Zootaxa 3815 (1): 051–067 ISSN 1175-5326 (print edition) www.mapress.com/zootaxa/ Article ZOOTAXA Copyright © 2014 Magnolia Press ISSN 1175-5334 (online edition) http://dx.doi.org/10.11646/zootaxa.3815.1.3 http://zoobank.org/urn:lsid:zoobank.org:pub:FADC54BE-301E-40E7-9029-159082652822

A diminutive new species of cave-dwelling Wolf (: Boie, 1826) from Peninsular Malaysia

L. LEE GRISMER1, EVAN S.H. QUAH2, SHAHRUL ANUAR M.S2,3, MOHD ABDUL MUIN2, PERRY L. WOOD, JR4 & SITI AZIZAH MOHD NOR2 1Department of Biology, La Sierra University, 4500 Riverwalk Parkway, Riverside, California 92515 USA. E-mail: [email protected] 2School of Biological Sciences, Universiti Sains Malaysia, 11800 USM, Pulau Pinang, Penang, Malaysia. E-mails: [email protected], [email protected], [email protected] 3Center for Marine and Coastal Studies, Universiti Sains Malaysia, 11800 Minden, Pulau Pinang, Malaysia 4Department of Biology, Brigham Young University, 150 East Bulldog Boulevard, Provo, Utah 84602 USA. E-mail: [email protected]

Abstract

A newly discovered, diminutive, cave-dwelling, lowland species of the colubrid snake Lycodon Boie is described from a limestone cave along the Thai-Malaysian border in the state of Perlis, northwestern Peninsular Malaysia. Lycodon cavernicolus sp. nov. is most closely related to L. butleri Boulenger, an endemic, upland, forest-dwelling species from Peninsular Malaysia of the fasciatus group but is separated from L. butleri and all other species of the L. fasciatus group and the closely related L. ruhstrati group by having the combination of 245 (male) and 232 (female) ventral scales; 113 (male) and 92 (female) paired, subcaudal scales; a single precloacal plate; nine or 10 supralabials; 10 or 11 infralabials; a maximum total length of 508 mm (female); a relative tail length of 0.25–0.27; an immaculate venter in juveniles and dark- brown, posterior, ventral scale margins in adults; and dorsal and caudal bands in juveniles white. The discovery of L. cav- ernicolus sp. nov. adds to a rapidly growing list of newly discovered from karst regions and limestone forests of Peninsular Malaysia, underscoring the fact that these areas should be studied before they are quarried as they harbor a significant portion of the Peninsular Malaysia’s herpetological diversity.

Key words: new species, Lycodon, karst, limestone, cave, conservation, endemic biodiversity, Peninsular Malaysia

Introduction

Karst formations are formed through the dissolution of layers of carbonate bedrock creating caves, sinkholes, and karst towers and compose some of Peninsular Malaysia’s most spectacular landscapes. Exposed karst surfaces are particularly subject to weathering and as such, karst towers and cliff faces are often marked by deep, corrugated surfaces resulting from years of erosion and fracturing. These erosive processes also create extensive networks of cave systems that permeate karst formations and create a unique microhabitat to which a number of organisms have become adapted (see Clements et al. 2006; Vermeulen & Whitten 1999). Despite the astonishing degree of floral endemism in karst habitats and their surrounding limestone forests (Kiew 1998) karst formation are generally not considered to harbor high numbers of endemic, terrestrial vertebrates (i.e. Alström et al. 2010; Jenkins et al. 2004; Woxvold et al. 2009) and as such, remain understudied by vertebrate systematists (see Grismer et al. 2014a). In contrast to this notion, however, our recent herpetological surveys of karst regions in Peninsular Malaysia have revealed 13 endemic, karst-adapted species of geckos (eight species of Cnemaspis—Grismer et al. 2008a,b, 2009; 2013; Wood et al. 2013 and five species of Cyrtodactylus—Grismer et al. 2012, 2014a,b,c) and a new species of limestone forest-adapted colubrid snake of the genus Dendrelaphis (Quah et al. in preparation). Remarkably, we have only explored approximately 2% of the known karst formations and associated limestone forests in Peninsular Malaysia (Price 2001) and anticipate that tens of additional new species will eventually be discovered as our explorations continue.

Accepted by Z. Nagy: 21 May 2014; published: 12 Jun. 2014 51 Our surveys of the karst systems associated with the Banjaran Nakawan mountain range separating Peninsular Malaysia from southern Thailand resulted in the discovery of a new species of karst-adapted Rock Gecko (Cnemaspis biocellata Grismer, Chan, Nasir & Sumontha) and Bent-toed Gecko (Cyrtodactylus astrum Grismer, Wood, Quah, Anuar, Muin, Sumontha, Norhayati, Bauer, Wangkulangkul, Grismer & Pauwels). Continuing our surveys in this mountain range, we discovered two individuals of a small species of lowland snake from the karstic cave Gua Wang Burma in the Perlis State Park, Perlis (Fig. 1). We determined that these individuals belonged to the genus Lycodon because they have dorsoventrally flattened heads; vertically elliptical pupils; large nostrils; strongly arched upper maxillary bones whose anterior sections angle inwards; curved, enlarged, anterior maxillary teeth followed by a diastema; 17 anterior and mid-body and 15 posterior rows of dorsal scales bearing eight weakly keeled, medial rows; and rounded, weakly notched, ventral scales (Lanza 1999). We added these specimens to the molecular phylogenetic data set of Siler et al. (2013) and determined they were conspecific and most closely related to L. butleri Boulenger, an upland species endemic to Peninsular Malaysia. However, these specimens differ from L. butleri and all other species of Lycodon in having a unique suite of morphological and color pattern characters. We therefore consider it a new species and describe it below.

