQ687441 GQ281779 JQ687457 JQ687416 Opisthotropis guangxiensis 1 Colubridae Natricinae GP746 JQ687447 GQ281776 JQ687462 JQ687422 Opisthotropis lateralis 1 Colubridae Natricinae GP646 JQ687445 GQ281782 JQ687461 JQ687420 Opisthotropis latouchii 1 Colubridae Natricinae GP647 JQ687446 GQ281783 JQ687421 Regina alleni Colubridae Natricinae AF402633 AF402916 AF384835 Regina grahami Colubridae Natricinae AF402635 AF402918 HQ121593 AF384837 Regina rigida Colubridae Natricinae AF402636 AF471120 AF471052 AF384838 Regina septemvittata Colubridae Natricinae AF402634 AF402917 AF384836 Rhabdophis nuchalis Colubridae Natricinae AF402624 AF402907 AF384826 Rhabdophis nuchalis 1 Colubridae Natricinae GP251 JQ687438 GQ281786 JQ687454 JQ687413 Rhabdophis subminiatus Colubridae Natricinae AF544776 AF544713 U49325 Rhabdophis subminiatus 1 Colubridae Natricinae GP57 JQ687436 GQ281777 JQ687411 Rhabdophis tigrinus Colubridae Natricinae AF236679 AF471119 AF471051 Rhabdophis tigrinus 1 Colubridae Natricinae GP613 JQ687444 GQ281785 JQ687460 JQ687419 Seminatrix pygaea Colubridae Natricinae AF402637 AF402920 AF384839 Sinonatrix aequifasciata 1 Colubridae Natricinae GP357 JQ687440 JQ687430 JQ687456 JQ687415 Sinonatrix annularis Colubridae Natricinae AF236677 AF544712 AF036022 Sinonatrix annularis 1 Colubridae Natricinae GP889 JQ687449 JQ687431 JQ687424 Sinonatrix percarinata 1 Colubridae Natricinae GP324 JQ687439 GQ281784 JQ687455 JQ687414 Sinonatrix percarinata 2 Colubridae Natricinae GP956 JQ687451 JQ687433 JQ687465 JQ687426 Storeria dekayi Colubridae Natricinae AF402639 AF471154 EF417389 AF471050 EF417436 EF417460 EF417365 Storeria occipitomaculata Colubridae Natricinae AF402638 AF402921 AF384840 U49323 Thamnophis atratus Colubridae Natricinae AF420085 AF420086 AF420087 AF420088 Thamnophis brachystoma Colubridae Natricinae AF420089 AF420090 HM630351 AF420092 Thamnophis butleri Colubridae Natricinae AF402640 AF402923 AF420093 HM630350 AF420095 Thamnophis Colubridae Natricinae EF460849 AF420108 AF420096 AF420097 AF420098 chrysocephalus Thamnophis couchii Colubridae Natricinae AF402653 AF402936 AF420104 AF384855 AF420106 Thamnophis cyrtopsis Colubridae Natricinae AF402641 EF417388 EF417412 EF417435 EF417459 EF417364 Thamnophis elegans Colubridae Natricinae AF402642 AF402925 AF420114 AF384844 AF420116 Thamnophis eques Colubridae Natricinae AF420117 AF420118 AF420119 AF420120 Thamnophis exsul Colubridae Natricinae AF420125 AF420126 AF420127 AF420128 Thamnophis fulvus Colubridae Natricinae AF420129 AF420130 AF420131 AF420132 Thamnophis gigas Colubridae Natricinae AF420133 AF420210 AF420209 AF414099 Thamnophis godmani Colubridae Natricinae AF471165 AF420135 AF420136 AF420137 AF420138 Thamnophis hammondii Colubridae Natricinae AF420139 AF420140 AF420141 AF420142 Thamnophis marcianus Colubridae Natricinae AF402643 AF402926 AF420144 AF384845 AF420146 Thamnophis melanogaster Colubridae Natricinae EF417386 EF417410 EF417433 EF417457 EF417362 Thamnophis mendax Colubridae Natricinae AF420151 AF420152 Thamnophis ordinoides Colubridae Natricinae AF402644 AF402927 AF420158 AF384846 AF420160 Thamnophis proximus Colubridae Natricinae AF402645 AF402928 AF420162 AF384847 AF420164 Thamnophis radix Colubridae Natricinae AF402651 AF402934 AF420170 HM630344 AF420172 Thamnophis ruﬁpunctatus Colubridae Natricinae AF420173 AF420174 AF420175 AF420176 Thamnophis sauritus Colubridae Natricinae AF420177 AF420178 AF420179 AF420180 Thamnophis scalaris Colubridae Natricinae AF420181 AF420182 AF420183 AF420184 Thamnophis scaliger Colubridae Natricinae AF420189 AF420190 AF420191 AF420192 Thamnophis sirtalis Colubridae Natricinae AF402647 DQ902094 AF402929 AF420194 DQ995396 AY136272 P. Guo et al. / Molecular Phylogenetics and Evolution 63 (2012) 825–833 829

Table 1 (continued)

