Research Article |
Corresponding author: Bryan L. Stuart ( bryan.stuart@naturalsciences.org ) Academic editor: Thomas Ziegler
© 2020 Thy Neang, Adam Henson, Bryan L. Stuart.
This is an open access article distributed under the terms of the CC0 Public Domain Dedication.
Citation:
Neang T, Henson A, Stuart BL (2020) A new species of Cyrtodactylus (Squamata, Gekkonidae) from Cambodia’s Prey Lang Wildlife Sanctuary. ZooKeys 926: 133-158. https://doi.org/10.3897/zookeys.926.48671
|
Cyrtodactylus phnomchiensis sp. nov. is described from Phnom Chi, an isolated mountain in Prey Lang Wildlife Sanctuary, Kampong Thom Province, Cambodia. The new species is recognized by having a unique combination of morphological characters, including snout-vent length 76.1–80.7 mm; paravertebral tubercles 31–36; ventral scales 45–54; enlarged femoral scales 0–8, without pores; enlarged precloacal scales 7–10, bearing pores 4–5 in males, pits 1–7 in females; the posterior border of nuchal loop unbroken and pointed, bordered anteriorly and posteriorly by a broad yellow or yellowish white band; and yellow spots on top of head. The new species also represents a divergent mitochondrial DNA lineage within the C. irregularis complex that is closely related to C. ziegleri, but the phylogenetic relationships among the new species and two divergent mitochondrial subclades within C. ziegleri are not resolved based on available sequence data. Cyrtodactylus phnomchiensis sp. nov. is the only member of the C. irregularis complex known to occur west of the Mekong River. The new species may be endemic to Phnom Chi, and likely faces imminent conservation threats.
Cyrtodactylus irregularis, C. ziegleri, Mekong River, Phnom Chi, Sphenomorphus preylangensis
Bent-toed Geckos of the genus Cyrtodactylus Gray are one of the most species-diverse genera of gekkonid lizards, with 292 recognized species (Uetz et al. 2020). Much of the diversity within Cyrtodactylus has been described only during the past decade and from mainland Southeast Asia (
Cyrtodactylus irregularis (Smith, 1921) was originally described from the Langbian Plateau near Da Lat, southern Vietnam. For nearly a century, C. irregularis was treated as a single, geographically widespread, but morphologically variable species. Recent taxonomic studies on variation in morphology and, usually, the mitochondrial cytochrome c oxidase subunit I (COI) gene (
During field surveys by Wild Earth Allies in June–July 2019, five specimens of the C. irregularis complex were collected in Cambodia on the western side of the Mekong River at Phnom Chi (Mountain) in Prey Lang Wildlife Sanctuary, Kampong Thom Province. Herein, we investigate the taxonomic status of the Phnom Chi specimens through comparisons of morphological and mitochondrial DNA data with other members of the C. irregularis complex.
Field work was conducted both day and night to search microhabitats for amphibians and reptiles at Phnom Chi. Specimens were collected by hand and kept overnight in individual plastic or cloth bags for photographing the following day. Specimens were euthanized by cardiac injection of high concentration of tricaine methanesulfonate (MS-222) and fixed in 10% formalin after preserving liver tissue in 20% DMSO-salt saturated storage buffer. After a minimum of three days of formalin-fixation, the specimens were soaked in water for six hours to remove formalin, and transferred to 70% ethanol for permanent storage. Specimens were deposited in the herpetological collection at the Centre for Biodiversity Conservation, Royal University of Phnom Penh, Cambodia (
Morphometric and meristic characters were measured and counted using a Nikon SMZ 645 dissecting microscope. Measurements were taken by hand with digital calipers to the nearest 0.1 mm (ratios calculated to 0.001). Measured characters were:
AG Axilla-groin distance, measured from the posterior margin of forelimb at its insertion point on the body to the anterior margin of hind limb at its insertion point on the body;
CrusL Crus length, measured from the knee to the base of the heel;
EarDH Ear diameter in horizontal distance, measured as the horizontal distance between anterior and posterior margins of the ear opening;
EarDV Ear diameter in vertical distance, measured as the vertical distance between dorsal and ventral margins of the ear opening;
END Eye-nostril distance, measured from the anterior margin of eye to the posterior margin of nostril;
ESD Eye-snout distance, measured from the anterior margin of eye to the tip of snout;
EyeD Eye diameter, measured as the horizontal distance from the anterior to the posterior margins of the eyeball;
Eye-EarD Eye-ear distance, measured from the posterior margin of eye to the anterior margin of ear opening;
ForeL Forearm length, measured from the posterior margin of elbow while flexed 90° to the wrist inflection;
HeadD Head depth, measured as the maximum depth of head from the occiput to the throat;
HeadL Head length, measured from the tip of snout to the posterior margin of the retroarticular process of the lower jaw;
HeadW Head width, measured as the maximum head width at the corners of the jaws;
IOD Interorbital distance, measured as the shortest distance between the anterior corners of the eyes;
IND Internarial distance, measured as the shortest distance between the nostrils;
SVL Snout to vent length, measured from the tip of the snout to the vent;
TaL Tail length, measured from the vent to the tip of the tail;
TaW Tail width, measured at the base of the tail immediately posterior to the post-cloacal swelling.