FIGURE 1. Distribution of Lycodon cavernicolus sp. nov. and L. butleri in Peninsular Malaysia. Stars designate type localities.

52 · Zootaxa 3815 (1) © 2014 Magnolia Press GRISMER ET AL. Material and methods

Morphological analysis. Morphological and color pattern data were taken from two specimens from Gua Wang Burma cave, Perlis and 12 specimens of Lycodon butleri from throughout its range. Data evaluated from species of the ruhstrati group were acquired from Vogel et al. (2009). Data presented for species of the fasciatus group were obtained from Zhang et al. (2011). Scale counts and scale nomenclature follow Vogel et al. (2009). All body measurements were made to the nearest millimeter. The number of ventral scales was counted according to Dowling (1951). Half ventrals were not counted, except if they were present on both sides (i.e. divided ventrals). The terminal scute (precloacal plate) was not included in the number of ventrals. The dorsal scale row counts are given at one head length behind the head, at midbody (i.e., at the level of the ventral scale corresponding to a one- half of the total number of ventrals), and at one head length before the vent. We considered sublabials as those scales that had more than one-half their length below a supralabial. The values for paired head characters are listed in Tables 2 and 3 in left/right order. The light-colored bands on the body (dorsal bands) and tail (caudal bands) were counted on one side. Indiscernible or incomplete bands were counted as a single band and bands that were partly fused were counted as two. The collar (i.e. the band on the back of the head) was not counted as a band and bands covering the precloacal plate were added to the dorsal band count. Ventral or belly pattern was considered either banded, spotted, banded anteriorly and spotted posteriorly, immaculate, or with dark-brown posterior edges on the ventral scales. Character abbreviations are SVL–Snout-vent length (mm), TaL–tail length (mm), TL–total length (mm), Rel–TL relative tail length TaL/TL, ASR–dorsal scale rows at neck, MSR–dorsal scale rows at midbody, PSR–dorsal scale rows before vent, Keel–number of keeled dorsal scale rows, VEN–number ventral plates, PreVEN–number of preventrals, VEN not–ventrals notched or not, VEN keel–ventrals keeled or not, SC–number of subcaudal scales, ANA–anal plate single or divided, Lor–number of loreal scales, Lo touch–loreal scale entering the orbit, SL–number of supralabials, SL/Eye–numbers of the supralabials entering orbit, Larg SL–largest supralabial, IL–number of infralabials, IL-tot–total number of infralabials, IL/1st child–number of infralabials in contact with the anterior chin shield, PreOc–number of preoculars, PostOc–number of postoculars. Museum abbreviations are BM–The Natural History Museum, London, UK; BNHS: Bombay Natural History; ZMB–Museum für Naturkunde, Berlin; LSUHC–La Sierra University Herpetological Collection, La Sierra University, Riverside, California, USA. USMHC–Universiti Sains Malaysia Herpetological Collection Molecular analysis. Sequence data from a 1,110 base pair fragment of the cytochrome b gene (cyt b) was obtained from six specimens of Lycodon butleri and the two specimens from the Gua Wang Burma which in total comprised the ingroup. Outgroup samples included 43 specimens from Siler et al. (2013) comprising 18 species. An additional outgroup sequence was obtained for L. albofuscus (USMHC 1457). New sequences used in this study are deposited in GeneBank (Table 1). Genomic DNA was isolated from liver tissue stored in 95% ethanol and extracted using the tissue protocol provided by the iNtRON Biotechnology G-spinTM total DNA Extraction Kit (Korea). Cyt b was amplified using a double-stranded Polymerase Chain Reaction (PCR) under the following conditions:1.0 µl genomic DNA (10–30 µg of DNA), 0.5 µl light strand primer (5μM) HI4910 5’–GACCTGTGATMTGAAAAACCAYC–3’ (Burbrink et al. 2000), 0.5 µl heavy strand primer (5μM) THRSN2 5’–CTTTGGTTTACAAGAACAATGCT–3’

(Burbrink et al. 2000), 2.5 µl of iNtRON MgCl2-free PCR Buffer, 2.5 µl of iNtRON 15 mM MgCl2, 0.2 µl iNtRON Taq DNA Polymerase, 0.5 µl dNTP Mix 10 mM each dNTP, and 17.3 µl nuclease free water. PCR reactions were completed using a G-Storm GS482 Thermal Cycler with the following reaction conditions: initial denaturation at 94°C for 2 min, second denaturation at 94°C for 35 s, annealing at 50°C for 35 s followed by an extension cycle at 72°C for 95 s +4 s per cycle for 32 cycles. PCR products were visualized using gel electrophoresis using a 2.0% agarose gel. PCR products having a distinct band with the correct molecular weight based on the standardized ladder were purified using a iNtRON Biotechnology MEGAquick-spinTM Total Fragment DNA Purification Kit (Korea) and sequenced through the MyTACG Bioscience Enterprise sequencing facility (). Sequences were analyzed from both the 3’ and 5’. Both forward and reverse sequences were assembled and edited in MEGA v5.05 (Tamura et al. 2011). Sequences were aligned by eye and to ensure the correct amino acid reading frame and MacClade v4.08 (Maddison & Maddison 2005) was used to check for premature stop codons and none were found.

