toward a phylogeny of the kukri snakes, genus …labs.eeb.utoronto.ca/murphy/pdfs of...

Asian Herpetological Research 2010, 1(1): 1-21 DOI: 10.3724/SP.J.1245.2010.00001

Toward a Phylogeny of the Kukri Snakes, Genus Oligodon

Marc D. Green1, Nikolai L. Orlov2 and Robert W. Murphy1,3* 1 Department of Natural History, Royal Ontario Museum, 100 Queen’s Park, Toronto, ON M5S 2C6 Canada 2 Zoological Institute, Russian Academy of Sciences, Universitetskaya nab 1, St. Petersburg 199034, Russia 3 State Key Laboratory of Genetic Resources and Evolution, Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming 650223, China

Abstract The South and Southeast Asian snake genus Oligodon, known for its egg-eating feeding behavior, has been a taxonomically and systematically challenging group. This work provides the first phylogenetic hypothesis for the ge-nus. We use approximately 1900 base pairs of mitochondrial DNA sequence data to infer the relationships of these snakes, and we examine congruence between the phylogeny and hemipenial characters. A hypothesis for the position of Oligodon within the Colubridae is also proposed. We discuss the implications of the phylogeny for previous taxonomic groupings, and consider the usefulness of the trees in analysis of behavior and biogeography of this genus.

Keywords Oligodon, genealogy, Colubridae, hemipenis, mitochondrial DNA, Southeast Asia, South Asia

1. Introduction

The genus Oligodon Fitzinger 1826, the kukri snakes, contains 70 species distributed throughout much of South and Southeast Asia (Green et al., in press). The common name derives from a distinctively shaped Nepalese knife, the kukri. The hind teeth of these snakes are broad and strongly recurved, much like the shape of the kukri. These non-venomous snakes are usually nocturnal and often brightly colored. They feed primarily on the eggs of birds and reptiles. The morphology of their teeth is very effective for cutting open eggs. Although some work on this aspect of their life history exists, little can be inferred about the evolution of this unique feeding strategy wi-thout a phylogeny.

Several attempts have been made to study the phy-logenetic relationships and taxonomy of Oligodon (Campden-Main, 1969; David et al., 2008b; Leviton, 1963a; Pope, 1935; Smith, 1943; Tillack and Günther, 2009; Wall, 1923; Wallach and Bauer, 1996) and these efforts have met with mixed success. Smith (1943) made the most ambitious review, examining 34 species and

*Corresponding author: Prof. Robert W. Murphy, holds faculty posi-tions at the University of Toronto, the Chinese Academy of Sciences, Hainan Normal University and East China Normal University, with his research focusing on evolutionary biology and conservation genetics. E-mail: [email protected] Received: 30 June 2010 Accepted: 20 July 2010

commenting on their relationships. However, his work predated the formalization of phylogenetic methods (Hennig, 1966). Very little progress has been made since then and Oligodon has yet to receive a modern phyloge-netic treatment using either morphology or molecules. Leviton (1963a) studied six Philippine taxa and Zhao et al. (1998) added new data in their examination of Chi-nese species. Neither study constructed a phylogenetic tree. Most works on Oligodon have not been comparative and they have focused on small, politically defined geo-graphic regions. Thus, in the absence of a comparative framework, some authors have perpetuated errors and confusion.

Oligodon’s preference for eggs as a source of food is well documented (Broadley, 1979; Coleman et al., 1993; Golf, 1980; Hu and Zhao, 1987; Minton and Anderson, 1963; Ota and Lin, 1994; Wall, 1921). Their egg-eating mechanisms and behavior are known (Coleman et al., 1993; Kiran, 1981; Kiran, 1982; Minton and Anderson, 1963; Ota and Lin, 1994). The snakes enthusiastically ingest the contents of opened eggs presented to them, and also ingest small reptile and bird eggs whole. An interesting and unusual aspect of this natural history is their apparent adaptation to taking eggs too large to swallow. They use their uniquely shaped teeth and jaw articulation to saw a hole in an egg-shell large enough to insert their head, and feed on the contents. Stomach content studies reveal that Oligodon also eat small rodents, tadpoles,

2 Asian Herpetological Research Vol. 1

frogs and insects (Batchelor, 1958; Cox, 1991; David and Vogel, 1996; Hu, 2001; Hwang et al., 1965; Meggitt, 1931; Schulz, 1988; Shi and Zheng, 1985; Toriba, 1987; Wall, 1921; Wall, 1923). Juveniles appear to be primarily insectivorous. It is not known whether they hunt or sca-venge for food.

The feeding habits of Oligodon have aroused interest in Chinese agriculture. At least one influential Asian re-ference stated they are harmful to crop production (Hu and Zhao, 1987), presumably because they eat the eggs of other beneficial insectivorous reptiles.

Some other behaviors of Oligodon may have an un-derlying phylogenetic basis. Most species are docile (David and Vogel, 1996; Pope, 1935; Stuebing and Inger, 1999; Wall, 1921). However, O. fasciolatus has an aggressive disposition (Smith, 1943; Taylor, 1965). It also displays an interesting defensive behavior in which a tail-up posture is taken and the bright red hemipenes are extruded repeatedly, contrasting noticeably with the white ventral scales (Wüster and Cox, 1992).

Feeding, behavior and the apparent large number of species in Oligodon provoke interest in its phylogenetic position within the Colubridae. Phylogeny is useful for discussing the evolution of oophagy in snakes (de Queiroz and Rodríguez-Robles, 2006) and phylogenetic relationships within Oligodon could provide insights into what drove the radiation within the clade. Phyllorhynchus uses a similar egg-contents extracting strategy for feeding (Gardner and Mendelson III, 2003), and it and Oligodon might share a close phylogenetic relationship (Dowling, 1974; Dowling et al., 1996; Lawson et al., 2005). How-ever, the position of these snakes within the larger Colu-bridae is unresolved (Dowling, 1974; Dowling et al., 1996; Kraus and Brown, 1998; Lawson et al., 2005; Lopez and Maxson, 1996; Zhang et al., 1984). Although de Queiroz and Rodriguez-Robles discuss Oligodon and Phyllorhynchus, they do not include them in their tree-based character-optimization analysis.

The geographic distribution of Oligodon is complex. Several species are widespread and many have overlap-ping ranges. Those ranges are all threatened due to the rapid pace of development in South and Southeast Asia (Dang and Nhue, 1995; de Silva, 1998). Many ranges are now fragmented and it is unclear what impacts this has had on the genetic diversity of isolated populations.

Although hemipenial defensive behaviors have not been extensively surveyed in Oligodon, the hemipenis itself has been used to infer historical relationships. Cope

(1895) drew the hemipenis of O. ancorus for use in the species’ description, suggesting that it might indicate historical relationships. Wall (1923) first looked at the hemipenes of Oligodon comparatively. Pope (1935) com-piled a list of hemipenial characters in an attempt to dis-tinguish Holarchus from Oligodon. However, he was unable to do so because O. bitorquatus, the type species for the genus, was intermediate in condition. Smith (1943) compiled observations on the hemipenes of Oligodon. Leviton (1963a) used hemipenial data to reconstruct the evolutionary history of the Philippine taxa. The groupings used today within Oligodon have been primarily based on Smith’s work and they have been taken to reflect histori-cal descent. Based on hemipenial morphology, Oligodon was considered to be a member of the Colubrinae (Zaher, 1999), now the Colubridae (Zaher et al., 2009). Herein, the diversity of Oligodon’s hemipenial morphology (Fi-gures 1–4) provided a foundation for character coding.

Figure 1 Photograph of one everted, bifurcate (forked) hemipenis of Oligodon chinensis (ROM34581)

Figure 2 Photograph of the everted hemipenes of Oligodon for-mosanus (ROM35634), showing the bifurcate (forked) condition and large papillae (arrows).

No. 1 Marc D. Green et al. Toward a Phylogeny of the Kukri Snakes, Genus Oligodon 3

Figure 3 Drawing of everted hemipenes of Oligodon albocinctus, showing the non-bifurcate (non-forked) condition. Small papillae present in O. albocinctus are not visible on this view. Some apical flounces and calyces are shown. Redrawn from Wall (1923; everted drawing), and Smith (1943; dissected view).

Figure 4 Drawing of everted hemipenes of Oligodon sublineatus, showing non-bifurcate (non-forked) condition. Spines are typical for many of the Indian Oligodon. Redrawn from Wall (1923).

Mitochondrial DNA (mtDNA) sequences can be used to reconstruct the phylogenetic relationships of snakes (de Queiroz and Lawson, 1994; Kraus and Brown, 1998; Lawson et al., 2005; Lopez and Maxson, 1996; Murphy et al., 2002; Rodríguez-Robles and de Jesús-Escobar, 1999; Slowinski and Lawson, 2002; Vidal et al., 2000). Herein, we present a molecular phylogeny for Oligodon based on mtDNA data and discuss correlation of rela-tionships and hemipenial morphology. Because Oligodon is such a speciose group, and tissue samples are not available for most species, this study suffers somewhat from an under-sampling of taxa. Nevertheless, it forms the foundation for future comparative work for exami-ning the evolution of structures, behaviors and biogeog-raphy in the group.

2 Materials and Methods

2.1 Specimens examined In total, 48 specimens of Oligodon were sequenced (Table 1), including three uni-dentified specimens, two of which appeared very similar to Oligodon chinensis, but would have represented a sub-stantial and unusual range extension for that species. The sequenced individuals represented 17 of 70 named spe-cies and the collecting localities were mapped (Figure 5). The sequence data for O. modestus were a synthesis of a partial sequence from ROM17490 for 12S rRNA, tRNAval and the 16S rRNA data of Lopez and Maxson (1996).

Table 1 A list of specimens chosen for DNA sequencing, their voucher numbers and locality data. For some specimens in the ROM collec-tion, denoted *, field numbers are provided in lieu of a catalogue number. ROM=Royal Ontario Museum; CAS=California Academy of Sci-ence; TNHM=Texas Natural History Museum; UMMZ=University of Michigan Museum of Zoology; USNM=United States National Mu-seum.