Scale counts are reported in right and left (R, L) order. The presence, absence and/or numbers of the following characters were recorded:
EFS Enlarged femoral scales;
EPrecS Enlarged precloacal scales;
FP Femoral pores;
InL Infralabials, counted as the number of scales from the first lower labial scale immediately posterior to mental to the last scale below posterior edge of the eyeball;
LDRT Longitudinal dorsal rows of enlarged tubercles, counted as the number of tubercles transversely across the dorsum between ventrolateral folds;
PrecG Precloacal groove;
PrecP Precloacal pores;
PostPSR Post precloacal scale rows;
PostSP Post cloacal spur;
PVT Paravertebral tubercles, counted as the number of enlarged tubercles in a straight line between limb insertions left of the vertebral column;
SDLF4 Subdigital lamellae beneath fourth finger, counted as the number of both expanded proximal subdigital lamellae from the base to the largest scale on the digital inflection, and unmodified distal lamellae beneath fourth finger to the claw sheath;
SDLT4 Subdigital lamellae on fourth toe, counted as the number of expanded proximal subdigital lamellae from the base to the largest scale on digital inflection and unmodified distal subdigital lamellae beneath fourth toe to the claw sheath;
SL Supralabials, counted as the number of scales from the first upper labial scale immediately posterior to rostral to the last scale below posterior edge of the eyeball;
VS Ventral scales, counted as the number of scales transversely across the ventral surface at midbody between ventrolateral folds.
Total genomic DNA was extracted from preserved liver tissue of two Phnom Chi specimens (
All available Cyrtodactylus COI sequences (n = 453), and the outgroup Hemidactylus frenatus (GenBank accession GQ245970), were downloaded from GenBank on 1 October 2019. The downloaded sequences were aligned and visually inspected in Sequencher to ensure that insertion-deletions did not disrupt the coding region. Preliminary phylogenetic analysis (not shown) was performed on the alignment under the parsimony criterion using a heuristic search with equal weighting of nucleotide substitutions in PAUP* version 4.0a165 (
The resulting pruned COI alignment contained 270 taxa and 717 characters, with no insertion-deletions. The alignment was partitioned by codon position, and the best-fit partitioning scheme and models of sequence evolution were selected using PartitionFinder 2 (
The Phnom Chi specimens could not be referred to any other named members of the C. irregularis complex owing to having a unique combination of morphological characters. These characters included body size such as having a relatively long body and tibia; scalation, such as the number of subdigital lamellae under the fourth finger and fourth toe, number of longitudinal dorsal and paravertebral rows of tubercles, number of ventral scales, number of enlarged precloacal scales and associated pores (in males) and pits (in females), absence of pores in their enlarged femoral scales, and size of the median subcaudal scale rows from other species in the complex; and pattern and coloration, including an unbroken nuchal loop bordered anteriorly and posteriorly by a broad yellow or yellowish white band, three or four dark brown body bands, and two or three yellowish white or light brown body bands, about half the width of the brown body bands, and yellow spots on top of the head.
