A new insular species of the Cyrtodactyluspulchellus group (Reptilia, Gekkonidae) from Tarutao Island, southern Thailand revealed by morphological and genetic evidence

Abstract The bent-toed geckos of the Cyrtodactyluspulchellus group are widely distributed along the Thai-Malay Peninsula. Although taxonomic and phylogenetic studies of this species group have been continuously conducted, only some populations from Thailand have been included, resulting in hidden diversity within this group. In this study, we used morphological and molecular data to clarify the taxonomic status and describe a new population from Tarutao Island, Satun Province, southern Thailand. Cyrtodactylusstellatussp. nov. can be distinguished from its congeners by the combination of the following morphological characters: body size; tuberculation; number of dark body bands, ventral scales, and femoroprecloacal pores in males; presence of precloacal pores in females; and scattered pattern on dorsum. Phylogenetic analyses of the mitochondrial ND2 gene recovered the new species as the sister species to C.astrum, with an uncorrected pairwise divergence of 9.78–12.37%. Cyrtodactylusstellatussp. nov. is currently only known from Tarutao Island, Thailand. The discovery of this species suggests that the diversity within the C.pulchellus group remains underestimated and future exploration of unsurveyed areas are needed to further the understanding of this group and its geographic range.


Introduction
Bent-toed geckos in the genus Cyrtodactylus Gray, 1827 are geographically widespread and inhabit lowland (e.g., peat swamps, karst formations, and limestone forests) to mountainous regions (> 1,500 m a.s.l) of South Asia to Melanesia, ranging from India, Myanmar, Thailand, Vietnam, Cambodia, Malaysia, Java, Papua New Guinea to northern Australia Nielsen and Oliver 2017;Pauwels et al. 2018;Purkayastha et al. 2020;Riyanto et al. 2020;Grismer et al. 2020aGrismer et al. , 2021aGrismer et al. , 2021b. This genus is the most diverse group of gekkotans, comprising 314 nominal species (Uetz et al. 2021). During the last two decades, the number of new species described in this genus has significantly increased with the exploration of unsurveyed karst formations (Luu et al. 2016;Nazarov et al. 2018;Davis et al. 2019;Grismer et al. 2018Grismer et al. , 2020b. Moreover, genetic data has become a useful tool for taxonomic studies, revealing hidden diversity within the genus (Murdoch et al. 2019;Chomdej et al. 2020;Neang et al. 2020;Riyanto et al. 2020;Kamei and Mahony 2021;Liu and Rao 2021). Recent molecular studies have further supported the monophyly of this genus based on the most complete phylogenetic analysis to date, and have recognized 31 species groups (Grismer et al. 2021b).
One clade of particular interest is the Cyrtodactylus pulchellus group. This relatively diverse group is distributed along the Thai-Malay Peninsula and has high morphological and molecular variation. Cyrtodactylus pulchellus Gray, 1827 was thought to be a single wide-ranging species across their distributional range, but following an integrative approach many new species have been described (e.g., C. bintangrendah , C. langkawiensis Grismer et al., 2012, and C. sharkari Grismer et al., 2014. This species group has been recovered as monophyletic and currently contains 16 recognized species, based on multiple phylogenetic studies (Grismer et al. , 2014(Grismer et al. , 2016Quah et al. 2019;Wood et al. 2020;Termprayoon et al. 2021). This group is distributed from the south of the Isthmus of Kra, southern Thailand to southern Peninsular Malaysia and some of its offshore islands (Grismer and Ahmad 2008;Sumontha et al. 2012;Grismer et al. 2012Grismer et al. , 2014Grismer et al. , 2016Quah et al. 2019;Wood et al. 2020;Termprayoon et al. 2021). During field surveys, specimens of the C. pulchellus group were collected from Tarutao Island, Satun Province, southern Thailand. Initially, these specimens were recognized as an insular population of C. astrum Grismer et al., 2012 due to their superficial resemblance in coloration pattern and dorsal tuberculation. A re-examination of these specimens showed morphological differences from its other congeners and mitochondrial DNA sequence data revealed corroborative evidence that the new population of Cyrtodactylus from Tarutao Island represents a distinct monophyletic lineage and is the sister species to C. astrum from the adjacent mainland. Based on integrative analyses, we considered this new Cyrtodactylus population from Tarutao Island as distinct and described it as a new species below.