Species Voucher reference Locality and GenBank accession numbers

Dinodon semicarinatus AB008539

Oligodon arnensis ROM 19155 Assam, India HM591499

O. barroni ROM 32464 Krong Pa, Gai Lai, Vietnam HM591523

O. chinensis ROM 25757* Chi Linh, Hia Duong, Vietnam

O. chinensis ROM 34540 Chi Linh, Hia Duong, Vietnam HM591527

O. chinensis ROM 35626 Quang Thanh, Cao Bang, Vietnam HM591526

O. chinensis ROM 26940* Quang Thanh, Cao Bang, Vietnam

O. chinensis ROM 31032 Tam Dao, Vinh Phu, Vietnam HM591524

O. chinensis ROM 30824 Pac Ban, Tuyen, Vietnam HM591525

O. sp. ROM 30971 24 km W of Con Cuong, Nghe An, Vietnam

O. sp. ROM 30970 24 km W of Con Cuong, Nghe An, Vietnam HM591528

O. cinereus ROM 25755* Chi Linh, Hia Duong, Vietnam

O. cinereus ROM 32462 Chi Linh, Hia Duong, Vietnam HM591501

O. cinereus ROM 29552 Tam Dao, Vinh Phu, Vietnam HM591502


(Continued Table 1)

Species Voucher reference Locality and GenBank accession numbers

O. cinereus ROM 37092 Cat Tien, Dong Nai, Vietnam HM591504

O. cinereus CAS 210159 Alaungdaw Kathapa Ntl. Pk., Sagaing Div., Myanmar HM591505

O. cinereus CAS 215597 Alaungdaw Kathapa Ntl. Pk., Sagging Div., Myanmar

O. cinereus CAS 205028 Rakhine Yoma Mtn. Rng., Rakhine State, Myanmar HM591507

O. cinereus CAS 213379 Hlaw Ga Park, Yangon Div., Myanmar HM591506

O. cinereus CAS 215261 Kalaw Tnshp., Shan State, Myanmar HM591508

O. cinereus ROM 30969 24km W of Con Cuong, Nghe An, Vietnam HM591503

O. cruentatus CAS 213271 Hlaw Ga Park, Yangon Div., Myanmar HM591517

O. cruentatus CAS 213408 Hlaw Ga Park, Yangon Div., Myanmar

O. cyclurus CAS 204963 Mwe Hauk, Ayeyarwady Div., Myanmar HM591535

O. cyclurus CAS 215636 Alaungdaw Kathapa Ntl. Pk., Sagging Div., Myanmar HM591536

O. formosanus ROM 25827* Chi Linh, Hia Duong, Vietnam

O. formosanus ROM 35806 Chi Linh, Hia Duong, Vietnam HM591532

O. formosanus ROM 35629 Quang Thanh, Cao Bang, Vietnam HM591533

O. formosanus ROM 35630 Quang Thanh, Cao Bang, Vietnam

O. formosanus ROM 30939 Ba Be, Cao Bang, Vietnam HM591531

O. formosanus ROM 30823 Pac Ban, Tuyen, Vietnam HM591529

O. formosanus ROM 30826 Tam Dao, Vinh Phu, Vietnam HM591530

O. maculatus TNHC 59846 Barangay Baracatan, Mindinao Is., Philippines HM591511

O. modestus ROM 17490 Valencia, Negros Is., Philippines HM591498

O. modestus 16S mRNA from Lopez and Maxson, 1995

O. ocellatus ROM 32261 Yok Don, Dak Lak, Vietnam HM591534

O. octolineatus UMMZ 201913 3 Km E of Tutong, Tutong Dist., Brunei HM591519

O. planiceps CAS 213822 Shwe Set Taw Wldlf. Sanc., Magwe Div., Myanmar HM591514

O. splendidus CAS 204855 Kyauk Se Tnshp., Mandalay Div., Myanmar HM591509

O. splendidus USNM 520626 2 Km WNW Chatthin, Chatthin Wldlf. Sanc., Burma HM591510

O. taeniatus USNM 520625 2 Km WNW Chatthin, Chatthin Wldlf. Sanc., Burma HM591520

O. taeniatus ROM 32260 Yok Don, Dak Lak, Vietnam HM591521

O. taeniatus ROM 37091 Cat Tien, Dong Nai, Vietnam HM591522

O. theobaldi CAS 210710 Naung U Tnshp., Mandalay Div, Myanmar HM591515

O. theobaldi CAS 213896 Shwe Set Taw Wldlf. Sanc., Magwe Div., Myanmar HM591516

O. torquatus CAS 210693 Pakokku Tnshp., Magwe Div., Myanmar HM591512

O. torquatus CAS 210695 Pakokku Tnshp., Magwe Div., Myanmar

O. torquatus CAS 215976 Min Gone Taung Wldlf Sanc, Mandalay Div., Myan-mar HM591513t

O. venustus ROM 19156 India HM591500

O. sp. ROM 27049* Quang Thanh, Cao Bang, Vietnam HM591518

Phyllorhynchus decurtatus ROM 14153 Ocotillo, Imperial County, California, United States HM591497

Pituophis lineaticollis AF512746 The sequenced specimens of Oligodon were primarily

from Vietnam and Myanmar. Specimens were selected to maximize species representation in the analysis. We also attempted to evaluate inter- and intra-population genetic variation within species when possible by including mul-tiple specimens.

Phyllorhynchus decurtatus, a putative close relative of Oligodon, was selected as the initial outgroup taxon and a specimen was sequenced de novo for the equivalent mtDNA regions used in the ingroup. Sequences from Gen-Bank (http://www.ncbi.nlm.nih.gov/GenBank) for more distantly related colubrids, including Dinodon semicarinatus


Figure 5 Composite satellite photograph of Southeast Asia, showing the relief and topographic features and the collection localities of specimens of Oligodon from Vietnam and Myanmar. The localities from Brunei, India and the Philippines are indicated by arrows in their approximate directions. CB=Ba Be & Quang Thanh, Cao Bang, Vietnam; PB=Pac Ban, Tuyen, Vietnam; CL=Chi Linh, Hia Duong, Vietnam; TD=Tam Dao, Vinh Phu, Vietnam; CC=24km W of Con Cuong, Nghe An, Vietnam; KP=Krong Pa, Gai Lai, Vietnam; YD=Yok Don, Dak Lak, Vietnam; CT=Cat Tien, Dong Nai, Vietnam; VP=Valencia, Negros Is., Philippines; BP=Barangay Baracatan, Mindinao Is., Philippines; TB=3 km E of Tutong, Tutong Dist., Brunei; IN=India; MH=Mwe Hauk, Ayeyarwady Div., Myanmar; HG=Hlaw Ga Park, Yangon Div., Myanmar; RY=Rakhine Yoma Mtn. Rng., Rakhine State, Myanmar; KS=Kyauk Se Tnshp., Mandalay Div., Myanmar; KT=Kalaw Tnshp., Shan State, Myanmar; SS=Shwe Set Taw Wldlf. Sanc., Magwe Div., Myanmar; NU=Naung U Tnshp., Mandalay Div, Myanmar; MGT=Min Gone Taung Wldlf Sanc, Mandalay Div,.Myanmar; PT=Pakokku Tnshp., Magwe Div., Myanmar; AK=Alaungdaw Kathapa Ntl. Pk., Sagging Div., Myanmar; CWS=2 km WNW Chatthin, Chatthin Wldlf. Sanc., Myanmar and Pituophis lineaticollis, were also used as additional outgroup taxa.

Tissue samples included either skeletal muscle or liver tissue preserved in 95% ethanol at the time of collection and subsequently stored either in ethanol or frozen at −80°C. Voucher specimens are preserved in the collec-tions of the California Academy of Sciences (CAS), the Royal Ontario Museum (ROM), Texas Memorial Mu-seum (TNHC), Museum of Zoology, University of Michigan (UMMZ) and National Museum of Natural History (USNM).

2.2 DNA extraction, amplification and sequencing Three contiguous mtDNA genes were selected for phy-logeny reconstruction, including 12S ribosomal RNA (rRNA), 16S rRNA and tRNA Valine (tRNAval). Total DNA was obtained from alcohol preserved and frozen tissue samples using a preparation of 500 µL 0.1 M STE, 50 µL 20% SDS and 12.5 µL 20 mg/mL Proteinase K (GIBCO) added to 0.2 g of well-macerated tissue. The preparation was incubated at 37°C for 24–48 h with se-veral intervals of mixing.

DNA was purified from the cell lysate using a phe-nol-chloroform-isoamyl alcohol (PCI), chloroform- iso-amyl alcohol (CI) extraction. A volume of 450 µL PCI was added to the digested DNA, mixed well and centri-fuged at 13000 x g for 5min. The supernatant was trans-ferred to a new tube and the PCI step was repeated 2–3 times. Subsequently, the DNA supernatant was cleared of remaining phenol with a 450 µL CI wash, cen-trifugation and final supernatant draw-off. Final volume of DNA suspension product was 30–120 µL. The pre-sence of unsheared mtDNA was tested by adding 10 µL of product to a 1% agarose/TAE buffered electrophoretic gel with ethidium bromide. Results were visualized on a ultra-violet light table and recorded with Polaroid film or a digital camera. The DNA extraction procedure was re-peated if the sample failed to amplify with the poly-merase chain reaction (PCR).

Specimens of O. arnensis, O. modestus and O. venus-tus were collected predating the practice of separately preserving tissues for molecular investigation, and were fixed with at least some formalin. Consequently, an ex-


traction protocol modified after Chatigny (2000) was used. Small pieces (>0.1 g) of loose or free tissue were harvested from the specimen/container. These were transferred to 1.5 mL tubes and frozen using a mixture of dry-ice and ethanol, and then ground to a fine powder using a pestle sized for 1.5 mL conical bottom centrifuge tubes. These powdered tissues were digested with the STE/SDS/Proteinase K buffer above, but with the fol-lowing changes: Proteinase K concentration was doubled; DTE (Dithioerythritol) was added as 2.5 µL of 20 mg/mL stock solution to reverse protein cross-linkages; incuba-tion temperature was increased to 55°C. The samples were incubated for 24 h, then an additional volume of Proteinase K and DTE was added and the samples were incubated for a further 48 h. The extracted DNA was subjected to the PCI/CI clean-up described above. This produced DNA from which short PCR amplicons could be made.

Primer 12S1L, located at approximately 410 base pairs (bp) from the beginning of the 12S ribosomal RNA (rRNA) region, and primer 16S2H, located approxi-mately 200 bp from the terminus of the 16SrRNA region, were used for amplification of purified mtDNA. These primers flanked a region of approximately 1800 conti-nuous bp and internal primers were used to sequence overlapping regions. However, in many cases, and espe-cially with the formalin-fixed samples, shorter amplicons were produced and sequenced leaving some internal se-quence corresponding to the priming region unknown. The complete list of primers used in this study and their relative positions are presented in Figure 6 and Table 2.

Figure 6 A representation of a linear double stranded section of the mitochondrial genome, including 12S rRNA, tRNAval and 16Sr RNA. Arrows represent relative location of primers used in this study.