The standard deviation of split frequencies among the four Bayesian runs was 0.006260 and the Estimated Sample Sizes (ESS) of parameters were ≥ 1,606, indicating that the four runs were sufficiently sampled and had converged. The Phnom Chi specimens represented a distinct mitochondrial lineage that did not match any other named species (Fig.
The Phnom Chi samples had uncorrected p-distances in COI of 4.3–6.2% from C. ziegleri (all samples) and 7.0–8.6% from C. bugiamapensis. Cyrtodactylus ziegleri (all samples) had uncorrected p-distances of 6.7–8.5% from C. bugiamapensis. Cyrtodactylus ziegleri subclade Z1 had uncorrected p-distances of 4.7–5.2% from C. ziegleri subclade Z2.
Map illustrating (1) the type locality of Cyrtodactylus phnomchiensis sp. nov. at Prey Lang Wildlife Sanctuary, Kampong Thom Province, Cambodia; (2) the type locality of C. ziegleri at Chu Yang Sin National Park, Dak Lak Province, Vietnam (
Upper (A) and lower (B) portions of a fifty percent majority-rule consensus phylogram resulting from partitioned Bayesian analysis of 717 aligned characters of the mitochondrial cytochrome c oxidase subunit I (COI) gene from geckos in the Cyrtodactylus irregularis group. The outgroup Hemidactylus frenatus (GenBank accession GQ245970) and exemplars of other Cyrtodactylus clades including C. auribalteatus (GenBank accession AP018116), C. badenensis (KF929505), C. chanhomeae (MF169908), C. interdigitalis (MF169919), C. intermedius (MF169920), C. jellesmae (MF169923), C. peguensis (AP018114), C. russelli (MF169938), and C. thirakhupti (AP018115) were also included in the analysis (not shown). Black circles at nodes indicate Bayesian posterior probabilities ≥ 0.99, and open circles at nodes indicate Bayesian posterior probabilities ≥ 0.95. Numbers at terminal tips are GenBank accession numbers.
On the basis of their distinctiveness in morphology and mitochondrial DNA, including from C. ziegleri to which they are phylogenetically related (but exact relationship unresolved; Fig.
All from Cambodia, Kampong Thom Province, Sandan District, Phnom Chi, Prey Lang Wildlife Sanctuary:
The specific epithet is taken from the type locality of Phnom Chi and the Latin suffix -ensis meaning “originating from.” The specific epithet is masculine in agreement with the gender of Cyrtodactylus.
Cyrtodactylus phnomchiensis sp. nov. is distinguished from the 19 other named species in the C. irregularis group (
Adult male with SVL 76.1 mm; head slightly elongate, HeadL 22.1 mm, about 30% of SVL, moderately widened, HeadW 14.1 mm, HeadW/HeadL 0.64, slightly depressed, HeadD 9.4 mm, HeadD/HeadL 0.43, distinct from neck, triangular in dorsal profile; snout rather elongated, rounded in rostral region, ESD 9.0 mm, slightly less than HeadD, ESD/HeadL 0.41, frontonasal region flattened, prefrontal region slightly concave, forming elongated medial rostral groove, canthus rostralis flattened, slightly angled between loreal region and rostral groove; lores posterior to nostrals depressed, anterior to orbit flattened; eye large, eyeball rounded, slightly protruding, EyeD 5.1 mm, shorter than the distance between eye and ear, Eye-EarD 5.7 mm, pupil vertical, covered by crenellate supraciliaries; ear opening oval, deeply sunk, rather small, elongated in oblique position, EarDV 1.2 mm, almost twice longer than its diameter in horizontal position, EarDH 0.7 mm; rostral large, subrectangular, height 1.9 mm, shorter than its width 3.6 mm, medially divided dorsally by a suture, reaching to about half way of rostral height, in contact with 1st SL and nostrils laterally, supranasals and internasal dorsally (Fig.