Species Family Subfamily Voucher 12S cmos COI cytb ND1 ND2 ND4 Thamnophis sumichrasti Colubridae Natricinae AF420197 AF420198 AF420199 AF420200 Thamnophis valida Colubridae Natricinae EF417385 EF417408 EF417432 EF417437 EF417360 Trachischium monticola 1 Colubridae Natricinae GP1487 JQ687435 JQ687428 JQ687453 Tropidoclonion lineatum Colubridae Natricinae AF420205 AF420206 AF420207 AF420208 Virginia striatula Colubridae Natricinae AF402650 AF402933 AF384852 Xenochrophis Colubridae Natricinae AF544780 AF544714 FJ416748 ﬂavipunctatus Xenochrophis piscator Colubridae Natricinae GQ225679 GQ225669 GQ225666 GQ225659 Xenochrophis punctulatus Colubridae Natricinae AF471106 AF471079 AY486996 AY487035 AY487074 Xenochrophis Colubridae Natricinae GQ225678 GQ225668 GQ225663 GQ225660 schnurrenbergeri Xenochrophis vittatus Colubridae Natricinae EF395871 EF395920 EF395895 Enhydris plumbea Homalopsidae Homalopsidae DQ343650 EF395934 DQ343650 DQ343650 DQ343650 DQ343650 DQ343650

Table 2 clade, we find that several widespread and diverse genera are not Results of AIC model selection conducted in monophyletic, these include Xenochrophis and Amphiesma. The MrModeltest for partitions of the dataset. genus Trachischium Günther, 1858 was erected based on the type Partition AIC model species Trachischium fuscum. There are five species in the genus 12S GTR + I + G and found throughout montane regions of the south slope of C-mos GTR + G Himalaya Mountains including, Nepal, India, Bangladesh, Bhutan C-mos, position 1 GTR + G and China (The Reptile Database: http://www.reptile-database. C-mos, position 1 GTR org). Their secretive nature and montane distribution account for C-mos, position 1 HKY COI HKY + I + G the lack of knowledge regarding their basic biology and systematic COI, position 1 F81 + I position. This genus has been placed under Colubridae, incertae COI, position 2 GTR + G sedis (The Reptile Database: http://www.reptile-database.org). COI, position 3 SYM + I Both BI and ML analyses presented here indicate the position of Cyt. b GTR + I + G Cyt-b, position 1 GTR + I + G Trachischium (T. monticola, Medog, Co., S. Xizang, China) to be sister Cyt-b, position 2 GTR + I + G to a clade composed of Amphiesma sauteri and A. craspedogaster (GP Cyt-b, position 3 GTR + I + G 139), to the exclusion of another sample of A. craspedogaster (GP ND1 GTR + I + G 600), deeply nested within Natricinae. ND1, position 1 GTR + I + G The complex phylogenetic relationships among Amphiesma ND1, position 2 HKY + I + G ND1, position 3 GTR + G with regard to Trachischium presents several systematic and taxo- ND2 GTR + I + G nomic problem requiring resolution. To correct the paraphyletic ND2, position 1 GTR + I + G Amphiesma, as defined presently, this genus could be divided into ND2, position 2 HKY + I + G two or more genera. The specimen GP 600 (YBU 071128), which ND2, position 3 GTR + I + G ND4 GTR + I + G has been tentatively identified as A. craspedogaster (Guo et al., ND4, position 1 GTR + I + G 2008), is likely a new species or may represent a unique genus, ND4, position 2 GTR + I + G depending on the position of T. monticola. Since the type species ND4, position 3 GTR + G of Amphiesma is A. stolatus (Duméril et al., 1854), the easiest way to correct A. craspedogaster and A. sauteri, which share a mrca with Trachistium monticola, would be to change the entire group to the genus Trachistium. However, extensive sampling of species within lnL = 168.8; RP = 0.37) at 27 Ma (95% CI: 22–33 Ma) and the À À Trachistium and Amphiesma, particularly the inclusion of the type other to Africa (Node 4; lnL = 167.9; RP = 0.88) at 26 Ma (95% CI À species of Trachistium (T. rugosum), would first need to be con- 20–32 Ma; Table 3). Dispersal to the New World (Node 3, Fig. 2; ducted before we can make a taxonomic recommendation. Addi- lnL = 168, RP = 0.82) occurred during the late Oligocene or Early À tionally, the diverse genus Xenochrophis is paraphyletic with Miocene (23 Ma; 95%CI 19–28 Ma). The dispersal to the Palearctic respect to Rhabdophis and Atretium. eventually produced the three extant species of Natrix, which is The second major clade, composed of the remaining