Double stranded mtDNA was amplified in 25 µL PCR

reactions (Saiki, 1989). The reactions were set up using 1 L of DNA (0.05 μg/µL), 1 µL of each primer (10 mM), 0.08 µL of a dNTP mix (containing 10 mM of each of dATP, dCTP, dGTP and dTTP) (Perkin-Elmer Cetus Kit), 0.2 µL of Amplitaq polymerase (Perkin-Elmer Cetus),

Table 2 Primer sequences used for PCR amplification and se-quencing of mtDNA in this study. See also Figure 6.

Name Sequence 5' to 3'

12S1L CCA ACT GGG ATT AGA TAC CCC ACT AT

12S2LM ACA CAC CGC CCG TCA CCC T

16S3H GTA GCT CAC TTG ATT TCG GG

16S4H CCA GCT ATC ACC AAG TTC GGT AGG CTT TTC

16S1LM CCG ACT GTT GAC CAA AAA CAT

16S1H GGC TAT GTT TTT GGT AAA CAG

16S5H CTA CCT TTG CAC GGT TAG GAT ACC GCG GC

16S2H CCG GAT CCC CGG CCG GTC TGA ACT CAG ATC ACG

2.5 L of reaction buffer (containing 67 mM Tris, pH 8.8, 2 mM MgCl) (Sigma) and 19 µL distilled deionized water. Thermocycling of the reaction mix was performed on MJ Research PTL-200 and Perkin-Elmer GeneAmp 9700 machines. Cycling parameters were set as follows: 38 or 40 cycles; denature at 92°C for 30 s; anneal at 45°C for 45 s; extension at 72°C for 45–90 s, and a final extension cycle at 72°C for 5 min. The annealing temperature was varied upward or downward to increase or decrease primer-DNA specificity when using different primer combinations with different samples. The extension time was increased when amplifying the longer fragments when using the 12s–1L primer to either 16s–4H or 16s–2H. Some samples could not be amplified for the entire ~1800 bp fragment; these were amplified in three overlapping segments of 12s–1L to 16s–4H, 12s–2LM to 16s–5H and 16s–1L to 16s–2H. Other less desirable am-plicons (12S–2LM to 16S–1H) were also used when the sample DNA proved refractory to amplification owing to variation in the sequence at the priming region. The com-bination of 12S–2LM and 16S–3H proved particularly effective for the formalin fixed samples.

Double stranded DNA PCR product was subjected to electrophoresis in a 100 mL 1% agarose gel with 10 µL ethidium bromide for 20–30 min at 120 V and 99 mA, and visualized under UV light. The amplified DNA frag-ments were cleaned of agarose and purified using either GeneClean (Bio101/Can Scientific), a standard QiaQuick (Qiagen) protocol, or with a protocol involving centri-fuging the band through an aerosol resistant p20/200 pi-pette tip at 13000 g for 5 min. The cleaned DNA was suspended in distilled deionized water.

DNA sequencing was performed using either 33P in-corporation and autoradiograph film or with dye-labeled nucleotides and fluorescent detection during electropho-resis. For radio-sequencing, a Thermo Sequenase 33P-


labeled cycle sequencing kit (Amersham) was used on Perkin-Elmer Cetus and Perkin-Elmer GeneAmp 9700 thermocyclers using the manufacturers’ protocols. The sequencing reaction mix consisted of 8 µL dH2O, 1 µL of light or heavy strand primer, 1.5 µL of buffer, 1.5 µL of Sequenase, 3 µL DNA and 6 µL termination mix, and was divided into four tubes (A, C, G, T). Subsequently, 0.37 µL of the appropriate 33P labeled ddNTP was added to each tube and overlaid with a drop of paraffin oil. Upon completion of cycling, 3 µL of stop solution (Amersham) was added to each tube and samples were stored at –20°C until electrophoresis (usually within 24 hr).

The products of the isotope-sequencing reactions were resolved in a 40 cm long 7M urea-5% polyacryla-mide-0.6% TBE buffer gel electrophoresed at 65 V for 2 hr or 6 hr (short run and long run). The gels were blot-ted onto filter paper, dried under heat and vacuum, and exposed for 2–5 d on Kodak XAR autoradiograph film. Gels were developed for 5 min and fixed for 2 min with Kodak developer and fixer.

Alternatively, some sequencing reactions were per-formed with the BigDye Terminator v 3.1 Cycle Se-quencing Kit (Applied Biosystems) using: 1 µL BigDye, 2 µL 5x BigDye Terminator Buffer, 2 µL ddH20, 1 µL of 10 mM primer solution, and 4 µL PCR product. Follow-ing the sequencing reactions, samples were cleaned and precipitated with sodium acetate and ethanol. These products were resolved on an ABI377 (Applied Biosys-tems) automated sequencer using the manufacturer pro-tocols.

2.3 DNA sequence alignment and phylogenetic analy-sis Sequences were either read from X-ray films ma-nually and entered into BioEdit 3.0 (Hall, 1997), or ex-tracted from ABI pherogram files, and initially aligned using ClustalW 1.7 (Thompson et al., 1994) with a gap-opening penalty of 15 and extension penalty of 6.66. The aligned sequences were imported into MacClade 4.0 (Maddison and Maddison, 1992) and final alignment was achieved by eye. Nucleotide ratios and proportions in-cluding transition/transversion ratios were calculated in MacClade. PAUP* 4.0 (Swofford, 2001) was used to calculate pairwise divergences among samples using un-corrected p-distances.

The aligned sequences were analyzed with PAUP*4.0b10 (Swofford, 2001) using maximum parsi-mony (MP) as the criterion for tree selection. Phyloge-netically uninformative sites were excluded, and heuristic searches were performed using random addition sequence with 10000 replicates while retaining minimal trees only,

tree bisection-reconnection branch swapping and collapse zero length branches. All multistate characters were evaluated as nonadditive (unordered) and unweighted. Gaps were treated as “missing data”.

Bayesian inference (BI) was also used to infer phy-logenetic relationships. MrModeltest v.2.3 (Nylander, 2004) was used to obtain evolutionary models using the Akaike Information Criterion (Akaike, 1974, 1979). BI was performed using MrBayes v.3.1.2 (Huelsenbeck and Ronquist, 2001). Two simultaneous runs of four chains were run for 3 X 106 generations, sampling every 100 generations. Stationarity was assessed when likelihood scores reached a stable equilibrium. A burn-in value of 7500 was implemented to discard topologies with low likelihood scores prior to generating a 50% majority rule consensus tree. Nodal support within the Bayesian con-sensus tree was assessed from the posterior distribution of topologies (Erixon et al., 2003).

Nodal support was evaluated using non-parametric bootstrap proportions (BSP; Felsenstein, 1985) and Bremer Decay Indices (Bremer, 1988). BSP was based on 10000 replicates calculated in PAUP* and nodes recei-ving >50% support were noted on the most parsimonious trees (MPTs). DIs were calculated by searching for trees in PAUP* using run instructions produced by AutoDecay v4 (Eriksson, 1998) and decay values were noted on the MPT on nodes also having >50% BSP.

Hemipenial characters were included in a total evi-dence approach because of their potential to be phyloge-netically informative. The data set was reduced to 17 in-group taxa by creating consensus DNA sequences for the nominal species. Variable base pair characters within species were coded as multi-state characters. Three un-ordered, unpolarized hemipenial characters were appended to the molecular data set. Hemipenial data for Phyllorhynchus were determined from literature (Klauber, 1935; Savage and Cliff, 1954) to be as follows: hemi-penis with spines, no apical papillae, and partially forked (or “lobed”, a condition in some Oligodon, but not any species for which sequence data were available). Zaher’s (1999) assignment of both these species to the Colubrinae (Colubridae) was based on them bearing distal calyces and the single sulcus in those Oligodon with undivided, unlobed (non-quadrapenes) organs.

Hemipenial characters were mapped on to phyloge-netic tress using MacClade. Specimens in Table 1 and literature records were used to determine character states. Hemipenial structure was described as three characters as follows: hemipenis with a forked division (quadrapenis),


unforked or partially forked; apical papillae present or absent; spines present or absent.

Available sequences from GenBank were used to test the relationship of Oligodon with Phyllorhynchus and to investigate where Oligodon might belong within the Colubridae. Table 3 was constructed to summarize 255 sequence fragments from eight genes plus known al-

lozyme, protein encoding loci shared between Oligodon and the other colubrids. These sequences were assembled into a file creating 48 synthetic genera. Concatenation produced 6971 aligned nucleotide positions. An un-weighted MP analysis in PAUP* with 10000 random ad-dition heuristic replicates and TBR branch swapping was applied to the data set.

Table 3 Colubroidea sequences from GenBank used to hypothesize the phylogenetic relationship of Oligodon to other colubrids. “x” de-notes a gene sequence that was not available for use in the sample, or synthetic representative. “+” denotes known protein loci data from Dowling et al. (1996), their coding used. “1” indicates a sequence obtained from Lopez and Maxson, 1995. 12S/16S sequences for Oligodon cinereus denoted “this study” are from specimen CAS205028. 12S/16S sequences for Pseudoxenodon bambusicola “this study” are ROM30825. 12S/16S sequences for Lycodon fasciatus “this study” are ROM25654. Higher taxonomic names follow Zaher et al. (2009).