Body slightly slender, AG 36.6 mm, nearly half SVL, AG/SVL 0.481 with well-defined narrow vertebral furrow posteriorly; scales on dorsum small, mostly homogenous, granular, interspersed with larger, low, weakly keeled, irregularly arranged, tubercles; longitudinal dorsal rows of enlarged tubercles approximately 18; paravertebral tubercles 32; tubercles on nape within dark brown nuchal loop, anterior dorsal surface at level above shoulder smaller, more rounded, sparser than those on mid-dorsum and posterior dorsal surface, more prominent, being denser, weekly keeled, more regularly arranged on sacral and tail base region; tubercles on lateral body sparsely; ventral scales small, not imbricate, those near midline larger than lateral and dorsal scales; scales on throat and gular region the smallest; faint ventrolateral folds with few emerged tubercles; ventral scales at midbody between ventrolateral folds 47; precloacal region moderately enlarged, a few rows of enlarged precloacal non-pore bearing scales anterior to pore bearing precloacal scales; enlarged precloacal scales 7, in angular series, bearing 5 pores, terminal scale on each side poreless; post precloacal scale rows 3, the first row immediately posterior to enlarged precloacal pore-bearing scales with six scales in angular series, the second row with four scales in angular series, the third row with three scale in straight line, the medial scale largest; femoral scales slightly enlarged (8R, 8L), distal scales more than twice the size of proximal scales, all smaller than those of pore-bearing precloacal scales, separated from precloacal scales by diastema; precloacal groove absent; fully everted hemipenes thick, 5.9 mm in length, two penes at each sheath, two sockets posterior to hemipenal bases (Fig.
Limbs rather slender; digits with strongly inflected interphalangeal joints; forelimbs bearing five relatively slender fingers, moderately bowed, ending with curved claws, ForeL/SVL 0.162; expanded proximal subdigital lamellae on fourth finger 6, unmodified distal subdigital lamellae on fourth finger 12, total subdigital lamellae on fourth finger 18; hind limbs bearing five relatively slender toes, strongly bent, ending with curved claws, CrusL/SVL 0.176; expanded proximal subdigital lamellae on fourth toe 7, unmodified distal subdigital lamellae on fourth toe 15, total subdigital lamellae on fourth toe 22; all digits lacking scansorial setae on ventral surface; scales on limbs small, interspersed with larger, low, conical, weakly keeled tubercles; scales on palmar and plantar surfaces small.
Tail moderately wide anteriorly, TaW 5.5 mm, segmented, cylindrical, becoming slender toward tip, regenerated posteriorly; dorsal caudal longitudinal tubercle rows at base of tail 8; 2 transverse rows of dorsal caudal tubercles at posterior margin of third band on tail, 22.7 mm from tail base; vertebral caudal surface with scattered bump at approximate intervals of 3 mm; subcaudal scale rows smooth, small, differing in size and irregular in shape, usually alternating between a single slightly enlarged and two smaller scales, 2 or 3 times larger than neighboring lateral caudal scales (Fig.
(Fig.
Morphometric and meristic characters of the type series are presented in Table
Mensural, meristic and color pattern characters of Cyrtodactylus phnomchiensis sp. nov. Abbreviations defined in the text. All specimens have regenerated portions of tails (*).