Asian, contrasted against the dispersal to the New World yielding the European, and all NW genera (Thamnophiini) is strongly supported 58 extant species of Thamnophiini, whereas the dispersal to Africa as monophyletic (Fig. 1). Similar to Pyron et al. (2011) and Alfaro yielded Natriciteres and Afronatrix on the African continent and and Arnold (2001), several NW thamnophiine genera are demon- Lycognathophis in the Seychelles. strated to be paraphyletic, including Thamnophis, Nerodia and Regi- na. If these results are supported by sampling of additional species, 4. Discussion loci, and coalescent-based species tree analyses, then major taxonomic rearrangements would be necessary to correct paraphyletic 4.1. Systematics of Natricinae genera in both clades.

Our results support previous ﬁndings of monophyly for the subfamily Natricinae (Lawson et al., 2005; Pyron et al., 2011). We dem- 4.2. Biogeographic implications onstrate that a basal split separates two diverse clades, the ﬁrst composed mostly of Asian genera and the three African genera: Biogeographic patterns that are similar with respect to time and Afronatrix, Amphiesma, Atretium, Lycognathophis, Macropisthodon, area suggest similar underlying causes (Donoghue, 2001; Donog- Natriciteres, Rhabdophis, Trachischium and Xenochrophis. Within this hue and Moore, 2003; Cox and Moore, 2010; Lomolino, 2010; 830 P. Guo et al. / Molecular Phylogenetics and Evolution 63 (2012) 825–833

Enhydris plumbea Imantodes cenchoa Coluber constrictor Thamnophis rufipunctatus 100/100 Adelophis foxi Thamnophis melanogaster Thamnophis valida Thamnophis exsul Thamnophis godmani Thamnophis mendax Thamnophis sumichrasti Thamnophis scalaris 100/100 Thamnophis scaliger Thamnophis cyrtopsis Thamnophis hammondii Thamnophis ordinoides Thamnophis gigas Thamnophis atratus Thamnophis couchii Thamnophis elegans 100/100 Thamnophis brachystoma Thamnophis butleri 100/100 Thamnophis radix Thamnophis eques Thamnophis marcianus Thamnophis chrysocephalus NW 97/100 Thamnophis fulvus Thamnophis sirtalis A1 Thamnophis proximus 100/100 Thamnophis sauritus Nerodia cyclopion Nerodia floridana 92/51 97/56 Regina grahami 100/95 Tropidoclonion lineatum Regina septemvittata 100/100 Nerodia erythrogaster Nerodia fasciata 93/78 Nerodia harteri Nerodia sipedon Nerodia rhombifer Nerodia taxispilota Clonophis kirtlandii 100/97 54/- Virginia striatula 60/- Seminatrix pygaea 94/62 Regina alleni 100/97100/100 Regina rigida Storeria dekayi 100/100 Storeria occipitomaculata 100/100 OW 100/100 Natrix maura Natrix natrix A2 100/100 Natrix tessellata 100/100 Opisthotropis guangxiensis Opisthotropis lateralis

79/- 100/100 Opisthotropis cheni OW 99/99 Opisthotropis latouchii Sinonatrix aequifasciata A3 100/100 100/100 Sinonatrix annularis Sinonatrix annularis 1 97/99 100/81 Sinonatrix percarinata 1 Sinonatrix percarinata 2 100/100 100/94 Afronatrix anoscopus Lycognathophis seychellensis Natricinae 92/77 45/73 Natriciteres olivacea 100/100 Macropisthodon rudis 1 Macropisthodon rudis 2 100/91 100/100 Amphiesma stolatum 0.1 Amphiesma stolatum 1 100/100 Xenochrophis schnurrenbergeri Xenochrophis piscator 100/97 1 86/- Atretium yunnanensis 100/80 Xenochrophis flavipunctatus 96/- Xenochrophis punctulatus 53/- Xenochrophis vittatus 100/100 100/100 Rhabdophis nuchalis 100/66 Rhabdophis nuchalis 1 -/- 100/100 100/100 Rhabdophis tigrinus Rhabdophis tigrinus 1 100/100 Rhabdophis subminiatus Rhabdophis subminiatus 1 Amphiesma craspedogaster 2 (GP 600) Trachischium monticola 100/99 1 (GP 139) 100/91 Amphiesma craspedogaster 100/100 Amphiesma sauteri

Fig. 1. Bayesian 50% majority-rule consensus phylogram inferred using the 19-partiton model for combined nuclear and mtDNA data in MrBayes. The values assigned to the internodes indicate posterior probability support (before slash) and ML bootstrap support (after slash). A node with support value < 50% was indicated by ‘‘-’’. Branch support indices are not given for some subclades to preserve clarity.