Genus Species 12S 16S c-MOS CytB NADH4 NADH1 NADH2 COI Proteinloci

Colubridae incertae sedis

Grayia ornata AF158434 AF158503 AF544684 AF544663 x x x

smythii DQ112080 DQ112077

Calamariidae

Calamaria pavimentata x x AF471103 AF471081 x x x x +

Colubridae

Ahaetulla fronticincta x x AF471161 AF471072 x x x x +

Boiga cynodon AF139568 Z46525 x x x +

dendrophila AF471128 AF471089 U49303 x x x

Cemophora coccinea x 1(RH54131) AF471132 AF471091 AF138754 x x x +

Coluber constrictor U96794 L01770 AY486937 AY486913 AY487040 AY486963 AY487001 AY122649 +

Coluber dorri x AY188081 AY188001 AY188040 AY487042 AY486964 AY487003 x

Coluber zebrinus x AY188084 AY188004 AY188043 AY487058 AY486980 AY487019 x

Coronella girondica AY122835 AY643353 AF471113 AF471088 AY487066 AY486988 AY487027 AY122751

Crotaphopeltis tornieri x x AF471112 AF471093 x x AF428016 x

Dasypeltis atra x AF471136 AF471065 x x x x

scabra x 1(RH64432) x x x x +

Dinodon rufozonatum AF471163 AF471063

semicarinatus AB008539 AB008539 AB008539 AB008539 AB008539 AB008539 AB008539 +

Dipsadoboa unicolor x x AF471139 AF471062 x x AF428017 x

Dispholidus typus x AY188051 AY187973 AY188012 U49302 x x x

Elaphe quatuor-lineata AY122798 AF215267 AY486955 AY486931 AY487067 AY486989 AY487028 AY122714 +

Eirenis modestus AY039143 AY376780 AY486957 AY486933 AY487072 AY486994 AY487033 AY039181

Gastropyxis smaragdina AF158435 AF158504 DQ112078 DQ112075 x x x x

Gonyosoma oxycephalum AY122679 Z46490 AF471105 AF471084 x x x AY122661 +

Hemerophis socotrae AY039132 AY188083 AY188003 AY188042 AY487055 AY486977 AY487016 AY039170

Hemorrhois nummifer AY039163 AY376771 AY376800 AY376742 AY487049 AY486971 AY4870010 AY039201

Hierophis jugularis AY039152 AY376769 AY486941 AY486917 AY487046 AY486968 AY487007 AY039190

Lycodon capucinus U49317 x x x

fasciatus This study This study x x x

laoensis Z46455 Z46485 x x x +

zawi AF471111 AF471040 x x x


(Continued Table 3)

Genus Species 12S 16S c-MOS CytB NADH4 NADH1 NADH2 COI Proteinloci

Lytorhynchus diadema AY647229 AY188064 AY187986 AY188025 x x x x

Macroprotodon cucullatus AY643296 AY643337 AF471145 AF471087 AY487064 AY486986 AY487025 x

Masticophis flagellum AY122668 1(SM) AY486952 AY486928 AY487061 AY486983 AY487022 AY122650 +

Oligodon cinereus This study This study AF471101 AF471033 x x x +

octolineatus U49316 x x x

Oocatochus rufodorsatus AY122800 x AF435015 AF036016 x x x AY122716

Opheodrys aestivus x 1(RH61940) AF471147 AF471057 x x x x +

Oxybelis aeneus AF158416 AF158498 AF471148 AF471056 x x x x +

Pantherophis obsoletus AY122844 Z46493 AF471140 AF283644 AF138757 x x AY122760 +

Philothamnus heterodermus x 1(RH61045) AF471149 AF471055 x x x x +

Phyllorhynchus decurtatus AF544783 AF544812 AF471098 & AF544728 AF471083 x x x x +

Platyceps rogersi AY039127 AY188082 AY188002 AY188041 AY487052 AY486974 AY487013 AY039165

Pseustes poecilonotus x 1(RH62677) x x x x x x +

Ptyas korros AF236680 AY486953 AY486929 AY487062 AY486984 AY487023 AY122652 +

mucosus 1(RH56036)

Rhinechis scalaris AY122802 x AY486956 AY486932 AY487068 AY486990 AY487029 AY122718 +

Salvadora mexicana x AY486958 AY486934 AY487075 AY486997 AY487036

grahamiae AY122847 1(RH56651) AY122662 +

Sonora semiannulata x x AF471164 AF471048 x x x x

Spalerosophis diadema AY039144 AF471155 AF471049 AY487059 AY486981 AY487020 AY039182 +

cliffordi 1(RH58139)

Spilotes pullatus x x AF471110 AF471041 x x x x

Tantilla melanocephala AF158424 AF158491 x x x x +

relicta AF471107 AF471045 x x x x

Telescopus fallax x AY188078 AF471108 AF471043 x x x x +

Thelotornis capensis x x AF471109 AF471042 x x x x

Toxicodryas pulverulenta x x AF471118 AF471047 x x x x

Natricidae

Natrix natrix AY122682 AF158530 AF471121 & AF544697 AF471059 AY873736 AY873760 AY870640 AY122664

Thamnophis godmani AF471165 AF420135 AF420138 AF420136 AF420137 x

sirtalis AF402647 1(LM3071) x +

Pseudoxenodontidae

Pseudoxenodon bambusicola This study This study x x x x

karlschmidti AF471102 AF471080 x x x x

Dipsadidae

Alsophis cantherigerus AF544694 x x x

portoricensis AF158448 AF158517 AF471126 AF471085 U49309 x x x +

3. Results

3.1 Sequence variability in Oligodon The aligned data set contained 1885 bp of homologous DNA sequence sites from 48 ingroup samples and three outgroup taxa. In

nine cases, multiple individuals of the same species and locality did not vary and, in these cases, only one sample was selected for phylogenetic analysis. In the case of O. cinereus CAS205028 from Rakhine Yoma, Myanmar and O. cinereus CAS213379 from Hlaw Ga, Myanmar,


no variation was discovered. Further, O. theobaldi CAS213896 from Shwe Set Taw, Myanmar and O. theobaldi CAS210710 from Naung U, Myanmar showed no sequence variation. However, these sequences were maintained in the analysis to maximize geographic coverage.

DNA extracted from O. arnensis, O. modestus and O. venustus proved difficult to PCR amplify. However, high quality sequences were obtained for approximately 500 bp (using 12S–LM and 16S–3H). About 400 addi-tional bp were added to O. modestus from the literature.

The aligned data set includes a contiguous region of 507 bp from 12SrRNA, 64 bp of tRNA valine (tRNAval) and 1314 bp of 16SrRNA for a total of 1875 bp. The number of variable sites for each of the three gene seg-ments is given in Table 4. The average and ranges of the nucleotide frequencies across the ingroup taxa are as fol-lows: A, 0.389 (0.380–0.400); C, 0.237 (0.224–0.246); G, 0.159 (0.157–0.172) and T, 0.214 (0.194–0.221). Overall, 713 (38%) sites vary among ingroup and outgroup taxa, of which 513 are potentially phylogenetically informative. Within the ingroup, 618 characters are variable, 439 of which are potentially phylogenetically informative.

The overall transition:transversion ratio was 1.13:1. Uncorrected p-distances, the pair-wise absolute number of bp differences, were summarized in Table 4, with average and range of sequence divergence within the in-group and an intergeneric comparison between outgroups and the ingroup samples. Eight species in the analysis had representatives from multiple localities included as terminal nodes. The p-distances for the average in-ter-locality variation for these samples were as follows: O chinensis 0.027, O. cinereus 0.036, O. cyclurus 0.040, O. formosanus 0.014, O. splendidus 0.011, O. taeniatus 0.029, and O. torquatus 0.018.

3.2 Phylogeny The unweighted maximum parsimony analysis using all DNA data resulted in one MPT (2124 steps, CI = 0.488, RI = 0.710) with many nodes receiving

strong support (Figure 7). The heuristic search found the tree island with the MPT in 99.4% of all replicates. Oli-godon, as a genus, received very strong support. It is the sister group of Phyllorhynchus + Spilotes.

The GTR+I+G model of nucleotide evolution was chosen for these data. The well resolved BI tree (Figure 8) was very similar to the MPT, differing only in the place-ment of O. modestus, O. maculatus and O. octolineatus.

Almost all species received high support (BS ≥98%, DI ≥10+) with localities/populations showing mono-phyly. Exceptions to this were O. taeniatus and O. chinensis. Both were monophyletic, but O. chinensis only received high support values when the specimens of Oli-godon sp. from Con Cuong, Vietnam were included. Support for monophyly in O. taeniatus was high (BS = 75%, DI = 4), even though the Chatthin (Myanmar) and Cat Tien, Yok Don (Vietnam) specimens represented ex-treme east-west sampling of the range and the localities showed considerable genetic divergence with respect to each other.

An O. cyclurus-group (sensu Smith, 1943) is resolved within the MPT, albeit with moderate support only (BS = 62%, DI = 4). It consists of O. cyclurus, O. ocellatus, O. formosanus, and O. chinensis, and within this group a clade of three Vietnamese-Chinese species—O. ocellatus, O. formosanus and O. chinensis—receives higher support. The BI analysis differed in not uniting O. cyclurus with O. ocellatus + O. formosanus + O. chinensis.

An O. taeniatus-group (sensu Smith, 1943), consisting of O. taeniatus and O. barroni (BS = 99%, DI = 12), is resolved and this is sister to the O. cyclurus-group in the MPT. Philippine and Sunda Island’s O. octolineatus and O. modestus show a weakly supported relationship with the O. cyclurus-group + O. taeniatus-group. The O. cy-clurus-group + O. taeniatus-group is only recovered in the BI tree if O. octolineatus is included.

Indian O. arnensis, a widespread species, and O. venustus, a restricted-range species, show a well-supported association

Table 4 A list of total sites, variable sites and uncorrected p-distances for the three mtDNA genes in the data set for snakes of the genus Oligodon. Inter-generic measures range and mean of divergences between the ingroup and outgroup samples, intra-ingroup measures range and mean of p-distances among ingroup samples.

Range in pairwise divergence (average pairwise divergence) Gene Total bp Variable sites

(potentially parsimony informative), Ingroup Inter-generic Intra-ingroup

12SrRNA 507 161 (116) 0.115–0.266 (0.166) 0.000–0.149 (0.083)

tRNAval 64 16 (11) 0.000–0.198 (0.114) 0.000–0.186 (0.063)

16SrRNA 1314 441 (312) 0.110–0.188 (0.141) 0.000–0.149 (0.087)

Total 1885 618 (439) 0.121–0.202 (0.147) 0.000–0.125 (0.084)


Figure 7 Maximum parsimony phylogeny of Oligodon. Values denoted above internodes are bootstrap proportions, numbers below inter-nodes are decay index values. with the O. cyclurus-group, O. taeniatus-group, O. octo-lineatus and O. modestus. The other major clade reco-vered in the tree consists mostly of mainland Asian taxa including O. cinereus (a widespread and highly variable species), O. splendidus, O. maculatus (Philippines), and the following species whose relationships are shown in parenthetical notation: ((O. torquatus, O. planiceps) (O. theobaldi, O. cruentatus)). This latter clade contains three taxa with quite restricted distributions and O. theobaldi, which is less restricted in distribution. The ranges of these species might overlap in Myanmar.