Voucher specimen |
|
|
|
|
|
Range |
---|---|---|---|---|---|---|
Type status | Holotype | Paratype | Paratype | Paratype | Paratype | |
Sex | Male | Male | Female | Female | Female | |
SVL | 76.1 | 79.0 | 77.3 | 80.7 | 76.7 | 76.1–80.7 |
TaL* | 75.1 | 56.9 | 66.6 | 79.1 | 64.6 | 56.9–79.1 |
TaW | 5.5 | 4.4 | 5.3 | 5.6 | 5.7 | 4.4–5.7 |
TaW/SVL | 0.072 | 0.056 | 0.069 | 0.069 | 0.074 | 0.056–0.74 |
ForeL | 12.3 | 13.2 | 12.1 | 13.7 | 11.7 | 11.7–13.7 |
ForeL/SVL | 0.162 | 0.167 | 0.157 | 0.170 | 0.153 | 0.153–0.170 |
CrusL | 13.4 | 14.9 | 13.8 | 16.1 | 13.2 | 14.2–16.6 |
Crus/SVL | 0.176 | 0.189 | 0.179 | 0.200 | 0.172 | 0.172–0.200 |
AG | 36.6 | 36.1 | 35.3 | 36.4 | 35.4 | 35.3–36.6 |
AG/SVL | 0.481 | 0.457 | 0.457 | 0.451 | 0.462 | 0.451–0.481 |
HeadL | 22.1 | 23.5 | 22.2 | 23.6 | 23.4 | 22.1–23.6 |
HeadL/SVL | 0.290 | 0.297 | 0.287 | 0.292 | 0.305 | 0.287–0.305 |
HeadW | 14.1 | 14.5 | 13.7 | 15.2 | 13.7 | 13.7–15.2 |
HeadD | 9.4 | 9.2 | 8.6 | 9.8 | 8.6 | 8.6–9.8 |
EyeD (eye diameter) | 5.1 | 5.1 | 4.8 | 4.8 | 4.5 | 4.5–5.1 |
EyeD/SVL | 0.067 | 0.065 | 0.062 | 0.059 | 0.059 | 0.059–0.067 |
Ear-EyeD (eye-ear distance) | 5.7 | 6.1 | 6.0 | 6.5 | 5.7 | 5.7–6.5 |
ESD (eye-snout distance) | 9.0 | 9.5 | 9.0 | 9.9 | 9.3 | 9.0–9.9 |
ESD/SVL | 0.118 | 0.120 | 0.116 | 0.123 | 0.121 | 0.116–0.123 |
END (eye-nostrial distance) | 6.6 | 6.9 | 6.4 | 7.0 | 7.0 | 6.4–7.0 |
IO (interorbital distance) | 5.5 | 4.8 | 5.0 | 5.7 | 5.2 | 4.8–5.7 |
IND (internarial distance) | 2.9 | 2.8 | 2.6 | 2.9 | 2.7 | 2.6–2.9 |
EarDV (vertical) | 1.2 | 1.3 | 1.2 | 1.3 | 1.3 | 1.2–1.3 |
EarDH (horizontal) | 0.7 | 1.1 | 0.7 | 0.7 | 0.7 | 0.6–1.1 |
Intersupranasal scales | 1 | 1 | 1 | 1 | 1 | 1 |
Supralabials (SL) | 12R/13L | 11R/11L | 11R/11L | 12R/12L | 12R/13L | 11–13 |
Infralabials (InL) | 9R/9L | 10R/9L | 10R/10L | 10R/9L | 10R/10L | 8–10 |
PVT | 32 | 31 | 36 | 34 | 32 | 31–36 |
LDRT | 18 | 20 | 20 | 20 | 19 | 18–20 |
VS | 47 | 47 | 52 | 45 | 54 | 45–54 |
Median subcaudal scales slightly enlarged | yes | yes | yes | yes | yes | yes |
SDLF4 | 18 | 20 | 18 | 19 | 19 | 18–20 |
SDLT4 | 22 | 23 | 20 | 21 | 21 | 20–23 |
EFS | 8R8L | 3R/3L | 0 | 7R/6L | 0 | 0–8 |
FP | 0 | 0 | 0 | 0 | 0 | 0 |
EPrecS | 7 | 9 | 7 | 10 | 9 | 7–10 |
PrecP | 5 | 4 | 4 | 1 | 7 | 1–7 |
PrecG | 0 | 0 | 0 | 0 | 0 | 0 |
PostPSR | 3 | 4 | 3 | 3 | 3 | 3–4 |
PostSP | 4 | 3 | 4 | 4 | 3 | 3–4 |
Number of dark brown body bands | 3 | 3 | 4 | 3 | 3 | 3–4 |
Femoral and precloacal scales continuous | no | no | no | no | no | no |
Yellowish spots on top of head | yes | yes | yes | yes | yes | yes |
Posterior border of nuchal loop pointed | yes | yes | yes | yes | yes | yes |
First body band complete | yes | yes | yes | yes | yes | yes |
Second to fourth body bands more irregular | yes | yes | yes | yes | yes | yes |
Yellowish white or light brown bands about half the width of dark brown body bands | yes | yes | yes | yes | yes | yes |
Number of yellowish white or light brown body bands | 2 | 2 | 3 | 2 | 2 | 2–3 |
Yellowish spot above ear opening | yes | yes | yes | yes | yes | yes |
Enlarged tubercle at corner of jaw | yes | Yes | yes | yes | yes | yes |
The new species is known only from the type locality at Phnom Chi in Prey Lang Wildlife Sanctuary, Kampong Thom Province, Sandan District, Cambodia. All individuals were found at night between 2001–2147 hr in evergreen-large dipterocarp dominated forest associated with rocky terrain (Fig.