Pyron and Burbrink, 2010). One common pattern in biogeography, (ratsnakes, crotalines, and Plestiodon skinks) likely originated in particularly for squamates, is the co-distribution of taxa across the Asia and dispersed through Beringia to North America at roughly Holarctic (Sanmartin et al., 2001). At least three common groups the same time, at the intersection of Miocene and Oligocene P. Guo et al. / Molecular Phylogenetics and Evolution 63 (2012) 825–833 831

Fig. 2. Chronogram with x-axis showing dates in millions of years (Ma) as produced using relaxed phylognetic methods in BEAST (see text for details and a discussion regarding uncertainty in dates at key nodes). Colored branches refer to the most likely geographic area estimate from the single best likelihood ancestral area estimate using the program Lagrange (see legend on tree and Appendix 1 for all possible reconstructions at each node). Uncertainty or combined geographic areas are represented by multicolored lineages. The range of each terminal taxon is represented by colored boxes after the species name. (For interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of this article.)

(Burbrink and Lawson, 2007; Wüster et al., 2008; Brandley The natricine watersnakes originated in Asia in the late Eocene et al., 2011). Here, we demonstrate a fourth example of a diverse or Oligocene, and dispersed to Africa in the Oligocene, and then and widespread Holarctic group with a similar biogeographic eventually to Palearctic and North America at the end of the Oligo- history. cene or early Miocene (Fig. 2). At this time, the Thulean Land Bridge 832 P. Guo et al. / Molecular Phylogenetics and Evolution 63 (2012) 825–833

Table 3 Results showing the likelihood (lnL) and relative probability (in parentheses) for the three most likely ancestral area reconstructions and divergence dates with uncertainty at ﬁve key internal nodes indicated in Fig 2. Reconstructions for the upper and lower branch emerging from each indicated node are given to the left and right of the bar, respectively, with areas coded as: A = Neotropical, B = Nearctic, C = Western Palearctic, D = Ethiopian (African), E = Eastern Palearctic, F = Oriental.