3.3 Hemipenial characters Reduction to 17 taxa allowed the branch-and-bound algorithm to be used for tree building, which resolved a single MPT of 1420 steps,

CI = 0.598, RI = 0.577 (Figure 9). The topology of the tree with the hemipenial characters was very similar to the MPT produced by the DNA data alone (Figure 7). The two trees differed only at weakly supported nodes. In one case, O. splendidus moved from being the sister of O. cinereus to becoming the sister of O. maculatus (as re-solved in the BI tree). In the second case, O. chinensis and O. formosanus obtained a sister relationship to O. ocellatus. Tree instability was not due to the addition of new data, but rather to the deletion of taxa with uncertain identity and those having sequences from combined indi-viduals. The branching pattern in Figure 9 was obtained in the molecular-only data set when one taxon of uncer tain identity (ROM27049) and one locality of O. formo-


Figure 8 Bayesian inference phylogeny for Oligodon. Values denoted behind the nodes are posterior probabilities. sanus (Quang Thanh, Cao Bang, Viet Nam) were deleted. Regardless, the hemipenial characters had the same transformations on both trees.

Optimizing the three hemipenial characters on the phylogeny (Figure 9, A–C) reveals that the bifurcate hemipenis arose once only. Apical papillae and spines appear to have more instances of parallel, homoplastic state changes, but they are apomorphic and homologous for some clades. Accelerated (ACCTRAN) and delayed (DELTRAN) transformation optimizations display the same mapping for the hemipenial characters with one exception. Mapping apical papillae character states pro-duces two equivocal internodes within the ingroup. Under

ACCTRAN these become unequivocal for the presence of papillae, under DELTRAN they are non-papillate.

3.4 Position of Oligodon in the Colubridae Meta-analysis of colubrine sequences recovered a single MPT on a tree island found in 40% of the replicates (Figure 10, length 14314 steps). The tree had a clade consisting of Oligodon + (Phyllorhynchus + Spilotes), the same as found in the MP analysis of Lawson et al. (2005).

4. Discussion

4.1 Molecular variation Slowly evolving 12S and 16S ribosomal RNA genes are appropriate for resolving older


Figure 9 Hemipenial character state mapping for Oligodon. A: presence, absence, or partially divided/forked hemipenis. B: presence or absence of apical papillae on hemipenis (under ACCTRAN, equivocal internodes are papillate, under delayed transformation optimization they are non-papillate). C: presence or absence of spines on hemipenis. Several clades have been labeled to facilitate discussion. All charac-ters optimized unordered. divergences (~150 Ma) (Mindell and Honeycutt, 1990). Divergence in the rRNA sequences is moderate for the 48

samples of Oligodon. Among 507 aligned nucleotide sites for 12S, 32% are variable and 23% are potentially infor-


Figure 10 Maximum parsimony analysis of 48 colubrids including Oligodon and outgroups using published sequences of 12SrRNA, 16SrRNA, nuclear gene c-Mos, Cytochrome b, NADH4, NADH1, NADH2, COI and allozyme, protein-coding loci (Acp-2, Ldh-2, Mdh-1 and Pgm from Dowling et al., 1996). mative. Of 1314 sites in 16S, 34% are variable and 24% are potentially informative. The ratio of A: C: G: T for Oligodon is similar to that of other snakes (Kraus and Brown, 1998; Rodríguez-Robles and de Jesús-Escobar, 1999; Slowinski and Lawson, 2002) and typical of mtDNA. Analysis of p-distances for each gene reveals a similar range and average of variation as described in other studies, such as on Bufo (Liu et al., 2000) and snakes (Vidal et al., 2000). Taken together, this suggests that these characters are useful for reconstructing history in Oligodon at the taxonomic level of species and possi-bly the haplo-history between populations.

Evaluation of mtDNA genes as a single data partition is appropriate because mtDNA genes do not freely seg-regate and recombine but instead are inherited as a single linkage group. The data could be partitioned by transcrip-tional unit, but it is not necessarily true that these data partitions are appropriate or “natural” (Eernisse and Kluge, 1993). There are no a priori grounds to believe that knowledge of the transcriptional structure indicates that the whole sequence is under the same evolutionary forces. The most appropriate scale for data partitioning is the individual base pair.

The finding of no intra-locality variation within spe-


cies for these genes is interesting, but can be explained several ways. It may be that faster evolving genes are needed to assess genetic variation at this level, that more samples need to be sequenced, or that populations may in fact be quite isolated from each other. Nuclear gene se-quences will be required to assess gene flow.

The p-distances within species of Oligodon with mul-tiple populations were less than 50% of the average p-distance among all populations and species. They ranged from being as low as 13% of the variation seen among the whole ingroup. The p-distances were generally correlated with increasing geographic distance between localities (r2=0.664) indicating that isolation-by-distance was a factor to consider.

4.2 Phylogeny and previous hypotheses of relation-ships Many relationships recovered in this phylogeny agree with prior proposals. Although several species of Oligodon were absent from this study, the corroboration between genetic characters and hemipenial morphology allowed for sound speculation about the relationships of several other taxa. Philippine taxa Taylor (1922–25, 1965), Taylor and Elbel, (1958) and Leviton (1963a, 1963b) propose that O. maculatus is most closely related to Philippine O. an-corus, and then to O. purpurascens, which is not included in this study. Oligodon maculatus is more closely related to O. splendidus and O. cinereus. Oligodon purpurascens is a widespread species that extends onto the Malay Pe-ninsula. Oligodon maculatus, O. splendidus and O. cinereus all have quite similar hemipenes in their be-ing papillate, non-spinose and proximally calyculate. The relationship of O. purpurascens, which also has large hemipenial papillae, with either O. cinereus or O. splen-didus is new. It is possible that the large papillae of O. purpurascens are homologous with those in a clade that includes widespread and complex O. cinereus. However, another set of relationships is also possible for O. purpurascens.

Leviton (1963a) reports that O. modestus is related to O. ancorus and O. purpurascens. However, our phylo-geny refutes a close relationship between O. maculatus (a close relative of O. purpurascens) and O. modestus. While they share large hemipenial papillae, this condition is not a defining synapomorphy. Philippine diversity might be best explained by multiple invasions of the is-lands by different clades. A close relationship between O. purpurascens and O. modestus would allow O. maculatus to represent an invasion by an ancestor more closely allied with the O. cinereus-group.

Oligodon cyclurus - O. taeniatus-Group Smith (1943) suggests that O. cyclurus and O. chinensis are most closely related, that O. taeniatus and O. barroni are most closely related, and that these two groups are sisters. He places several other species in these groups as well. The larger grouping, which Smith refers to as the O. cyclurus - O. taeniatus complex, is recovered in the MPT but not in the BI tree. This study contains a fair sampling of the species that Smith assigns to these groups. Notwith-standing, there are several recent taxonomic changes in this group.

David et al. (2008b) describe three new species, O. deuvei, O. pseudotaeniatus, and O. moricei, and asso-ciate them with O. taeniatus. The discovery of these new species suggests that genetic information might be help-ful for identifying cryptic species of Oligodon. Our study includes a specimen of O. taeniatus from Myanmar (Chatthin Wildlife Sanctuary, USNM520625). This ani-mal is discussed by David et al. (2008b) who report that it was incorrectly identified by Zug (1998; catalogue number incorrect in publication). The voucher specimen is lost (David et al. 2008b). However, Zug’s description (Zug, pers. comm.) and the sequence data leave little doubt that it is O. taeniatus. Specimen USNM520624, which David et al. refer to as “most likely” being O. theobaldi is not included herein.

Several recent publications (David et al., 2004, 2008a,b; Pauwels et al., 2002, 2003; Stuart and Emmett, 2006; Stuart et al., 2006; Teynie et al., 2004) draw atten-tion to an opinion (Wagner, 1975) that O. fasciolatus is a species distinct from O. cyclurus. Although this study does not include O. fasciolatus, that species occurs in the center of what was once the nearly pan-Southeast Asian range of O. cyclurus. Future studies should include populations of this taxon and additional populations of O. cyclurus to evaluate the validity of this change. This study includes another taxon historically included in the variation of O. cyclurus, i.e., O. ocellatus, as well as O. cyclurus sensu stricto. O. ocellatus does not form the sister group of O. cyclurus.

This study’s sampling of the O. cyclurus - O. taeniatus group contains two specimens of uncertain identity (Oli-godon sp., from the vicinity of Con Cuong). Genetically these are quite similar to O. chinensis, yet the locality is far outside that species’ known range. These snakes might represent an undescribed species. A morphological examination of the specimens is underway. Oligodon cinereus This species has a considerable geographic range and is highly variable in coloration.


Samples of this species are from the western and eastern areas of its range, but not the center. Predictably, the p-distance between these samples is moderately high (0.036) relative to other species in this study. The sam-ples used here include some of the color varieties: ROM37092 from Cat Tien National Park, southern Viet-nam, represents the “palidocinctus” color form, while the specimens from Chi Linh, northern Vietnam, represent the moderately inornate form (Smith’s type II). Other specimens are closest to Smith’s form III. Given that O. taeniatus is a species complex (David et al., 2008b) and because the Myanmar and Vietnamese haplotypes are not sister lineages, it is likely that O. cinereus is also a com-plex of cryptic species. Additional sequenced specimens will be required to solve this problem, especially speci-mens from the central region of the range. Oligodon torquatus - O. planiceps - O. cruentatus - O. theobaldi Smith (1943) proposes a close relationship among O. torquatus, O. planiceps, O. cruentatus and O. theobaldi, but is unsure of where to place O. torquatus. His grouping is supported here and O. torquatus is re-solved as the sister taxon of O. planiceps. The autapo-morphic loss of hemipenial spines in O. torquatus serves to diagnose that species. This group differs from other spinose hemipenial Oligodon by the presence of apical papillae. Smith describes additional details of spine size and placement, which taken together indicates that the character coding of spinose/not-spinose might be overly simplistic.

One unidentified specimen (ROM27049, Oligodon sp.) is associated with this clade, albeit with low measures of support. Examination of the specimen is ongoing. It could represent a major range extension of a known spe-cies, or perhaps represents an undescribed species. Its hemipenes are unknown, but are predicted to be spinose and have papillae.

4.3 Hemipenes and phylogeny Fifty of 70 recognized species have described hemipenes. Of the remaining spe-cies, most are rare (from a highly restricted range, and/or not very abundant) or very rare (in some instances known only from the types). Pope (1935) reports hemipenial data for 12 Chinese species of Oligodon. Smith (1943) de-scribes the condition in 33 species, including nine sur-veyed by Pope, and divides Oligodon into seven groups based on hemipenial condition. Leviton (1963) adds data for four Philippine species and proposes recognizing three hemipenial groups of Oligodon. Leviton (1953, 1960) notes the hemipenial condition in O. annamensis. Zhao and Jiang (1981) describe and illustrate the organ of

O. multizonatus. David et al. (2008a,b) note the organ when describing new species.

A survey of three museum catalogues (FMNH: Alan Resetar, pers. comm.; CAS: Joseph B. Slowinski, pers. comm.; and ROM), indicates that a majority of Oligodon are not sexed. This situation precludes group assignment of many species of Oligodon. Unfortunately, this pivotal character is present in males only and its diagnostic fea-tures have been neglected in authoritative but less aca-demically oriented literature (Chan-Ard et al., 1999; Cox et al., 1998; Manthey and Grossmann, 1997). Hemipenial morphology is either neglected or not examined com-paratively in several important academic treatments (Campden-Main, 1969, 1970, 1970 (1984); in den Bosch, 1994; Lazell et al., 1999; Mahendra, 1984; Tillack and Günther, 2009; Yang and Inger, 1986; Zhao et al., 1998).