Cyrtodactylus phnomchiensis sp. nov. is distinguishable from all 19 other members of the C. irregularis group by a unique combination of morphological characters (and in mitochondrial DNA; Fig.
Cyrtodactylus phnomchiensis sp. nov. differs from C. bidoupimontis Nazarov, Poyarkov, Orlov, Phung, Nguyen, Hoang & Ziegler, 2012 by having ventral scales 45–54 (vs. 38–43 in bidoupimontis), precloacal pits in females 1–7 (vs. absent in bidoupimontis), dark brown body bands larger than yellowish white or light brown dorsal bands (vs. dark brown bands, when present, narrower than light yellow dorsal bands in bidoupimontis), and distinct large yellow band on anterior margin of dark brown nuchal loop (vs. narrow light margin in bidoupimontis), and yellow spots on top of head (vs. dark spots in bidoupimontis).
Cyrtodactylus phnomchiensis sp. nov. differs from C. buchardi by having SVL 76.1–80.7 mm (vs. 60–65 mm in buchardi), SDLF4 18–20 (vs. 14 in buchardi), SDLT4 20–23 (vs. 12–14 in buchardi), ventral scales 45–54 (vs. 30 in buchardi), LDRT 18–20 (vs. 25 in buchardi), precloacal pores in males 4–5 (vs. 9 in buchardi), and irregular dorsal body bands (vs. blotches in buchardi).
Cyrtodactylus phnomchiensis sp. nov. differs from C. bugiamapensis Nazarov, Poyarkov, Orlov, Phung, Nguyen, Hoang & Ziegler, 2012 by having LDRT 18–20 (vs. 20–24 in bugiamapensis), ventral scales 45–54 (vs. 36–46 in bugiamapensis), precloacal pores in males 4–5 (vs. 7–11 in bugiamapensis), SDLF4 18–20 (vs. 15–17 in bugiamapensis), SDLT4 20–23 (vs. 17–20 in bugiamapensis), CrusL/SVL in adult specimens 0.172–0.200 (vs. 0.144–0.157 in bugiamapensis), large nuchal loop bordered anteriorly and posteriorly by broad yellow bands (vs. narrow nuchal loop bordered by distinct narrow white lines in bugiamapensis), dark brown body bands 3–4 (vs. seven highly irregular dark blotches with light margins in bugiamapensis), and top of head with yellowish spots (vs. distinct dark brown spots in bugiamapensis).
Cyrtodactylus phnomchiensis sp. nov. differs from C. caovansungi Orlov, Nguyen, Nazarov, Ananjeva & Nguyen, 2007 by having SVL 76.1–80.7 mm (vs. 90.4–94.0 mm in caovansungi), ventral scales 45–54 (vs. 38–44 in caovansungi), femoral pores absent (vs. 6 in caovansungi), precloacal pores in males 4–5 (vs. 9 in caovansungi), SDLF4 18–20 (vs. 22 in caovansungi), and enlarged subcaudals absent (vs. present in caovansungi).
Cyrtodactylus phnomchiensis sp. nov. differs from C. cattienensis Geissler, Nazarov, Orlov, Böhme, Phung, Nguyen & Ziegler, 2009 by having SVL 76.1–80.7 mm (vs. 69.0 mm maximum in cattienensis), ventral scales 45–54 (vs. 28–42 in cattienensis), SDLF4 18–20 (vs. 12–16 in cattienensis), and SDLT4 20–23 (vs. 14–19 in cattienensis).
Cyrtodactylus phnomchiensis sp. nov. differs from C. cucdongensis Schneider, Phung, Le, Nguyen & Ziegler, 2014 by having SVL 76.1–80.7 mm (vs. 55.8–65.9 mm in cucdongensis), ventral scales 45–54 (vs. 35–44 in cucdongensis), SDLF4 18–20 (vs. 13–18 in cucdongensis), and SDLT4 20–23 (vs. 15–20 in cucdongensis).