Sampling
Field surveys were conducted on Tarutao Island, Mueang Satun District, Satun Province, southern Thailand from November 2017 to November 2019 (Fig. 1). Specimens of the C. pulchellus group were collected from karst forest at night (1900-2200 h). Ecological data (air temperature and relative humidity) were recorded using a Kestrel 4000 Weather Meter, and habitat use of each specimen was noted. Geographical coordinates and elevation were recorded using a Garmin GPSMAP 64s. For molecular studies, liver tissue was taken from each euthanized specimen, individually preserved in 95% ethyl alcohol, and stored at -20 °C. Specimens were initially fixed in 10% formalin and later transferred into 70% ethyl alcohol for permanent storage. Voucher specimens were deposited in the herpetological collections of the Zoological Museum, Kasetsart University, Thailand (ZMKU). Additional preserved specimens were examined in the holdings of the Thailand Natural History Museum (THNHM), Thailand, and the La Sierra University Herpetological Collection (LSUHC), La Sierra University, Riverside, California, USA.

DNA extraction and PCR amplification
Total genomic DNA was extracted from ethanol-preserved liver tissue of five Cyrtodactylus specimens from Tarutao Island (Table 1) using a NucleoSpin Tissue Kit (Macherey-Nagel GmbH & Co. KG, Germany). A fragment of mitochondrial NADH dehydrogenase subunit 2 (ND2) gene and its flanking tRNAs was amplified using a double-stand Polymerase Chain Reaction (PCR) under the following conditions: initial denaturation at 94 °C for 4 min, followed by 35 cycles of denaturation at 94 °C for 30 sec, annealing at 48-52 °C for 30 sec, extension at 72 °C for 1 min 30 sec, and final extension at 72 °C for 7 min using the primers Metf6 (5' AAGCTTTCGGGCCCA-TACC 3'; Macey et al. 1997), and COIH (5' AGRGTGCCAATGTCTTTGTGRTT 3'; Macey et al. 1997). PCR products were purified using NucleoSpin Gel and PCR Clean-Up kit (Macherey-Nagel GmbH & Co. KG, Germany). Purified products were sequenced for both strands using the same amplifying primers on an ABI 3730XL DNA Sequencer (Applied Biosystems, CA, USA). Sequences were visually checked and aligned in Geneious R11 (Biomatters, Ltd, Auckland, New Zealand). The proteincoding region of ND2 was translated to amino acids and checked to confirm the lack of premature stop codons. All sequences were deposited in GenBank under the accession numbers OK094494-OK094503 (Table 1).

Phylogenetic analyses
Phylogenetic trees were reconstructed using two different methods, Maximum Likelihood (ML) and Bayesian Inference (BI). The best substitution model for each partition was determined using the Bayesian Information Criterion (BIC) under the greedy search algorithm as implemented in PartitionFinder2 on XSEDE (Lanfear et al. 2016). The selected models for ML and BI were TIM+G for 1 st and 2 nd codon positions of ND2, TVM+I+G for 3 rd codon position of ND2 and TRN+I+G for tRNAs. The ML analysis was performed in IQ-TREE web server v1.6.12 (Trifinopoulos et al. 2016) with 1,000 bootstrap replicates using ultrafast bootstrap approximation (Minh et al. 2013). The BI analysis was performed in MrBayes 3.2.6 on XSEDE (Ronquist et al. 2012) using the CIPRES Science Gateway v3.3 (Miller et al. 2010). Two simultaneous runs were performed with four chains per run, three hot and one cold under the default settings. The analysis was run for 10,000,000 generations and sampled every 1,000 generations from the Markov chain Monte Carlo (MCMC), with the first 25% of each run discarded as burn-in. Stationarity and the effective sample sizes (ESS) for all parameters were assessed in Tracer v1.7.1. (Rambaut et al. 2018). Nodes with ultrafast bootstrap support (UFB) of ≥ 95 and Bayesian posterior probabilities (BPP) of ≥ 0.95 were considered to be strongly supported (Huelsenbeck and Ronquist 2001;Wilcox et al. 2002;Minh et al. 2013). Intraspecific and interspecific uncorrected pairwise genetic divergences (p-distance) were calculated in MEGA X 10.0.5 using the pairwise deletion option for the treatment of gaps and missing data in the dataset (Kumar et al. 2018).