Using hemipenial microstructure, Smith (1943) di-vides the spinose hemipenial snakes into four groups. Herein, two spinose groups are recovered, consistent with Smith’s system, and there is no refutation for the other two groups, although Smith reports that spines are more plastic than either papillae or division of the organ into quadrapenes. Leviton (1963) proposes three groups as follows: deeply forked with papillae present or absent; unforked, papillae usually present, spines usually absent; and unforked, spines usually present, papillae usually absent. Only one of these groups is recovered herein. Leviton groups all species with deeply forked hemipenis together, but indicates that the individuals of both other groups can have either spines or papillae. Our analysis suggests that the two clades independently acquired spines. The hemipenes of most species in one clade are without spines but with papillae and this is comprised of O. theobaldi, O. cruentatus, O. planiceps, O. torquatus, O. cinereus, and other unsampled taxa. The other clade perhaps contains species with spines but without papillae on their hemipenes, including O. arnensis, O. venustus, and others. However, these two groupings fail to capture all species of Oligodon that also have unforked hemipe-nes.

Figure 9 shows that taxa tend to undergo subsequent modification of papillae and spines, especially secondary loss. However, neither character is freely plastic. Secon-dary loss or reduction of these ornamentations may be a key that reinforces speciation.

Hemipenial characters appear to be robust and diag-nostic (Leviton, 1960; Pope, 1935; Smith, 1943; Wall, 1923), providing relief from reliance on color and scala-tion for species identification and phylogenetic corrobo-


ration (Eernisse and Kluge, 1993; Kluge, 1997; Siddall and Kluge, 1997). The heritability of color and scalation characters remains a question in Oligodon in particular (Smith, 1943), and, in general, their use in phylogenetics often rests on invalid evolutionary assumptions (Murphy and Doyle, 1998).

4.4 Color, confusion and biogeography Much confu-sion in Oligodon occurs because of extensive intraspeci-fic variation in color and patterning. Most species of Oligodon have a fairly invariant color pattern, but a few species, namely O. cyclurus, O. fasciolatus, O. cinereus and O. taeniolatus, are highly variable. The venters are described for some species of Oligodon, but not others. In some species, individuals may be checkered or not. The dorsal color of Oligodon ranges from brown and black to green and orange with highlights in black, white, red and yellow. Overall color can change with age and contrast in pattern can be lost. The overall dorsal pattern can be very plain with loose, indistinct reticulation confined to scale margins, as in one form of both O. cyclurus and O. for-mosanus. It can also be more complex with varying numbers of longitudinal stripes, crossbars, or spots ran-ging in size and shape. Some species combine elements of all of the above. Taken together, O. cyclurus, O. taeniolatus and O. fasciolatus have five distinct color morphs, and O. cinereus has four. Geographic overlap and intergradations of the color morphs are poorly under-stood.

The highly variable species are also widespread. Oli-godon cyclurus and O. fasciolatus range from India eastward through Thailand (at approximately 11°15′N), to Vietnam and southwestern China (Yunnan). A separa-tion of about 500 km on the Thai peninsula separates O. fasciolatus from O. purpurascens, which continues throughout Indonesia. The distribution of O. cinereus closely matches cyclurus+fasciolatus. Ologodon taeniolatus ranges throughout India to eastern Iran and northward to southern Turkmenistan.

Many of Smith’s (1943) seven hemipenial groupings are corroborated here. None are refuted. If his assign-ments are mostly correct, then several clades will contain a morphologically variable widespread species with a range that will encompass that of other clade members. This morphologically variable species will also encom-pass much of the color and pattern variation found in other clade members. Thus, the sympatric species will have similar colors and patterns.

Pleistocene ecological changes (Hall and Holloway,

1998) may be responsible for the isolation of populations of widespread species that subsequently accrued mor-phological changes. This is consistent with the idea that many of the extant species of Oligodon appear to have a subset of the variation of one of the widespread species. Oligodon can be found on many Pleistocene landbridge islands. Certainly, glacial cycles can play a role in speci-ation throughout the Philippines and Indonesia (Karns et al., 2000).

4.5 Taxonomic implications Kukri snakes were origi-nally partitioned into two genera based on dentition. The genus Simotes Duméril, Bibron et Duméril, 1854 was established to include kukri snakes that had palatine and pterygoid teeth (Günther, 1864), whereas Oligodon Fit-zinger 1826 was reserved for those lacking this dentition (Günther, 1864; Boulenger, 1894). Holarchus (Cope, 1886; Pope, 1935; Stejneger, 1907) was used later to in-clude most of the same species originally assigned to Simotes (Duméril et al., 1854), but this group was rede-fined (by Pope) based on the absence of hemipenial spines and the highly correlated condition of having a divided anal scale. As late as 1970, Deuve continued to recognize Holarchus and Leviton (1963) speculated that there might be two clades of Oligodon, one that could carry the name Holarchus. Most of the kukris surveyed herein were originally assigned to Simotes based on den-tition. Oligodon modestus was an important exception in lacking palatine and pterygoid teeth. Based on hemipenial and anal scale conditions, two clades in this study could have been assigned to Oligodon: clade A and clade C, Figure 9. However, these two lineages were not mono-phyletic, and the clade that would carry the name Oli-godon was uncertain because the type species of Oli-godon, O. bitorquatus, was not included. Regardless, this arrangement precluded a monophyletic Simotes and/or Holarchus. The division of Oligodon drawing on any previous formulation would have required the erection of several additional genera.

Nomenclatorial stability should be a primary concern. A phylogenetically informed taxonomy contains the most information within a hierarchical system (Brooks and McLennan, 1991, 2002; Farris, 1967; Wiley, 1980). Until a far more complete sampling of species is possible, Oli-godon should not be subdivided into multiple genera. Oligodon: Position within the Colubridae/Colubrinae To determine whether Oligodon is an unusually speciose genus of snakes, it is necessary to determine its sister group relationships, and not just an appropriate outgroup (Brooks and McLennan, 1991). It is also important to


establish the sister group of Oligodon because outgroup choice can be critical in phylogenetic reconstruction (Milinkovitch and Lyons-Weiler, 1998; Tarrío et al., 2000; Ware et al., 2008).

Relationships within the Colubridae are discussed elsewhere (Dowling, 1974; Dowling et al., 1996; Kraus and Brown, 1998; Lawson et al., 2005; Lopez and Maxson, 1996; Zhang et al., 1984) and they will be the subject of future studies. Oligodon is included in several of these studies, yet no one uses all of the available data from molecular investigations to hypothesize the position of Oligodon relative to other colubrines. Kelly et al. (2003) use much of the available molecular data in a meta-analysis of snakes, but they include too few colu-brines (and no Oligodon) to address this question. Most studies agree that Oligodon falls within the Colubrinae. Two studies (Dowling et al., 1996; Lawson et al., 2005) suggest that Oligodon is related to New World Phyl-lorhynchus, but they disagree on the placement of these two genera within the Colubrinae. The close relationship between a relatively speciose Asian Oligodon and a rela-tively depauperate North American genus (Phyllorhyn-chus) remains an unintuitive result.

This analysis (Figure 10) did not refute a close rela-tionship between Oligodon and Phyllorhynchus (+Spilotes). However, this clade did not share a close relationship with other New World colubrines. It diverged early from the remaining Colubrinae. This finding was consistent with proposal of the Oligodontini by Dowling (1974), Dowling and Duellman (1978) and Dowling et al. (1996). The positioning of Oligodon herein was also con-sistent with that of Zhang and Zhao (1984).

5. Concluding Remarks

Oligodon is a genus worthy of future investigation. The rarity of specimens in collections and in the field across most, if not all of their distributions, will continue to be a key challenge to future work. There are relatively few specimens of Oligodon in museum collections, and even fewer have tissues separately preserved for molecular investigation. Political instability across much of the range of these species in the twentieth century often forces researchers to take an artificially geographically restricted approach to work. Notwithstanding these limi-tations, this study produces several interesting results.

The hemipenial morphology of Oligodon is phyloge-netically constrained and, therefore, informative. The hemipenes are useful in species identification, and it is

tantalizing to speculate that the ornamented organ may play a role in speciation or maintenance of species. Speculation on cladistic relationships by Smith (1943), based on hemipenial morphology, are validated herein, even though challenged by others. Thus, new species of Oligodon should be based on a type series that includes males with described hemipenes.

Splitting the genus Oligodon is not justified at this time. This would likely result in taxonomic instability.

The genus Oligodon may be old. Its age might be as-sociated with its species diversity. Burbrink and Lawson (2007) discuss the possible land bridges between Asia and North America that were traversable by snakes. If Phyllorhynchus arrived in North America in the early Oligocene along with rat snakes, then its sister group, Oligodon, may be at last 20 Ma old. During this long period of isolation, Oligodon could have experienced repeated habit fragmentations and reconnections. As mo-lecular clock methods improve, this possibility will be-come testable. The inclusion of Spilotes in a clade with Phyllorhynchus and Oligodon should be treated with great skepticism. The nucleotide dataset for Spilotes in the colubrid analysis, is the smallest of all taxa.

This phylogeny may be applied to future surveys of feeding and defensive behaviors. Egg-eating and defen-sive displays in this genus may not be uniform. If they are not ubiquitous, then the diversity can be mapped to this phylogeny thus inferring phylogenetically unique events. Better defined distributions will facilitate bio-geographic investigations of Oligodon and these might reveal historical connections between areas and identify barriers to dispersal in South and Southeast Asia. Range overlap between kukri snakes that are relatively distantly related might reveal historical biogeographic patterns in sub-clades. Phylogeny and biogeography of Oligodon could improve our understanding of speciation modes and they could be taken into consideration when planning for conservation in South and Southeast Asia.