A NEW CAVE-DWELLING LYCODON FROM MALAYSIA Zootaxa 3815 (1) © 2014 Magnolia Press · 53 TABLE 1. Samples used in the molecular analyses with GenBank accession numbers. LSUHC = La Sierra University Herpetological Collection, Riverside, California, USA. LSUHC Species Locality GenBank Voucher Accession Nos. 9985 Lycodon cavernicolus sp. nov. Peninsular Malaysia, Perak, Gua Wang Burma KJ607889 10500 Lycodon cavernicolus sp. nov. Peninsular Malaysia, Perak, Gua Wang Burma KJ607890 9136 Lycodon butleri Peninsular Malaysia, Perak, Bukit Larut KCO10353 9137 Lycodon butleri Peninsular Malaysia, Perak, Bukit Larut KJ607891 8365 Lycodon butleri Peninsular Malaysia, Perak, Bukit Larut KJ607892 10791 Lycodon butleri Peninsular Malaysia, Perak, Bukit Larut KJ607893 9241 Lycodon butleri Peninsular Malaysia, Perak, Bukit Larut KCO10354 8066 Lycodon butleri Peninsular Malaysia,Pahang, Fraser's Hill KCO10352

Phylogenetic analysis. We performed both Maximum Likelihood (ML) and Bayesian Inference (BI) analyses for phylogenetic reconstructions. Estimated models of molecular evolution were obtained from Siler et al. (2013:Table 1). For both the ML and BI analyses, datasets were partitioned by codon position (three partitions). The ML analysis was performed in RAxML HPC 7.5.4 (Stamatakis 2006) with a search of 200 iterations for the most likely tree and a separate analyses was ran for 1000 bootstrap pseudoreplicates via the rapid hill-climbing algorithm (Stamatakis et al. 2008). After booth runs were finished the bootstrap values were placed on the best tree. A partitioned Bayesian analysis was executed in MrBayes v3.2 (Ronquist et al. 2012). Two simultaneous runs were performed with eight chains per run, seven hot and one cold for 10,000,000 generations and sampling every 1000 generations from the Markov Chain Monte Carlo (MCMC). After 10,000,000 generations the standard deviation of split frequencies was well below 0.01 and chain convergence was assumed. The first 25% of each run was discarded as burn-in and a 50% majority ruled consensus tree was constructed using the sum function in MrBayes v3.2 (Ronquist et al. 2012). Nodal support for Bayesian posterior probabilities (BPP) above 0.95 and ML bootstrap support values (ML) greater than 70 are considered to be well supported (Huelsenbeck et al. 2001; Wilcox et al. 2002). Uncorrected percent sequence divergences were calculated in Mega v5.2.2 (Tamura et al. 2011).

Results

The molecular analysis thus far supports Vogel et al.’s (2009) hypothesized close relationship of the Lycodon ruhstrati group (L. rushstrati [Fischer]; L. davidi Vogel, Nguyen, Kingsada, & Ziegler; L. futsingensis [Pope], L. ophiophagus Vogel, David, Pauwels, Sumontha, Norval, Hendrix, Vu & Ziegler; L. paucifasciatus [Rendahl], and L. multifasciatus [Maki]) and species, that according to Vogel & David (2010), Vogel & Luo (2011) and Zhang et al. (2011), are generally considered to be part of the L. fasciatus group (L. butleri Boulenger; L. fasciatus [Anderson]; L. gongshan Vogel & Jian; and L. luichengchaoi Jun, Ke, Vogel & Dingqil) although genetic data from additional species of both groups are still needed (Fig. 2). Into the fasciatus group, we place the new population from Gua Wang Burma cave because it has the diagnostic character of a single, elongate, loreal scale that enters the orbit (Vogel et al. 2009; Fig. 3) and is the sister species of L. butleri (Fig. 2). The Gua Wang Burma population is differentiated from all members of the fasciatus group by a number of morphological and color pattern characteristics (Tables 2, 3) and has a p-distance of 9.3% from L. butleri (Table 4). As such it is described below as a new species:

54 · Zootaxa 3815 (1) © 2014 Magnolia Press GRISMER ET AL. FIGURE 2. Maximum Likelihood (-lnL 12988.774621) phylogenetic reconstruction of the genus Lycodon Boie, 1826. Bayesian posterior probabilities (BPP) and Maximum Likelihood (ML) bootstrap values (BPP/ML), respectively, are listed at the nodes.

A NEW CAVE-DWELLING LYCODON FROM MALAYSIA Zootaxa 3815 (1) © 2014 Magnolia Press · 55 TABLE 2. Diagnostic characters separating Lycodon cavernicolus sp. nov. from other species in the fasciatus group. Abbreviations are listed in the Materials and Methods. cavernicolus butleri fasciatus gongshan luichengchaoi sp. nov. VEN 245/232 220–227 182–219 210–216 202–206 SC 113/92 81–96 65–88 92–96 68/77 ANA single single single single divided SL 9 or 10 8 or 9 8 8 7 or 8 IL 10 or 11 9 or 10 8–10 8 7–9 Max total length /508 876/719 894/679 963/762 676/615 Rel TL 0.25–0.27 0.21–0.28 0.19–0.23 0.22–0.23 0.20–0.22 Dor-bands 36–45 28–37 19–37 32–40 40–45 Tail bands 29–41 17–23 7–21 9 10–15 Belly pattern 0, dark posterior banded & spotted banded banded banded edges on ventrals Band color white (juvenile) whitish whitish whitish yellow