Morphology
The morphological characters and their definition used in this study were modified from previous studies of the C. pulchellus group (Grismer and Ahmad 2008;Grismer et al. 2012Grismer et al. , 2014Grismer et al. , 2016Quah et al. 2019;Wood et al. 2020), and abbreviations are derived from Grismer et al. (2018Grismer et al. ( , 2020c. All mensural characters were taken with digital calipers to the nearest 0.01 mm on the left side, while scale counts were made on both sides when possible. Scalation and external morphology were evaluated under a Nikon SMZ745 dissecting microscope. Measurement and meristic characters are shown in Table 2, and external morphological characters evaluated are described below. External morphological characters examined in the C. pulchellus group were the degree of body tuberculation, weak tuberculation referring to dorsal body tubercles that are low and rounded whereas prominent tuberculation refer to tubercles that are raise and keeled; the presence or absence of tubercles on the dorsal and ventral surface of the forearms; the presence or absence of tubercles in the gular region, throat, and ventrolateral body folds; the width of the dark body bands relative to the width of the interspace between the bands; the presence or absence of dark pigmentation infused in the white caudal bands of adults; the presence or absence of a precloacal depression or groove; the presence or absence of scattered white/yellow tubercles on the dorsum; and the presence or absence of white tail tip on the posterior portion of the original tail in hatchlings and juveniles. Color pattern characteristics were taken from digital images of live specimens in both sexes and of all possible age classes prior to preservation.

Statistical analyses
All analyses were performed using the base statistical software in R v3.6.1 (R Core Team 2019). To eliminate bias of sexual dimorphism, adult males and females were analyzed separately. Morphological analyses were run on 15 mensural characters. Tail length (TL) was not included due to their different condition (e.g., original, regenerated, and broken). All measurements of each species were size-adjusted in order to remove potential effects of allometry using the following allometric equation: X adj = log[X ± β(SVL ± SVL mean )], where X adj = adjusted value; X = measured value; β = unstandardized regression coefficient for each OTU; SVL = measured snout-vent length; SVL mean Table 2. Measurement and meristic characters used in this study, with abbreviations and explanations.