Node Ancestral Area Divergence Dates (mean and 95% CI) 1 [F|BCF] – 169.3 (0.21); [F|BCE] – 169.5 (0.17) 33 Ma (95% CI 28–40) 2 [BC|F] – 168.8 (0.37); [BC|E] – 169.2 (0.25) 27 Ma (95% CI: 22–33) 3 [C|B] – 168 (0.82); [CE|B] – 170.3 (0.08) 23 Ma (95%CI 19–28) 4 [D|F] – 167.99 (0.88); [D|EF] – 170.1 (0.10) 26 Ma (95% CI 20–32) 5 [A|AB] – 168 (0.80); [A|B] – 169.6 (0.17) 6 Ma (95% 5–8) connecting the Western Palearctic to the Nearctic would have likely A second major question is why there were not return coloniza- been too cold and disconnected for ectotherms, whereas Beringia, tions from the New World to the OW? Certainly, bidirectional dis- connecting the Eastern Palearctic to the Nearctic, was composed persal across the Holarctic is possible for ectotherms, as seen in of warmer meosphytic forests (Sanmartin et al., 2001; Burbrink toads, ranid and hylid frogs, which crossed the between the OW and Lawson, 2007). Therefore, the natricine dispersal into the NW and NW several times (Smith et al., 2005; Wiens et al., 2006, provides yet another line of evidence of support for the Cenozoic 2009; Pramuk et al., 2008). One possibility for the squamates is Beringian Dispersal Hypothesis, and rejects the Thulean and de Geer that return colonization from the NW was simply not possible Land Bridges as potential routes of colonization. Although the dis- due to degrading habitat at the end of Miocene in Beringia, an tances between the Western Palearctic and Nearctic were massive expectation of the CBDH. Another possibility is that competition in the Miocene for dispersal across the Atlantic Ocean, such routes from an already diverse fauna in the OW excluded recolonization have been demonstrated for some lizard groups (Carranza et al., of Asia. A final explanation is that some or all of the groups did 2000; Carranza and Arnold, 2003; Vidal et al., 2008; Gamble et al., recolonize Asia, but did not radiate and subsequent extinction 2011). While we have no evidence for long-distance oceanic dis- eliminated all traces of back colonization. persal in natricines (or other colubrids), Thamnophis validus has dis- Here, we offer a fourth case of intercontinental Holarctic dispersed across the Sea of Cortez and Nerodia clarkii across the Florida persal in a squamate group during the late Oligocene/early Mio- Straits from Florida to Cuba (de Queiroz and Lawson, 2008). It thus cene. We suggest that this common pattern is likely due to the appears that the natricines, along with the ratsnakes, crotalines, and existence of diverse trans-Beringian squamate communities, sup- Plestiodon skinks share similar areas of origins, routes, and times of ported by the warm mixed-mesophytic forests that dominated dispersal, suggesting a more common mechanism is at play rather the region during the early Cenozoic (Sanmartin et al., 2001; Bur- random oceanic dispersal. While origination in tropical Asia simply brink and Lawson, 2007). Subsequent collapse of these ecosystems suggests that the most recent common ancestor of the group was resulted in the extinction of these communities (as evidenced by found in this area, the dispersal to North America in the late Oligo- the present lack of squamate taxa in the region), and an intermedi- cene may indicate several alternative processes. ate range disjunction separating the temperate Asian and North First, it might suggest that environmental conditions were able American faunas. We refer to this shared underlying biogeographic to support populations and communities of squamates only at that distribution engine as the Cenozoic Beringian Dispersal Hypothesis particular time and not before (or after). This explanation does not (CBDH), and suggest that it may be responsible for similar distribu- seem plausible, given that Beringia continuously connected the tional patterns in other organisms, such as plants and other ani- OW and NW and apparently was covered in a continuous belt of mals (Sanmartin et al., 2001). This research further highlights the warmer boreotropical forest throughout the first half of the Ter- importance of comparative biogeographic analyses for illuminating tiary before changing over to a coniferous/deciduous hardwood the underlying causes of shared spatial and temporal patterns forest (mesophytic