Lycodon cavernicolus sp. nov. Gua Wang Burma Wolf Snake Figs. 3–5

Holotype. Adult female (LSUHC 9985) collected on 12 March 2011 by Evan S.H. Quah and Shahrul Anuar M.S. from Gua Wang Burma, Perlis State Park, Perlis, Peninsular Malaysia (6°41.594N 100°11.400E at 175 m in elevation). Paratype. Juvenile male (LSUHC 10500) has the same data as the holotype except for being collected on 23 May 2010. Diagnosis. Lycodon cavernicolus sp. nov. is separated from all other species of the L. ruhstrati and L. fasciatus groups by having the combination of an elongate loreal scale that enters the orbit; 245 (male) and 232 (female) ventral scales; 113 (male) and 92 (female) paired subcaudal scales; a single precloacal plate; nine or 10 supralabials; 10 or 11 infralabials; a maximum total length of 508 mm for the single female; relative tail length 0.25–0.27; venter immaculate as juveniles and with dark edging on the posterior margins of the ventral scales in adults; and bands in juveniles that are white (Tables 2,3). Description of holotype (Figs. 3–5). Head flattened anteriorly, distinct from the neck; snout elongate; nostril oval, large, in the middle of the nasal; eye large, with a vertically elliptic pupil; rostral triangular, hardly visible from above; nasal vertically divided by a furrow along posterior margin of nostril; two square internasals, in wide, medial contact, and in contact with two large, subrectangular prefrontals posteriorly; single, azygous, subpentagonal frontal, longer than wide; two large, elongate parietals, contacted laterally by upper anteriror and posterior temporals and a larger paraparietal; 1/1 wide, triangular supraocular; 1/1 small preocular, located above the posterior portion of loreal; 2/2 postoculars of similar size; 1/1 narrow, elongate loreal entering orbit, in contact with second, third, and fourth supralabials ventrally, the prefrontal and preocular dorsally, the nasal anteriorly; 9/10 supralabials all higher than wide except last scale in the series; first and second supralabials in contact with nasal; fourth, fifth, and sixth supralabials entering orbit; seventh supralabial largest; upper row of two (long anterior and short posterior) temporals; lower row of three (two anterior and one posterior) temporals; a middle posterior temporal ventral to upper posterior temporal and dorsal to lower posterior temporal; posterior temporals smaller than anterior temporals; 11/11 infralabials; first pair of infralabials separated medially by deep, medial groove; first five infralabials in contact with first pair of chinshields; anterior and posterior pair of chinshields elongate, generally same size and shape, and bearing a deep, medial groove that is confluent with groove separating first pair of infralabials.

56 · Zootaxa 3815 (1) © 2014 Magnolia Press GRISMER ET AL. Body elongate, somewhat laterally compressed; SVL 406 mm; TaL 102 mm; TL 508 mm. 232 ventrals (plus two preventrals), 92 paired, subcaudals; anal single; dorsal scales in 17–17–15 rows, the eight medial rows weakly keeled; vertebral row not enlarged; no apical pits.

FIGURE 3. Dorsal (upper) and lateral (lower) head views of the holotype of Lycodon cavernicolus sp. nov. LSUHC 9985 showing the location of head scales. F = frontal; IF = infralabial; IN = internasal; L = loreal; LAT = lower anterior temporal; LPT = lower posterior temporal; M = mental; MPT = middle posterior temporal; N = nasal; P = parietal; PF = prefrontal; PO = postocular; PP = postparietal; PrO = preocular; R = rostral; SL = supralabial; SO = suprocular; UAT = upper anterior temporal; and UPT = upper posterior temporal.

A NEW CAVE-DWELLING LYCODON FROM MALAYSIA Zootaxa 3815 (1) © 2014 Magnolia Press · 57 FIGURE 4. Left: ventral pattern of the holotype of Lycodon cavernicolus sp. nov. LSUHC 9985. Right: ventral pattern of the paratype LSUHC 10500. Photos by L.L. Grismer

Coloration in life (Figs. 4, 5). Body and tail nearly uniformly light-brown; body bearing 36 faint, lighter colored bands; tail bearing 29 similarly colored bands; head coloration same as that of the faint bands; venter ground color beige; posterior edges of ventral scales edged in light-brown, generally beginning with ventral scale 40; subcaudals mottled with light-brown. Variation (Figs. 4, 5). Differences in scalation and color pattern between the holotype, the paratype, and 12 individuals of Lycodon butleri from throughout its range are listed in Table 3. The paratype is a hatchling and bears a bold, contrasting, dorsal color pattern similar to that of juvenile Lycodon butleri (Fig. 5). Its venter however, is nearly immaculate unlike the holotype whose ventrals are edged posteriorly with dark-brown (Fig. 4). It also has 45 irregularly shaped, whitish bands with darkened centers on the body and 41 similarly colored caudal bands. A wide, white band covers the occipital and posterior temporal regions. The anterior 11 bands on the body are more widely separated and distinct than the posterior body and caudal bands. Presumably, the banding pattern fades considerably with maturity as in L. butleri but not to the extent observed in L. cavernicolus sp. nov. (Figs. 5, 7). Etymology. The specific epithet “cavernicolus” is an adjective derived from the Latin caverna meaning “cave” and the Latin cola meaning “dweller of” and refers to this species being a cave-dweller. Natural history. Both the holotype and paratype were found deep within Gua Wang Burma cave approximately 200 m from the cave entrance (Fig. 6). Both specimens were found at approximately 1100 hrs. The holotype was observed scaling a vertical wall approximately 2 m above the ground while exploring nooks and crevices. She was gravid but expelled two eggs soon after capture. The paratype was found crawling over a slanting cave wall approximately 3 m above the ground in a more exposed area. Other species of amphibians and reptiles observed in the cave or near the cave entrance were the bufonid Phrynoides aspera, the gekkonids Cnemaspis biocellata, Cyrtodactylus astrum, and C. macrotuberculatus and the colubrid Othriophis taeniurus ridleyi.