Acknowledgements This research formed part of the requirements for the fulfillment of a M.Sc. Degree at the University of Toronto for MDG. Ross MacCulloch care-fully proofread the manuscript. The research was sup-ported by a Natural Sciences and Engineering Research Council (Canada) Discovery Grant A3148 to RWM. Fieldwork in Southeast Asia was supported by NSERC, the Royal Ontario Museum (ROM) Foundation and the ROM Members Volunteer Committee. All fieldwork was conducted using Animal Use Protocols approved by the ROM Animal Care Committee.


References

Akaike H. 1974. A look at the statistical model identification. IEEE Trans Automat Contr, 19: 716–723

Akaike H. 1979. A Bayesian extension of the minimum AIC pro-cedure of autoregressive model fitting. Biometrika, 66: 237–242

Batchelor D. M. 1958. Some notes on the snakes of Asahan, Ma-lacca. Malayan Nat J, 12: 103–111

Bremer K. 1988. The limits of amino acid sequence data in angio-sperm phylogenetic reconstruction. Evolution, 42: 795–803

Broadley D. G. 1979. Predation on reptile eggs by African snakes of the genus Prosymna. Herpetologica, 35: 338–341

Brooks D. R., McLennan D. A. 1991. Phylogeny, Ecology and Behavior. Chicago: University of Chicago Press, 441 pp

Brooks D. R., McLennan D. A. 2002. The Nature of Diversity: An Evolutionary Voyage of Discovery. Chicago: University of Chi-cago Press, 676 pp

Campden-Main S. M. 1969. The status of Oligodon taeniatus (Guenther), 1861, and Oligodon mouhoti (Boulenger), 1914 (Serpentes, Colubridae). Herpetologica, 25: 295–299

Campden-Main S. M. 1970. The identity of Oligodon cyclurus (Cantor, 1839) and revalidation of Oligodon brevicauda (Stein-dachner, 1867) (Serpentes: Colubridae). Proceedings of the Biological Society of Washington, 82: 763–766

Campden-Main S. M. 1970 (1984). A Field Guide to the Snakes of South Vietnam. NY: Herpetological Search Service & Exchange, Lindenhurst, 114 pp

Chan-Ard T., Grossmann W., Gumprecht A., Schulz K. D. 1999. Amphibians and Reptiles of Peninsular Malaysia and Thailand: An Illustrated Checklist. Amphibien und Reptilien der Halbinsel Malaysia und Thailands - Eine Illustrierte Checkliste. Wuerselen: Bushmaster Publications, 240 pp

Chatigny M. E. 2000. The extraction of DNA from formalin-fixed, ethanol-preserved reptile and amphibian tissues. Herpetol Rev, 31: 86–87

Coleman K., Rothfuss L. A., Ota H., Kardong K. V. 1993. Kinematics of egg-eating by the specialized Taiwan snake Oli-godon formosanus (Colubridae). J Herpetol, 27: 320–327

Cope E. D. 1886. An analytical table of the genera of snakes. Pro-ceedings of the American Philosophical Society 23: 479–499

Cope E. D. 1895. The classification of the Ophidia. Trans Am Phi-los Soc, 18: 186–219

Cox M. J. 1991. The Snakes of Thailand and Their Husbandry. Malabar, Florida: Robert E. Krieger, 526 pp

Cox M. J., Dijk P. P. V., Nabhitabhata J., Thirakhupt K. 1998. A Photographic Guide to Snakes and Other Reptiles of Peninsular Malaysia, Singapore and Thailand. Sanibel Island, FL: Ralph Curtis, 144 pp

Dang P. N., Nhue T. H. 1995. Environmental pollution in Vietnam, 129–200. In C. V. Sung (ed.), Environment and Bioresources of Vietnam, Present Situation and Solutions, Hanoi: The Gioi

David P., Cox M. J., Pauwels O. S. G., Chanhome L., Thirakhupt K. 2004. When a bookreview is not sufficient to say all: an in-depth analysis of a recent book on the snakes of Thailand, with an updated checklist of the snakes of the King-dom. Nat Hist J Chulalongkorn Univ, 4: 47–80

David P., Vogel G. 1996. The Snakes of Sumatra: An Annotated Checklist and Key With Natural History Notes. Frankfurt am

Main: Edition Chimaira, 260 pp David P., Vogel G, Pauwels O. S. G. 2008a. A new species of the

genus Oligodon Fitzinger, 1826 (Squamata: Colubridae) from southern Vietnam and Cambodia. Zootaxa, 1939: 19–37

David P, Vogel G. Van Rooijen J. 2008b. A revision of the Oli-godon taeniatus (Gunther, 1861) group (Squamata: Colubridae), with the description of three new species from the Indochinese region. Zootaxa, 1965: 1–49

de Queiroz A., Lawson R. 1994. Phylogenetic relationships of the garter snakes based on DNA sequence and allozyme variation. Biol J Linn Soc, 53: 209–229

de Queiroz A., Rodríguez-Robles J. A. 2006. Historical contin-gency and animal diets: The origins of egg eating in snakes. Am Nat, 167: 684–694

de Silva A. (ed) 1998. Biology and Conservation of the amphibians, reptiles and their habitats in South Asia. Peradeniya, Sri Lanka: ARROS, 364 pp

Deuve J. 1970. Serpents du Laos. Mémoire O.R.S.T.O.M. (39): 1–251

Dowling H. G. 1974. A provisional classification of snakes. HISS Yearbook Herpetol, 167–170

Dowling H. G., Duellman W. E. 1974–1978. Systematic Herpe-tology: A Synopsis of Families and Higher Categories. New York: HISS Publications

Dowling H. G., Hass C. A., Hedges B., Highton R. 1996. Snake relationships revealed by slowly-evolving proteins: A prelimi-nary survey. J Zool, 240: 1–28

Duméril A. M., Bibron C. G., Duméril A. H. A. 1854. Erpétologie Générale ou Histoire Naturelle Complète des Reptiles. Paris: Tome septième. Librarie Encyclopédique de Roret

Eernisse D. J., Kluge A. G. 1993. Taxonomic congruence versus total evidence, and amniote phylogeny inferred from fossils, molecules and morphology. Mol Biol Evol, 10: 1170–1195

Eriksson T. 1998. AutoDecay ver. 4.0 (program distributed by the author). Stockholm: Bergius Foundation, Royal Swedish Aca-demy of Sciences

Erixon P., Svennblad B., Britton T. Oxelman B. 2003. Reliability of Bayesian posterior probabilities and bootstrap frequencies in phylogenetics. Syst Biol, 52: 665–673

Farris J. S. 1967. The meaning of relationship and taxonomic pro-cedure. Syst Zool, 16: 44–51

Felsenstien J. 1985. Confidence limits on phylogenies: an ap-proach using the bootstrap. Evolution, 39: 783–791

Fitzinger L. J. 1826. Neue Classification der Reptilien nach ihren Natürlichen Verwandtschaften. Vienna: Heubner J. G., 66 pp

Gardner S. A., Mendelson III J. R. 2003. Diet of the leaf-nosed snakes, Phyllorhynchus (Squamata: Colubridae): squamate-egg specialists. Sw Nat, 48: 550–556

Golf M. L. 1980. A new species of Ophimegistus (Acari: Parmegis-tidae) from a Malaysian kukri snake. Pac Insects, 22: 380–384

Green M. D., Murphy R. W., Orlov N. L. In press. Checklist and key to the species of the kukri snakes, genus Oligodon. Russ J Herpetol

Günther A. C. L. G. 1864. The Reptiles of British India. London: Ray Society, 452 pp

Hall R., Holloway J. D. (eds) 1998. Biogeography and geographic evolution of SE Asia. Leiden: Backhuys Publications

Hall T., 1997. BioEdit Sequence Alignment Editor (program dis-


tributed by the author). Department of Microbiology North Carolina State University.

Hennig W. 1966. Phylogenetic Systematics. Urbana: University of Illinois Press, 263 pp

Hu C. H., 2001. Endangered species in a biodiverse web: Conser-vation of Green Sea Turtle impacts native egg-eating snake In Society for Conservation Biology (Conference), Hilo, Hawaii

Hu S., Zhao E. M. 1987. [Atlas of China Animals–Amphibians and Reptiles (2nd Edition)]. Beijing: Science Publishing House, 110 pp (In Chinese)

Huelsenbeck J. P., Ronquist F. 2001. MRBAYES: Bayesian in-ference of phylogeny. Bioinformatics 17: 754–755

Hwang M. H., Ho S. S., Chou S. A., Hsieh T. T., Hu B. C. 1965. [Feeding habits of snakes from Chekiang]. Acta Zool Sinica, 17: 137–146 (In Chinese)

in den Bosch H. A. J. 1994. On the juvenile forms of Oligodon waandersi (Bleeker, 1860). Mitteilungen aus dem Zoologischen Museum in Berlin, 70: 301–309

Karns D. R., O'Bannon A., Voris H. K., Weigt L. A. 2000. Bio-geographical implications of mitochondrial DNA variation in the bockadam snake (Cerberus rynchops, Serpentes: Homolapsinae) in Southeast Asia. J Biogeography, 27: 391–402

Kelly C. M. R., Barker N. P., Villet M. H. 2003. Phylogenetics of advanced snakes (Caenophidia) based on four mitochondrial genes. Syst Biol, 52: 439–459

Kiran U. 1981. A new structure in the lower jaw of colubrid snakes. Snake, 13: 131–133

Kiran U. 1982. Functional morphology of the mid-mandibular articulation in Oligodon arnensis (Shaw) (Serpentes: Colubri-dae). Snake 14: 83–90

Klauber L. M. 1935. Phyllorhynchus, the leaf-nosed snake. Bulle-tin of the Zool Soc San Diego, 12: 1–31

Kluge A. G. 1997. Testibility and the refutation and corroboration of cladistic hypotheses. Cladistics, 13: 81–96

Kraus F., Brown W. M. 1998. Phylogenetic relationships of colubroid snakes based on mitochondrial DNA sequences. Zool J Linn Soc, 122: 455–487

Lawson R., Slowinski J. B., Crother B. I., Burbrink F. T. 2005. Phylogeny of the Colubroidea (Serpentes): New evidence from mitochondrial and nuclear genes. Mol Phylogenetics Evol, 37: 581–601

Lazell J., Kolby J., Lin Y. M., Zhuang D. H., Lu. W. 1999. Rep-tiles and amphibians from Nan Ao Island, China. Postilla, 217: 1–18

Leviton A. E. 1953. A new snake of the genus Oligodon from An-nam. J Washington Acad Sci, 43: 422–424

Leviton A. E. 1960. Notes on the second specimen of the snake Oligodon annamensis Leviton. Wasamann J Biol, 18: 305–307