Cyrtodactylus phnomchiensis sp. nov. differs from C. cryptus Heidrich, Rösler, Vu, Böhme & Ziegler, 2007 by having precloacal pores in males 4–5 (vs. 9–11 in cryptus), and precloacal pits in females 1–7 (vs. absent in cryptus).
Cyrtodactylus phnomchiensis sp. nov. differs from C. dati Ngo, 2013 by having SVL 76.1–80.7 mm (vs. 70.1 mm maximum in dati), regenerated TaL 56.9–79.1 mm vs. (vs. 50.3 mm maximum, non-regenerated TaL in dati), femoral pores in both sexes absent (vs. present in dati), nuchal loop continuous (vs. broken in dati), and dark brown body bands (vs. irregular dark brown blotches on body in dati).
Cyrtodactylus phnomchiensis sp. nov. differs from C. gialaiensis Luu, Dung, Nguyen, Le & Ziegler, 2017 by having SVL 76.1–80.7 mm (vs. 62.8 mm maximum in gialaiensis), precloacal pores in males 4–5 (vs. 9–10 in gialaiensis), SDLF4 18–20 (vs. 14–15 in gialaiensis), and SDLT4 20–23 (vs. 15–17 in gialaiensis).
Cyrtodactylus phnomchiensis sp. nov. differs from C. huynhi Ngo & Bauer, 2008 by having SDLF4 18–20 (vs. 14–17 in huynhi), AGL/SVL 0.451–0.481 (vs. 0.370–0.428 in huynhi), ventral scales 45–54 (vs. 43–46 in huynhi), precloacal pores in males 4–5 (vs. 7–9 in huynhi), dark brown body bands 3–4 (vs. 5–6 in huynhi); femoral pores in both sexes absent (vs. 3–8 in huynhi), and nuchal loop bordered anteriorly and posteriorly by broad yellow bands (vs. narrow cream margin in huynhi).
Cyrtodactylus phnomchiensis sp. nov. differs from C. irregularis by lacking enlarged triangular tubercles at base of tail (vs. present in irregularis), CrusL/SVL 0.172–0.200 (vs. 0.138–0.156 in irregularis), LDRT 18–20 (vs. 22–24 in irregularis), paravertebral tubercles 31–36 (vs. 38–48 in irregularis); ventral scales 45–54 (vs. 38–45 in irregularis), SDLF4 18–20 (vs. 15–16 in irregularis), SDLT4 20–23 (vs. 18–20 in irregularis), dark brown body bands 3–4 (vs. 5–7, mostly as irregular blotches in irregularis), and yellowish spots on top of head (vs. distinct dark brown spots in irregularis).
Cyrtodactylus phnomchiensis sp. nov. differs from C. kingsadai Ziegler, Phung, Le & Nguyen, 2013 by having SVL 76.1–80.7 mm (vs. 83.0–94.0 mm in kingsadai), enlarged femoral scales 0–8 (vs. 9–12 in kingsadai), precloacal pore in males 4–5 (vs. 7–9 in kingsadai), and subcaudal scales not enlarged (vs. enlarged in kingsadai).
Cyrtodactylus phnomchiensis sp. nov. differs from C. phuocbinhensis Nguyen, Le, Tran, Orlov, Lathrop, MacCulloch, Le, Jin, Nguyen, Nguyen, Hoang, Che, Murphy & Zhang, 2013 by having SVL 76.1–80.7 mm (vs. 46.0–60.4 in phuocbinhensis); precloacal pits in females 1–7 (vs. absent in phuocbinhensis), top of head with yellow spots (vs. dark brown spots in phuocbinhensis), and dark brown body bands (vs. two dark brown longitudinal stripes or blotches in phuocbinhensis).
Cyrtodactylus phnomchiensis sp. nov. differs from C. pseudoquadrivirgatus Rösler, Vu, Nguyen, Ngo & Ziegler, 2008 by having yellow spots on top of head (vs. dark blotches on top of head in pseudoquadrivirgatus) and dark brown body bands (vs. highly irregular body blotches in pseudoquadrivirgatus).