58 · Zootaxa 3815 (1) © 2014 Magnolia Press GRISMER ET AL. A NEW CAVE-DWELLING LYCODON FROM MALAYSIA Zootaxa 3815 (1) © 2014 Magnolia Press · 59 FIGURE 5. Upper: holotype of Lycodon cavernicolus sp. nov. LSUHC 9985. Middle: paratype of L. cavernicolus sp. nov. LSUHC 10500. Photos by E.S.H. Quah. Lower: juvenile L. butleri LSUHC 9421 from Bukit Larut, Perak. Photo by L.L. Grismer.

60 · Zootaxa 3815 (1) © 2014 Magnolia Press GRISMER ET AL. FIGURE 6. Microhabitat inside the Gua Wang Burma cave at the type locality in Perlis State Park, Perlis. Photos by L.L. Grismer.

The gravid holotype indicates that the reproductive season of Lycodon cavernicolus sp. nov. extends through March. The only potential food source we found deep within cave is Cyrtodactylus astrum (juvelies) which are also known to occur on the cave walls (Grismer et al. 2012). Comparisons. Lycodon cavernicolus sp. nov. is distinguishable from all species of the ruhstrati group by having a single loreal scale that enters the orbit as opposed to the loreal scale not entering the orbit. From the species of the fasciatus group it differs by having more ventral scales (232–245 vs. 182–227 collectively); more subcaudal scales in the male (113 vs. 65–92 collectively); a much smaller adult female total length (508 vs. 615–762 collectively); more caudal bands (29–41 vs. 7–23 collectively); and a belly pattern that lacks wide, dark bands or dark spots (Table 2). Lycodon cavernicolus sp. nov. is further separated from its closest relative L. butleri by the loreal and internasals being separat as opposed to contacting and having an uncorrected p-distance of 9.3% (Table 4).

Discussion

Lycodon cavernicolus sp. nov. is not the first species of Lycodon purported to be associated with limestone habitats. Vogel et al. (2012) indicated that L. davidi seems to be associated with limestone formations based on the discovery of the holotype crawling on the forest floor outside a cave entrance. Unlike Peninsular Malaysia’s six forest-dwelling species of Lycodon (Fig. 8), we believe there is little doubt as to the cave-microhabitat specificity of L. cavernicolus sp. nov. Both specimens were found on vertical limestone surfaces 2–3 m above the cave floor in total darkness approximately 200 m from the cave entrance. The banded color pattern is essentially lost in adulthood which is adaptive for substrate matching and the considerably smaller adult size as compared to those of its closest relatives in the ruhstrati and fasciatus groups would be adaptive for living in a microhabitat with a reduced prey base. The molecular analysis indicates that Lycodon butleri from Fraser’s Hill, Pahang is nearly genetically identical to, and embedded within, the population of L. butleri from Bukit Larut, Perak which occurs on a different mountain range approximately 110 km to the northwest (Fig. 1). Loredo et al. (2013) noted the same pattern of low genetic

A NEW CAVE-DWELLING LYCODON FROM MALAYSIA Zootaxa 3815 (1) © 2014 Magnolia Press · 61 62 · Zootaxa 3815 (1) © 2014 Magnolia Press GRISMER ET AL. FIGURE 7. Adult Lycodon butleri. Upper: LSUHC 8066 from Fraser’s Hill, Pahang. Photo by L.L. Grismer. Middle: ZMB 57114 from Cameron Highlands, Pahang. Photo by G. Vogel. Lower: LSUHC 9137 from Bukit Larut, Perak.

A NEW CAVE-DWELLING LYCODON FROM MALAYSIA Zootaxa 3815 (1) © 2014 Magnolia Press · 63 FIGURE 8. Upper left: Lycodon albofuscus LSUHC 10918 from Fraser’s Hill, Pahang. Upper right: L. laoensis LSUHC 10026 from Perlis State Park, Perlis. Middle left: L. butleri LSUHC 9137 from Bukit Larut, Perak. Middle right: L. subcinctus LSUHC 10910 from Hutan Lipur Lata Belatan, Terengganu. Lower left: L. capucinus LSUHC 8964 from Pulau Perhentian Besar, Terengganu. Lower right: L. effraenis LSUHC 7734 from Endau-Rompin, Johor. Photos by L.L. Grismer. divergence and morphological similarity in the allopatric, upland populations of the Slug Asthenodipsas lasgalenensis Loredo, Wood, Quah, Anuar, Greer, Norhayati & Grismer and A. vertebralis (Boulenger) from the same mountain ranges. Loredo et al. (2013) hypothesized that their isolation on separate mountain ranges resulted from their association with the recent upslope migrations of cool-adapted forests to upland regions as a result of rising, lowland temperatures following the Last Glacial Maximum, approximately 11 thousand years before present (Woodruff 2010). The phylogeographic pattern, low genetic diversity, and morphological similarity observed here in L. butleri (Table 4; Fig. 7) match those of A. lasgalenensis and A. vertebralis and we hypothesize they are the result of the same historical, environmental processes. We further hypothesize that during one of the earlier, glacial periods when upland forests and their associated fauna were widespread in cooler, lowland regions, members of a population ancestral to L. cavernicolus sp. nov. and L. butleri utilized the Gua Wang Burma cave and became isolated in this cooler microhabitat during a subsequent rise in temperature and upslope retreat of cool-adapted forests. This sister species relationship between upland and lowland taxa is a common pattern beginning to emerge in all the major mountain ranges across Peninsular Malaysia as more species-level phylogenies are generated. For example, the two upland species Cyrtodactylus bintangtinggi Grismer, Wood, Quah, Anuar, Muin, Sumontha,