forests) in the Oligocene (Wolfe, 1987; Sanmar- among taxa. tin et al., 2001). Either type of habitat would have likely supported this community of squamates at any time prior to their actual dates of dispersal. Without further information regarding habitats Acknowledgments in Beringia throughout the late Eocene and Oligocene, this hypothesis is difficult to assess. A second possibility that might account This project was funded in part by the National Natural Science for the similar timing of dispersal to the New World concerns the Foundation of China (NSFC 30670236, NSFC 30970334, NSFC time it takes for these organisms to diversify and actually disperse 31071892), the program for New Century Excellent Talents in Uni- from tropical Asia across Beringia. That is, given the timing of versity (NCET-08-0908), the Ministry of Education of the People’s origin and dispersal capabilities of each group, they all took a Republic of China, the Sichuan Youth Sciences & Technology Foun- similar amount of time to diversify in Asia and subsequently dation (08ZQ026-006), and the US NSF (DBI-0905765 to R.A.P.). migrate across Beringia. Unfortunately, testing this is difficult Analyses were facilitated by a grant of computer time from the City without knowing dispersal capabilities of these groups and a fossil University of New York High Performance Computing Center, record to document the ever extending range of these taxa through which is supported by US National Science Foundation Grants time. Finally, it is possible that the similar timing is coincidental, CNS-0855217 and CNS-0958379. Numerous individuals helped and that one or more of these squamate groups actually reached with the collection and provision of tissue samples, they are H. the NW earlier, went extinct, and simply recolonized again, giving Zhao, M. He, G.X. Zhu. the appearance a simultaneous dispersal of squamate communities. This would suggest that dispersal and extinction were continuous through Beringia and the New World and that the extant Appendix A. Supplementary material diverse monophyletic groups found there today are simply representatives of the last dispersal and radiation of taxa that failed to Supplementary data associated with this article can be found, in go extinct. Distinguishing among these hypotheses is difficult, the online version, at http://dx.doi.org/10.1016/j.ympev.2012. and requires an actual fossil record in Beringia. 02.021. P. Guo et al. / Molecular Phylogenetics and Evolution 63 (2012) 825–833 833

References Malnate, E.V., 1960. Systematic division and evolution of the colubrid snake genus Natrix, with comments on the subfamily Natricinae. Proc. Acad. Nat. Sci. Phila. 112, 41–71. Alfaro, M.E., Arnold, S.J., 2001. Molecular systematics and evolution of Regina and Mittelbach, G.G., Schemske, D.W., Cornell, H.V., Allen, A.P., Brown, J.M., Bush, M.B., the Thamnophiine snakes. Mol. Phylogenet. Evol. 21, 408–423. Harrison, S.P., Hurlbert, A.H., Knowlton, N., Lessios, H.A., McCain, C.M., McCune, Arèvalo, E., Davis, S.K., Sites, J.W., 1994. Mitochondrial DNA sequence divergence A.R., McDade, L.A., McPeek, M.A., Near, T.J., Price, T.D., Ricklefs, R.E., Roy, K., Sax, and phylogenetic relationships among eight chromosome races of the D.F., Schluter, D., Sobel, J.M., Turelli, M., 2007. Evolution and the latitudinal Sceloporus grammicus complex (Phrynosomatidae) in central Mexico. Syst. diversity gradient: speciation, extinction and biogeography. Ecol. Lett. 10, 315– Biol. 43, 387–418. 331. Barry, C., 2001. The biogeographic regions reconsidered. J. Biogeogr. 28, 511–523. Nilsen, T.H., 1978. Lower Tertiary laterite on the Iceland-Faeroe Ridge and the Blackburn, T.M., Gaston, K.J., 2004. Macroecology. Basic Appl. Ecol. 5, 385–387. Thulean land bridge. Nature 274, 786–788. Brandley, M.C., Wang, Y.Z., Guo, X.G., de Oca, A.N.M., Feria-Ortiz, M., Hikida, T., Ota, Nylander, J.A.A., Ronquist, F., Huelsenbeck, J.P., Nieves-Aldrey, J.L., 2004. Bayesian H., 2011. Accommodating heterogenous rates of evolution in molecular phylogenetic analysis of combined data. Syst. Biol. 53, 47–67. divergence dating methods: an example using intercontinental dispersal of Posada, D., 2008. JModelTest: phylogenetic model averaging. Mol. Biol. Evol. 25, plestiodon (Eumeces) Lizards. Syst. Biol. 60, 3–15. 