Cyrtodactylus phnomchiensis sp. nov. differs from C. sangi Pauwels, Nazarov, Bobrov & Poyarkov, 2018 by having SVL 76.1–80.7 mm (vs. 56.3 mm maximum in sangi), paravertebral tubercles 31–36 (vs. 27–29 in sangi), ventral scales 45–54 (vs. 37 in sangi), precloacal pores in males 4–5 (vs. 7 in sangi), and first dark brown body band complete, second, third, and fourth more irregular (vs. highly irregular bands in sangi).
Cyrtodactylus phnomchiensis sp. nov. differs from C. takouensis Ngo & Bauer, 2008 by having LDRT 18–20 (vs. 9–10 smooth tubercles in takouensis); ventral scales 45–54 (vs. 39–40 in takouensis), SDLF4 18–20 (vs. 16–17 in takouensis), SDLT4 20–23 (vs. 18–20 in takouensis), and dark brown canthal stripe absent (vs. present in takouensis).
Cyrtodactylus phnomchiensis sp. nov. differs from C. taynguyenensis Nguyen, Le, Tran, Orlov, Lathrop, MacCulloch, Le, Jin, Nguyen, Nguyen, Hoang, Che, Murphy & Zhang, 2013 by having supralabials 11–13 (vs. 8–9 in taynguyenensis), precloacal pits in females present (vs. absent in taynguyenensis), SDLF 18–20 (vs. 13–18 in taynguyenensis), top of head with yellow spots (vs. dark brown blotches in taynguyenensis), and dark brown body bands (vs. black irregular blotches margined by light brown in taynguyenensis).
Cyrtodactylus phnomchiensis sp. nov. differs from C. yangbayensis Ngo & Chan, 2010 by having SDLT4 20–23 (vs. 15–17 in yangbayensis) and lacking enlarged subcaudal scales (vs. present in yangbayensis).
Cyrtodactylus phnomchiensis sp. nov. is most closely related in mitochondrial DNA to C. ziegleri Nazarov, Orlov, Nguyen & Ho, 2008 (Fig.
Mitochondrial DNA serves as a useful but imperfect tool for delimiting species within the C. irregularis complex (
Unfortunately, our phylogenetic analysis of the COI gene does not resolve the relationships among C. phnomchiensis sp. nov. and the two subclades of C. ziegleri (subclades Z1 and Z2; Fig.
Phnom Chi consists of an isolated small rocky mountain (peak of 652 m elevation) and a few associated smaller hills, altogether encompassing an area of approximately 4,464 ha within the Prey Lang Wildlife Sanctuary in Kampong Thom and Kratie provinces, Cambodia. The base and lower elevations of Phnom Chi have dry and mixed deciduous forest, whereas upper elevations have large dipterocarp-dominated evergreen and semi-evergreen forest. The current habitat remains in relatively good condition, but this long-overlooked site needs urgent conservation attention. Local communities utilize Phnom Chi for resource extraction, notably the tapping of liquid resin from large dipterocarp trees on the mountain, and small-scale, illegal gold extraction around the base, in addition to forest burning during the dry season (possibly by resin tappers). A small pagoda at the base of the mountain and the scenic beauty of the area (Fig.
Species diversity of Cyrtodactylus in Cambodia is likely to be significantly underestimated. Recently, five species were described within the C. intermedius complex from the Cardamom Mountains of southwestern Cambodia (
Wild Earth Allies is grateful to the General Department of Administration for Nature Conservation and Protection, Ministry of Environment of Cambodia, for providing permission to conduct research in Prey Lang Wildlife Sanctuary. Rangers of the Environmental Department of Kampong Thom Province, En E (Research Officer, Wild Earth Allies), and community members living at the foot of Phnom Chi assisted with field work. Field work was implemented with funding support from Wild Earth Allies. Sereivathana Tuy (Cambodia Director, Wild Earth Allies) provided technical support to the project. Hannah E. Som assisted with sequencing DNA and preparing figures. Thomas Ziegler and two anonymous reviewers improved the manuscript.