64 · Zootaxa 3815 (1) © 2014 Magnolia Press GRISMER ET AL. Norhayati, Bauer, Wangkulangkul, Grismer & Pauwels and Cnemaspis mcguirei Grismer, Grismer, Wood & Chan from Bukit Larut in the Banjaran Bintang mountain range are the sister species of Cyrtodactylus bintangrendah Grismer, Wood, Quah, Anuar, Muin, Sumontha, Norhayati, Bauer, Wangkulangkul, Grismer & Pauwels and Cnemaspis grismeri Wood, Quah, Anuar, & Muin, respectively (Grismer et al. 2012; Grismer et al. in prep.), from the adjacent lowlands. Similarly, in the Banjaran Titiwangsa mountain range, the upland Cyrtodactylus trilatofasciatus Grismer, Wood, Quah, Anuar, Muin, Sumontha, Norhayati, Bauer, Wangkulangkul, Grismer, & Pauwels is the sister species to the lowland C. sharkari Grismer, Wood, Anuar, Quah, Muin, Mohamed, Chan, Sumarli, Loredo & Heinz (Grismer et al. 2014b) and the upland Cnemaspis sp. Grismer, Wood, Grismer, Chan, Quah, Anuar, Riyanto, Norhayati, Muin, Sumontha & Pauwels is the sister species to the lowland C. bayuensis Grismer, Grismer, Wood, & Chan (Grismer et al. in prep.). This same pattern is also beginning to emerge in the Banjaran Timur mountain range in northeastern Peninsular Malaysia where the upland Cyrtodactylus tebuensis Grismer, Anuar, Muin & Quah is the sister species of the lowland C. sworderi (Smith) (Grismer et al. 2013). The discovery of Lycodon cavernicolus sp. nov. adds to a growing body of evidence that karst regions should be better studied prior to being quarried for cement. If reptiles are an indication of the hidden diversity within these unique habitats, then limestone forests may be some of the most biotically rich habitats in Peninsular Malaysia with a level of herpetological endemism commensurate with that of Malaysia’s islands (Chan et al. 2010; Grismer 2011a,b; Grismer et al. 2011).

Key to the species of Lycodon from Peninsular Malaysia

1. Loreal and preocular present ...... 2 Either loreal or preocular absent ...... 6 2. Loreal in contact with eye ...... 3 Loreal not in contact with eye; dorsal scales smooth ...... 5 3. Precloacal plate entire ...... 4 Precloacal plate divided; dorsal scales strongly keeled; ventrals keeled and notched; body elongate and slender; adults uniform grey ...... albofuscus 4. Loreal in contact with internasals; eight or nine supralabials; nine or 10 infralabials; dark ground color with 28–36 lighter bands on body; 14-23 light bands on tail ...... butleri Loreal not in contact with internasals; nine or 10 supralabials; 10 or 11 infralabials; dull-brown color with 36–45 faint bands on body; 29–41 light bands on tail ...... cavernicolus sp. nov. 5. No light cross-bands on body and tail; fine reticulated pattern on dorsum; light band on nape...... capucinus Body and tail bearing yellow bands on dark ground color; belly immaculate ...... laoensis 6. Preocular absent; prefrontals and loreals contact eye; precloacal divided ...... subcinctus Loreal absent; prefrontals contact 2nd and 3rd supralabials; anal entire ...... effraenis

Acknowledgements

We thank the Perlis State Forestry Department for their permission to conduct research in the Perlis State Park. Gernot Vogel read an early draft of the manuscript and provided data on several specimens of Lycodon butleri. This research was supported in part by grants to LLG from the College of Arts and Sciences, La Sierra University and Universiti Sains Malaysia Grant 811191 to Shahrul Anuar. Field work for EQSH was partially supported by the USM Fellowship Scheme.

References

Alström, P., Davidson, P., Duckworth, J.W., Eames, J.C., Trai, T.L., Nguyen, C., Ollson, U., Robinson, C. & Timmins, R. (2010) Description of a new species of Phylloscopus warbler from and Laos. Ibis, 152, 145–168. Burbrink, F.T., Lawson, R. & Slowinski, J.B. (2000) Mitochondrial DNA Phylogeography of the polytypic North American rat snake (Elaphe obsoleta): a critique of the subspecies concept. Evolution, 54, 2107–2118. http://dx.doi.org/10.1111/j.0014-3820.2000.tb01253.x Chan, K.O., van Rooijen, J., Grismer, L.L., Belabut, D., Akil, M.A.M.M., Jamaludin, H., Gregory, R. & Norhayati A. (2010) First report on the herpetofauna of Pulau Pangkor, Perak, Malaysia. Russian Journal of Herpetology, 17, 139–146.