1253–1256. Brown, J.M., Lemmon, A.R., 2007. The importance of data partitioning and the utility Pramuk, J.B., Robertson, T., Sites, J.W., Noonan, B.P., 2008. Around the world in 10 of Bayes factors in Bayesian phylogenetics. Syst. Biol. 56, 643–655. million years: biogeography of the nearly cosmopolitan true toads (Anura: Burbrink, F.T., Lawson, R., 2007. How and when did Old World ratsnakes disperse Bufonidae). Global Ecol. Biogeogr. 17, 72–83. into the New World? Mol. Phylogenet. Evol. 43, 173–189. Pyron, R.A., Burbrink, F.T., 2009a. Can the tropical conservatism hypothesis explain Burbrink, F.T., Lawson, R., Slowinski, J.B., 2000. Mitochondrial DNA phylogeography temperate species richness patterns? An inverse latitudinal biodiversity of the polytypic North American rat snake (Elaphe obsoleta): a critique of the gradient in the New World snake tribe Lampropeltini. Global Ecol. Biogeogr. subspecies concept. Evolution 54, 2107–2118. 18, 406–415. Carranza, S., Arnold, E.N., 2003. Investigating the origin of transoceanic Pyron, R.A., Burbrink, F.T., 2009b. Neogene diversification and taxonomic stability in distributions: mtDNA shows Mabuya lizards (Reptilia, Scincidae) crossed the the snake tribe Lampropeltini (Serpentes: Colubridae). Mol. Phylogenet. Evol. Atlantic twice. Syst. Biodivers. 1, 275–282. 52, 524–529. Carranza, S., Arnold, E.N., Mateo, J.A., López-Jurado, L.F., 2000. Long-distance Pyron, R.A., Burbrink, F.T., 2010. Hard and soft allopatry: physically and ecologically colonization and radiation in gekkonid lizards, Tarentola (Reptilia: Gekkonidae), mediated modes of geographic speciation. J. Biogeogr. 37, 2005–2015. revealed by mitochondrial DNA sequences. Proc. Roy. Soc. B 267, 637–649. Pyron, R.A., Burbrink, F.T., 2012. Extinction, ecological opportunity, and the origins Cook, J.A., Hoberg, E.P., Koehler, A., Henttonen, H., Wickström, L., Haukisalmi, V., of global snake diversity. Evolution 66 (1), 163–178. Galbreath, K., Chernyavski, F., Dokuchaev, N., Lahzuhtkin, A., MacDonald, S.O., Pyron, R.A., Burbrink, F.T., Colli, G.R., de Oca, A.N.M., Vitt, L.J., Kuczynski, C.A., Wiens, Hope, A., Waltari, E., Runck, A., Veitch, A., Popko, R., Jenkins, E., Kutz, S., Eckerlin, J.J., 2011. The phylogeny of advanced snakes (Colubroidea), with discovery of a R., 2005. Beringia: intercontinental exchange and diversification of high latitude new subfamily and comparison of support methods for likelihood trees. Mol. mammals and their parasites during the Pliocene and Quaternary. Mammal Phylogenet. Evol. 58, 329–342. Study 30, S33–S44. Rage, J.C., 1988. The oldest known colubrid snakes. State of the art. Acta Zool. Cox, C.B., Moore, P.D., 2010. Biogeography: An Ecological and Evolutionary Cracov. 31, 457–474. Approach. Wiley, Hoboken, NJ. Ree, R.H., Moore, B.R., Webb, C.O., Donoghue, M.J., 2005. A likelihood framework for de Queiroz, A., Lawson, R., Lemos-Espinal, J.A., 2002. Phylogenetic relationships of inferring the evolution of geographic range on phylogenetic trees. Evolution 59, north American garter snakes (Thamnophis) based on four mitochondrial genes: 2299–2311. how much DNA sequence is enough? Mol. Phylogenet. Evol. 22, 315–329. Ree, R.H., Smith, S.A., 2008. Maximum likelihood inference of geographic range de Queiroz, A., Lawson, R., 2008. A peninsula as an island: multiple forms of evolution by dispersal, local extinction, and cladogenesis. Syst. Biol. 57, 4–14. evidence for overwater colonization of Baja California by the gartersnake Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes 3: Bayesian phylogenetic inference Thamnophis validus. Biol. J. Linn. Soc. 95, 409–424. under mixed models. Bioinformatics 19, 1572–1574. Donoghue, M.J., 2001. The integrative nature of historical biogeography. Am. Zool. Rossman, D.A., Eberle, W.G., 1977. Partition of the genus Natrix, with preliminary 41, 1430. observations on evolutionary trends in Natricine snakes. Herpetologica 33, 34– Donoghue, M.J., Moore, B.R., 2003. Toward an integrative historical biogeography. 