Leviton A. E. 1963a. Contribution to a review of Philippine snakes, I. The snakes of the genus Oligodon. Philippine J Sci, 91: 459–484

Leviton A. E. 1963b. Remarks on the zoogeography of Philippine terrestrial snakes. Proceedings of the California Academy of Sciences, 4th ser. 31: 369–416

Liu W. Z., Lathrop A., Fu J., Yang D., Murphy R. W. 2000. Phylogenetic relationships among East Asian bufonids inferred from mitochondrial DNA sequences (Anura: Amphibia). Mol Phyl Evol, 14: 423–435

Lopez T. J., Maxson L. R. 1996. Mitochondrial DNA sequence variation and genetic differentiation among colubrine snakes (Reptilia: Colubridae: Colubrinae). Biochem Syst Ecol, 23: 487–505

Maddison W. P., Maddison D. R., 1992. MacClade. Sinauer Associates. Mahendra B. C. (ed) 1984. Handbook of the snakes of India, Ceylon, Burma, Bangladesh, and Pakistan, Agra, 412 pp

Manthey U., Grossmann W. 1997. Amphibien & Reptilien Südostasiens. Natur und Tier-Verlag, Münster, 512 pp

Meggitt. 1931. Insectivorus Snakes. Nature, 128: 413 Milinkovitch M. C., Lyons-Weiler J. 1998. Finding Optimal In-

group Topologies and Convexities When the Choice of Out-groups is Not Obvious. Mol Phyl Evol, 9: 348–357

Mindell D. P., Honeycutt R. L. 1990. Ribosomal RNA in verte-brates: evolution and phylogenetic applications. Ann Rev Ecol Syst, 21: 541–566

Minton, S. A., Anderson J. A. 1963. Feeding habits of the kukri snake, Oligodon taeniolatus. Herpetologica, 19: 147

Murphy R. W., Doyle K. D. 1998. Phylophenetics: frequencies and polymorphic characters in genealogical estimation. Syst Biol, 47: 737–761

Murphy R. W., Fu J., Lathrop A., Feltham J. V., Kovac V. 2002. Phylogeny of the rattlesnakes (Crotalus and Sistrurus) inferred from sequences of five mitochondrial DNA genes, 69–92. In M. Höggren, G. W. Schuett and H. W. Green (eds.), Biology of the Vipers, Carmel, Indiana: Biological Sciences Press

Nylander J. A. A. 2004. MrModeltest v2. (Program distributed by the author). Evolutionary Biology Centre, Uppsala University

Ota H., Lin J. T. 1994. Digestive tract contents of Oligodon for-mosanus and their implications on the adaptation for reptile-egg eating in snakes. J Taiwan Mus, 47: 75–78

Pauwels O. S. G., David P., Chimsunchart C., Thirakhupt K. 2003. Reptiles of Phetchaburi Province, western Thailand: a list of species, with natural history notes, and a discussion on the biogeography at the Isthmus of Kra. Nat Hist J Chulalongkorn Univ, 3: 23–53

Pauwels O. S. G., Wallach V., David P., Chanhome L. 2002. A new species of Oligodon Fitzinger, 1826 (Serpentes, Colubridae) from southern Peninsular Thailand. Nat Hist Chulalongkorn Univ, 2: 7–18

Pope C. H. 1935. The Reptiles of China. Turtles, Crocodilians, Snakes, Lizards. New York: American Museum of Natural His-tory, 604 pp

Rodríguez-Robles J. A., de Jesús-Escobar J. M. 1999. Molecular systematics of New World lampropeltinine snakes (Colubridae): implications for biogeography and evolution of food habits. Biol J Linn Soc, 68: 355–385

Saiki R. K. 1989. The design and optimization of the PCR, 7–16. In H. A. Erlich (ed.), PCR Technology. New York: Stockton Press

Savage J. M., Cliff F. S. 1954. A new snake, Phyllorhynchus arenicola, from the Gulf of California, Mexico. Proc Biol Soc Washington, 67: 69–76

Schulz K. D. 1988. Experiences with the kukri snakes of the genus Oligodon (Boie 1827). Snake Keeper 2: 5–7

Shi P., Zheng W. R. 1985. Studies on feeding habits of the snakes of Mount Wuyi. Acta Herpetol Sinica [new ser.], 4: 149–152


Siddall M. E., Kluge A. G. 1997. Probabalism and phylogenetic inference. Cladistics 13: 313–336

Slowinski J. B., Lawson R. 2002. Snake phylogeny: evidence from nuclear and mitochondrial genes. Mol Phyl Evol, 24: 194–202

Smith M. A. 1943. The Fauna of British India Including Ceylon and Burma. Reptilia and Amphibia Volume III.—Serpentes. London: Taylor and Francis, 583 pp.

Stejneger L. 1907. Herpetology of Japan and adjacent territory. Bull US Nat Mus, 58: 1–577

Stuart B. L., Emmett D. A. 2006. A collection of amphibians and reptiles from the Cardamom Mountains, southwestern Cambodia. Fieldiana, 109: 1–27

Stuart B. L., Sok K., Neang T. 2006. A collection of amphibians and reptiles from hilly eastern Cambodia. Raffles Bull Zool, 54: 129–155

Stuebing R. B., Inger R. F. 1999. A Field Guide to the Snakes of Borneo. Kota Kinabalu: Natural History Publications (Borneo), 254 pp

Swofford D. L., 2001. PAUP* Phylogenetic Analysis Using Parsi-mony (*and other methods). Ver. 4beta 8. Sinauer Associates

Tarrío R., Rodríguez-Trelles F., Ayala F. J. 2000. Tree rooting with outgroups when they differ in their nucleotide composition from the ingroup: The Drosophila saltans and willistoni groups, a case study. Mol Phyl Evol, 16: 344–349

Taylor E. H. 1922–25 (1966). Additions to the herpetological fauna of the Philippine Islands, I–IV. Asher, Amsterdam

Taylor E. H. 1965. The serpents of Thailand and adjacent waters. Univ Kansas Sci Bull, 45: 609–1096

Taylor E. H., Elbel R. E. 1958. Contribution to the herpetology of Thailand. Univ Kansas Sci Bull, 38: 1033–1189

Teynie A., David P., Ohler A., Luanglath K. 2004. Note on a collection of amphibians and reptiles from South Laos, with a discussion of the occurrence of Indo-Malayan species. Hama-dryad, 29: 33–62

Thompson J. D., Higgins D. G., Gibson T. J. 1994. Clustal W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucl Acids Res, 22: 4673–4680

Tillack F., ünther R. 2009. Revision of the species of Oligodon from Sumatra and adjacent islands with comments on the taxo-nomic status of Oligodon subcarinatus (Günther, 1872) and Oligodon annulifer (Boulenger, 1893) from Borneo (Reptilia, Squamata, Colubridae). Russ J Herpetol, 16: 265–294

Toriba M. 1987. Feeding behavior of two species of the genus Oligodon from China. Snake, 19: 5–9

Vidal N., Kindl S., Wong A., Hedges B. 2000. Phylogenetic rela-tionships of Xenodontine snakes infered from 12S and 16S ri-bosomal RNA sequences. Mol Phyl Evol, 14: 389–402

Wagner F. W. 1975. A revision of the Asian colubrid snakes Oli-godon cinereus (Günther), Oligodon joynsoni (Smith), and Oli-

godon cyclurus (Cantor), A revision of the Asian colubrid snakes Oligodon cinereus (Günther), Oligodon joynsoni (Smith), and Oligodon cyclurus (Cantor). M.Sc. Thesis. Baton Rouge Louisiana State University, 97 pp

Wall F. 1921. Ophidia Taprobanica or the Snakes of Ceylon. Colombo: Colombo Mus. (H. R. Cottle, govt. printer), 581 pp

Wall F. 1923. A review of the Indian species of the genus Oligodon suppressing the genus Simotes (Ophidia). Calcutta: Records of the Indian Museum, 25: 305–334

Wallach V., Bauer A. M. 1996. On the identity and status of Si-motes semicinctus Peters, 1862 (Serpentes: Colubridae). Hama-dryad, 21: 13–18

Ware J. L., Litman J. R., Klass K. D., Spearman L. A. 2008. Relationships among the major lineages of Dictyoptera: The ef-fect of outgroup selection on dictyopteran tree topology. Syst Entomol, 33: 429–450

Wiley E. O. 1980. Is the evolutionary species fiction? A considera-tion of classes, individulas and historical entities. Syst Zool, 29: 76–80

Wüster W., Cox M. J. 1992. Defensive hemipenis display in the kukri snake Oligodon cyclurus. J Herpetol, 26: 238–241

Yang D., Inger R. F. 1986. Translation of: Key to the Snakes and Lizards of China. Zhao Ermi & Jiang Yaoming. 1977. Chengdu Institute of Biology, Academia Sinica. Smithsonian Herpetol Infor Service (71): 1–21

Zaher H. 1999. Hemipenial morphology of the South American xenodontine snakes, with a proposal for a monophyletic Xeno-dontinae and a reappraisal of colubroid hemipenes. Bull Am Mus Nat Hist, 240: 1–168

Zaher H., Grazziotin F. G., Cadle J. E., Murphy R. W., de Moura-Leite J. C., Bonatto S. L. 2009. Molecular phylogeny of advanced snakes (Serpentes, Caenophidia) with an emphasis on South American xenodontines: a revised classification and descriptions of new taxa. Papeis Avulsos de Zoologia (Sao Paulo) 49: 115–153

Zhang F. J., Hu S. Q., Zhao E. M. 1984. Comparative studies and phylogenetic discussions on hemipenial morphology of the Chinese Colubrinae (Colubridae). Acta Herpetol Sinica [new ser.], 3: 23–44

Zhao E. M., Huang M., Zong Y. (eds) 1998. Fauna Sinica: Reptilia Volume 3 Squamata, Serpentes. Beijing: Science Press, 522 pp

Zhao E. M., Jiang Y. M. 1981. Studies on amphibians and reptiles of Mt. Gongga Shan, Sichuan, China. I. A new species and a new subspecies of snakes from Sichuan. Acta Herpetol Sinica [old ser.], 5: 53–58

Zug G . R., Win H., Thin T., Min T. Z., Lhon W. Z., Kyaw K. 1998. Herpetofauna of the Chatthin Wildlife Sanctuary, north-central Myanmar with preliminary observations of their Natural History. Hamadryad, 23: 111–120

toward a phylogeny of the kukri snakes, genus …labs.eeb.utoronto.ca/murphy/pdfs of...

Documents