A NEW CAVE-DWELLING LYCODON FROM MALAYSIA Zootaxa 3815 (1) © 2014 Magnolia Press · 65 Clements, R., Sodhi, N.S., Schilthuizen, M. & Ng, P.K.L. (2006) Limestone karsts of Southeast Asia: imperiled arks of biodiversity. Bioscience, 56, 733–742. http://dx.doi.org/10.1641/0006-3568(2006)56[733:lkosai]2.0.co;2 Dowling, H.G. (1951) A proposed standard system of counting ventral in snakes. Journal of Herpetology, 1, 97–99. Drummond, A.J., Ashton, B., Buxton, S., Cheung, M., Cooper, A., Duran, C., Field, M., Heled, J., Kearse, M., Markowitz, S., Moir, R., Stones-Havas, S., Sturrock, S., Thierer, T. & Wilson, A. (2011) Geneious V5.4. Available from: http:// www.geneious.com/ (accessed 3 June 2014) Grismer, L.L. (2011a) Lizards of Peninsular Malaysia, Singapore and Their Adjacent Archipelagos. Edition Chimaira, Frankfurt am Main, 728 pp. Grismer, L.L. (2011b) Field Guide to the Amphibians and Reptiles of the Seribuat Archipelago, Peninsular Malaysia. Edition Chimaira, Frankfurt am Main, 258 pp. Grismer, L.L., Anuar, S., Muin, M.A., Quah, E.S.H. & Wood, P.L. Jr. (2013) Phylogenetic relationships and description of a new upland species of Bent-toed Gecko (Cyrtodactylus Gray, 1827) of the C. sworderi complex from northeastern Peninsular Malaysia. Zootaxa, 3616 (3), 239–252. http://dx.doi.org/10.11646/zootaxa.3616.3.2 Grismer, L.L,, Belabut, D.M, Quah, E.S.H., Chan, K.O., Wood, P.L. Jr. & Hasim, R. (2014c) A new species of karst forest- adapted Bent-toed Gecko (genus Cyrtodactylus Gray, 1827) belonging to the C. sworderi complex from a threatened karst forest in Perak, Peninsular Malaysia. Zootaxa, 3755 (5), 434–446. http://dx.doi.org/10.11646/zootaxa.3755.5.3 Grismer, L.L., Chan, K.O., Nurolhuda, N. & Sumontha, M. (2008a) A new species of karst dwelling gecko (genus Cnemaspis Strauch 1887) from the border region of Thailand and Peninsular Malaysia. Zootaxa, 1875, 51–68. Grismer, L.L., Grismer, J.L., Wood, P.L. Jr. & Chan, K.O. (2008b) The distribution, , and redescription of the geckos Cnemaspis affinis (Stoliczka 1887) and C. flavolineata (Nicholls 1949) with descriptions of a new montane species and two new lowland, karst-dwelling species from Peninsular Malaysia. Zootaxa, 1931, 1–24. Grismer, L.L., Grismer, J.L., Wood, P.L. Jr., Ngo, V.T. & Chan, K.O. (2011) Herpetology on the fringes of the Sunda Shelf: a discussion of discovery, taxonomy, and biogeography. Bonner Zoologische Monographien, Bonn, 57, 57–97. Grismer, L.L., Norhayai, A., Chan, K.O., Belabut, D., Muin, M.A., Wood, P.W. Jr. & Grismer, J.L. (2009) Two new diminutive species of Cnemaspis Strauch 1887 (: Gekkonidae) from Peninsular Malaysia. Zootaxa, 2019, 40–56. Grismer, L.L., Wood, P.L. Jr., Anuar, S., Quah, E.S.H., Muin, M.A., Maketab, M., Chan, K.O., Heinz, H.M., Sumarli, A.S.-I., Loredo, A.I. & Heinz, H. (2014b) The phylogenetic relationships of three new species of the Cyrtodactylus pulchellus complex (Squamata: Gekkonidae) from poorly explored regions in northeastern Peninsular Malaysia. Zootaxa, 3786 (3), 359–381. http://dx.doi.org/10.11646/zootaxa.3786.3.6 Grismer, L.L., Wood, P.L. Jr., Chan, K.O., Anuar, S. & Muin, M.A. (2014a) Cyrts in the city: A new Bent-toed Gecko (Genus Cyrtodactylus) is the only endemic species of vertebrate from Batu Caves, Selangor, Peninsular Malaysia. Zootaxa, 3774 (4), 381–394. http://dx.doi.org/10.11646/zootaxa.3774.4.6 Grismer, L.L., Wood, P.L. Jr., Maketab, M., Chan, K.O., Heinz, H.M., Sumarli, A.S.-I., Chan, J.A. & Loredo, A.I. (2013) A new species of karst-adapted Cnemaspis Strauch, 1887 (Squamata: Gekkonidae) from a threatened karst region in Pahang, Peninsular Malaysia. Zootaxa, 3746 (3), 463–472. http://dx.doi.org/10.11646/zootaxa.3746.3.5 Grismer, L.L., Wood, P.L. Jr., Quah, E.S.H., Shahrul, A., Muin, M.A., Sumontha, M., Norhayati, A., Bauer, A.M., Wangkulangkul, S., Grismer, J.L. & Pauwels, O.S.G. (2012) A phylogeny and taxonomy of the Thai-Malay Peninsula Bent-toed Geckos of the Cyrtodactylus pulchellus complex (Squamata: Gekkonidae): combined morphological and molecular analyses with descriptions of seven new species. Zootaxa, 3520, 1–55. Huelsenbeck, J.P., Ronquist, F., Nielsen, R. & Bollback, J.P. (2001) Bayesian Inference of Phylogeny and Its Impact on Evolutionary Biology. Science, 294, 2310–2314. [Washington D.C.] Jenkins, P.D., Kilpatrick, C., William, C., Robinson, M.F. & Timmins, R.J. (2004) Morphological and molecular investigations of a new family, genus and species of rodent (Mammalia: Rodentia: Hystricognatha) from Lao PDR. Systematics and Biodiversity, 2, 419–454. http://dx.doi.org/10.1017/s1477200004001549 Kiew, R. (1998) Limestone, Quartzite and Ultramafic Vegetation. In: Soepadmo, E. (Ed.), The Encyclopedia of Malaysia: Plants. Editions Didier Miller, Singapore, pp. 26–27. Lanza, B. (1999) A new species of Lycodon from the Philippines, with a key to the genus. Tropical Zoology, 12, 89–104. Loredo, A.I., Wood, P.L., Jr., Quah, E.S.H., Anuar, S.H., Greer, L., Norhayati, A. & Grismer, L.L. (2013) Cryptic speciation within Asthenodipsas vertebralis (Boulenger, 1900) (Squamata: Pareatidae), the description of a new species from Peninsular Malaysia, and the resurrection of A. tropidonotus (Lidth de Jude, 1923) from Sumatra: an integrative taxonomic analysis. Zootaxa, 3664 (4), 505–524. http://dx.doi.org/10.11646/zootaxa.3664.4.5 Maddison, D.R. & Maddison, W.P. (2005) Macclade 4: Analysis of Phylogeny and Character Evolution. Version 4.08a. Available from: http://macclade.org/ (accessed 3 June 2014)