43. Integr. Comp. Biol. 43, 261–270. Rossman, D.A., Ford, N.B., Seigel, R.A., 1996. The Garter Snakes: Evolution and Drummond, A.J., Ho, S.Y.W., Phillips, M.J., Rambaut, A., 2006. Relaxed phylogenetics Ecology (Animal Natural History Series). University of Oklahoma Press, Norman, and dating with confidence. PLoS Biol. 4, 699–710. pp. 336. Drummond, A.J., Rambaut, A., 2007. BEAST: Bayesian evolutionary analysis by Sambrook, J., Russell, D.W., 2002. Molecular Cloning: A Laboratory Manual. Cold sampling trees. BMC Evol. Biol. 7, 214. Spring Harbor Laboratory Press, New York. Duméril, A.M.C.G., Bibron, G, Duméril, A.H.A., 1854. Erpétologie Générale ou Sanmartin, I., Enghoff, H., Ronquist, F., 2001. Patterns of animal dispersal, vicariance Histoire Naturelle Complète des Reptiles. Tome septième. Première Partie, and diversification in the Holarctic. Biol. J. Linn. Soc. 73, 345–390. Comprenant l’Histoire des Serpents Non Venimeux. Librairie Encyclopédique de Smith, S.A., Stephens, P.R., Wiens, J.J., 2005. Replicate patterns of species richness, Roret, Paris. vii, [4], xvi, 780 pp. historical biogeography, and phylogeny in Holarctic treefrogs. Evolution 59, Edgar, R.C., 2004. MUSCLE: a multiple sequence alignment method with reduced 2433–2450. time and space complexity. Bmc Bioinformatics 5, 1–19. Stamatakis, A., Hoover, P., Rougemont, J., 2008. A rapid bootstrap algorithm for the Enghoff, H., 1995. Historical biogeography of the holarctic: area relationships, RAxML web-servers. Syst. Biol. 75, 758–771. ancestral areas, and dispersal of non-marine animals. Cladistics 11, 223–263. Szyndlar, Z., 2009. Snake fauna (Reptilia: Serpentes) from the Early/Middle Miocene Gamble, T., Bauer, A.M., Colli, G.R., Greenbaum, E., Jackman, T.R., Vitt, L.J., Simons, of Sandelzhausen and Rothenstein 13 (Germany). Palaeontol. Z. 83, 55–66. A.M., 2011. Coming to America: multiple origins of New World geckos. J. Evol. Vidal, N., Azvolinsky, A., Cruaud, C., Hedges, S.B., 2008. Origin of tropical American Biol. 24, 231–244. burrowing reptiles by transatlantic rafting. Biol. Lett. 4, 115–118. Guo, P.S., Huang, J.R., Fu, S.Y., Liu., 2008. Two snakes new to Xizang Autonomous Vitt, L.J., Caldwell, J.P., 2009. Herpetology. Elsevier, Burlington, MA. Region. Sichuan J. Zool. 27 (4), 658–659. Wallace, A.R., 1876. The Geographical Distribuion of Animals. 2 vol. Macmillan, Gibbons, J.W., Dorcas, M.E., 2004. North American Watersnakes: A Natural History London. (Animal Natural History Series). University of Oklahoma Press, Norman. Webb, C.O., Losos, J.B., Agrawal, A.A., 2006. Integrating phylogenies into community Holman, J.A., 2000. Fossil Snakes of North America: Origin, Evolution, Distribution, ecology. Ecology 87, S1–S2. Paleoecology. Indiana University Press, Bloomington. Wiens, J.J., Donoghue, M.J., 2004. Historical biogeography, ecology and species Huelsenbeck, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogenetic richness. Trends Ecol. Evol. 19, 639–644. trees. Bioinformatics 17, 754–755. Wiens, J.J., Graham, C.H., Moen, D.S., Smith, S.A., Reeder, T.W., 2006. Evolutionary Kreft, H., Jetz, W., 2010. A framework for delineating biogeographical regions based and ecological causes of the latitudinal diversity gradient in hylid frogs: on species distributions. J. Biogeogr. 37, 2029–2053. Treefrog trees unearth the roots of high tropical diversity. Am. Nat. 168, 579– Kozak, K.H., Wiens, J.J., 2010. Niche conservatism drives elevational diversity 596. patterns in Appalachian salamanders. Am. Nat. 176, 40–54. Wiens, J.J., Sukumaran, J., Pyron, R.A., Brown, R.M., 2009. Evolutionary and Lawson, R., 1987. Molecular studies of Thamnophiine snakes: 1. The phylogeny of biogeographic origins of high tropical diversity in Old World frogs (Ranidae). the genus Nerodia. J. Herpetol. 21, 140–157. Evolution 63, 1217–1231. Lawson, R., Slowinski, J.B., Crother, B.I., Burbrink, F.T., 2005. Phylogeny of the Wolfe, J.A., 1987. Late Cretaceous-Cenozoic history of deciduousness and the Colubroidea (Serpentes): new evidence from mitochondrial and nuclear genes. terminal Cretaceous event. Paleobiology 13, 215–226. Mol. Phylogenet. Evol. 37, 581–601. Wüster, W., Peppin, L., Pook, C.E., Walker, D.E., 2008. A nesting of vipers: phylogeny Lomolino, M.V., 2010. Biogeography. Sinauer Associates, Sunderland, MA. and historical biogeography of the Viperidae (Squamata: Serpentes). Mol. Lomolino, M.V., Riddle, B.R., Brown, J.H., 2006. Biogeography. Sinauer Associates, Phylogenet. Evol. 49, 445–459. Sunderland, Mass.