66 · Zootaxa 3815 (1) © 2014 Magnolia Press GRISMER ET AL. Price, L. (2001) Caves and Karst of Peninsular Malaysia. Gua Publications, Malaysia, pp. 3–98. Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Höhna, S., Larget, B., Liu, L., Suchard, M.A. & Huelsenbeck, J.P. (2012) Mrbayes 3.2: Efficient Bayesian Phylogenetic Inference and Model Choice across a Large Model Space. Systematic Biology, 61, 539–542. http://dx.doi.org/10.1093/sysbio/sys029 Siler, C.D., Oliveros, C.H., Santanen, A. & Brown, R.M. (2013) Multilocus phylogeny reveals unexpected diversification patterns in Asian Wolf Snakes (Genus Lycodon). Zoologica Scripta, 42, 262–277. http://dx.doi.org/10.1111/zsc.12007 Stamatakis, A. (2006) Raxml-Vi-Hpc: Maximum Likelihood-Based Phylogenetic Analyses with Thousands of Taxa and Mixed Models. Bioinformatics, 22, 2688–2690. http://dx.doi.org/10.1093/bioinformatics/btl446 Stamatakis, A., Hoover, P. & Rougemont, J. (2008) A Rapid Bootstrap Algorithm for the Raxml Web Servers. Systematic Biology, 57, 758–771. http://dx.doi.org/10.1080/10635150802429642 Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M. & Kumar, S. (2011) Mega5: Molecular Evolutionary Genetics Analysis Using Maximum Likelihood, Evolutionary Distance, and Maximum Parsimony Methods. Molecular Biology and Evolution, 28, 2731–2739. http://dx.doi.org/10.1093/molbev/msr121 Tweedie, M.W. F. (1983) The Snakes of Malaya. 3rd edition. Singapore National Printers, Singapore, pp. 167, Pls. 1–12. Vermeulen, J. & Whitten, T. (1999) Biodiversity and Cultural Property in the Management of Limestone Resources – Lessons from East Asia. World Bank, Washington, D.C., 120 pp. Vogel, G. & David, P. (2010) A new species of the genus Lycodon (Boie, 1826) from Yunnan Province, (Serpentes: Colubridae). Bonn Zoological Bulletin, 57, 289–296. Vogel, G., David, P., Pauwels, O.S.G., Sumontha M., Norval G., Hendrix, R., Vu, N.T. & Ziegler, T. (2009) A revision of Lycodon ruhstrati (Fischer 1886) auctorum (Squamata Colubridae), with the description of a new species from Thailand and a new subspecies from the Asian mainland. Tropical Zoology, 22, 131–182. Vogel, G. & Luo, J. (2011) A new species of the genus Lycodon (Boie, 1826) from the southwestern China (Squamata: Colubridae). Zootaxa, 2807, 29–40. Vogel, G., Nguyen, T.Q., Kingsada, P. & Ziegler, T. (2012) A new species of the genus Lycodon Boie, 1826 from Laos (Squamata: Colubridae). North-Western Journal of Zoology, 8, 344–352. Wilcox, T.P., Zwickl, D.J., Heath, T.A. & Hillis, D.M. (2002) Phylogenetic Relationships of the Dwarf Boas and a Comparison of Bayesian and Bootstrap Measures of Phylogenetic Support. Molecular phylogenetics and evolution, 25, 361–371. http://dx.doi.org/10.1016/s1055-7903(02)00244-0 Wood, P.L. Jr., Quah, E.S.H., Anuar, S. & Muin, M.A. (2013) A new species of lowland karst dwelling Cnemaspis Strauch 1887 (Squamata: Gekkonidae) from northwestern Peninsular Malaysia. Zootaxa, 3691 (5), 538–558. http://dx.doi.org/10.11646/zootaxa.3691.5.2 Woodruff, D.S. (2010) Biogeography and conservation in Southeast Asia: how 2.7 million years of repeated environmental fluctuations affect today's patterns and the future of the remaining refugial-phase biodiversity. Biodiversity Conservation, 19, 919–941. http://dx.doi.org/10.1007/s10531-010-9783-3 Woxvold, I.A., Duckworth, J.W. & Timmins, R.J. (2009) An unusual new bulbul (Passeriformes: Pycnotidae) from the limestone karst of Lao PDR. Forktail, 25, 1–12. Zhang, J., Jiang, K., Vogel, G. & Rao, D. (2011) A new species of the genus Lycodon (Squamata: Colubridae) from province, China. Zootaxa, 2982, 59–68.

A NEW CAVE-DWELLING LYCODON FROM MALAYSIA Zootaxa 3815 (1) © 2014 Magnolia Press · 67