Abbreviations
Characters Measurement SVL Snout-vent length, taken from the tip of snout to the vent TW Tail width, taken at the base of the tail immediately posterior to the postcloacal swelling TL Tail length, taken from vent to the tip of the tail, original or regenerated FL Forearm length, taken from the posterior margin of the elbow while flexed 90º to the inflection of the flexed wrist TBL Tibia length, taken from the posterior surface of the knee while flexed 90º to the base of the heel AG Axilla to groin length, taken from the posterior margin of the forelimb at its insertion point on the body to the anterior margin of the hind limb at its insertion point on the body HL Head length, the distance from the posterior margin of the retroarticular process of the lower jaw to the tip of the snout HW Head width, measured at the angle of the jaws HD Head depth, the maximum height of head from the occiput to the throat ED Eye diameter, the greatest horizontal diameter of the eyeball EE Eye to ear distance, measured from the anterior edge of the ear opening to the posterior edge of the eyeball ES Eye to snout distance, measured from anterior most margin of the eyeball to the tip of snout EN Eye to nostril distance, measured from the anterior margin of the eyeball to the posterior margin of the external nares IO Inter orbital distance, measured between the anterior edges of the orbit EL Ear length, the greatest vertical distance of the ear opening The number of subdigital lamellae beneath the fourth toe, counted from the base of the first phalanx to the claw FPP The total number of precloacal and femoral pores in male (i.e., the sum of the number of femoral and precloacal scales bearing pores combined as a single meristic referred to as the femoroprecloacal pores) PP The number of precloacal pores in female BB The number of dark body bands between limb insertions DCB The number of dark caudal bands on the original tail = overall average SVL of each OTU (Thorpe 1975(Thorpe , 1983Turan 1999;Lleonart et al. 2000)-implemented through the R package GroupStruct (Chan and Grismer 2021).
Morphological measurements of C. astrum, C. dayangbuntingensis Quah et al., 2019, C. langkawiensis andC. lekaguli Grismer et al., 2012 were obtained from their original descriptions Quah et al. 2019). Additional preserved specimens of C. astrum (from Malaysia) and C. lekaguli (topotypes) were examined and included in the analyses (Appendix I). Morphometric adjustments were conducted separately on each species and then concatenated into a single data frame to ensure there was no interspecific conflation of variation (Reist 1985;McCoy et al. 2006). Specimens were assigned into five groups (= species) based on phylogenetic analyses which are Cyrtodactylus Tarutao Island samples (N = 5 males, 5 females), C. astrum (N = 5 males, 3 females), C. dayangbuntingensis (N = 2 males), C. langkawiensis (N = 2 males, 4 females), and C. lekaguli (N = 7 males, 9 females). Principal components analysis (PCA) was performed on size-adjusted data for each sex using FactoMineR package (Lê et al. 2008) and were visualized with the R package ggplot2 (Wickham 2016). For univariate analysis, Shapiro-Wilk test was used to evaluate data to meet normality assumptions (p ≥ 0.05) and Levene's test for testing for equality of variance (p ≥ 0.05). Morphological differences were compared using Analysis of variance (ANOVA) or Kruskal-Wallis test. ANOVA was conducted on normally distributed data with homogeneous variances and were subjected to Tukey HSD post hoc tests (Tukey's test) to determine which characters had statistically different mean values for which pairs of species if ANOVA had a p-value of less than 0.05. Kruskal-Wallis test was performed on non-normally distributed data and followed by a post hoc Dunn's multiple comparison (Dunn's test). Due to limited sample sizes, C. dayangbuntingensis (N = 2 males) and males of C. langkawiensis (N = 2 males) were excluded from the univariate analysis.

Phylogenetic relationships
The aligned matrix contained 1,429 mtDNA characters from 93 individuals of the C. pulchellus group and nine individuals of outgroup species (Table 1). The standard deviation of split frequencies among the two simultaneous BI runs was 0.002676, and the ESS values of all parameters were greater than or equal to 2,494.4. The maximum likelihood value of the best ML tree was lnL = -15,115.412.

Morphology
The PCA was conducted on members from Clade A. The plots on the first two PC axes showed that the Tarutao Island specimens are clustered separately from other species in both sexes (Fig. 4). In male, the first two principal components explained 66.95% of the morphological variation (Table 4). The first principal component (PC1) accounted for 45.88% of the variation and was heavily loaded on FL adj , TBL adj , HW adj , HD adj , EEadj , ES adj , and EN adj ; and the PC2 accounted for 21.07% of the variation and was heavily loaded on TW adj , AG adj , IO adj , and EL adj . PC analysis of females accounted for 56.74% of the variation in first two components. The PC1 accounted for 34.81% of the variation and was heavily loaded on TBL adj , HW adj , EE adj , ES adj , and EN adj ; and the PC2 accounted for 21.93% of the variation and was heavily loaded on TW adj , IO adj and IN adj .
The univariate analyses (ANOVA or Kruskal-Wallis test) were significantly different (p < 0.05) in most morphological characters among the members of Clade A (except C. dayangbuntingensis). In the comparison of adult males, the Tarutao Island population was significantly different from C. astrum and C. lekaguli in twelve morphological characters (ANOVA or Kruskal-Wallis test, p < 0.001-0.006) except AG adj , ED adj and IN adj (ANOVA or Kruskal-Wallis test, p = 0.051-0.122). Subsequent Tukey's test or Dunn's test demonstrated that Tarutao Island population was significantly different from C. astrum in SVL adj , FL adj , TBL adj , HL adj , HW adj , HD adj , EE adj , ES adj , and EN adj ; and C. lekaguli in SVL adj , TW adj , FL adj , HL adj , HW adj , HD adj , EE adj , ES adj , IO adj , and EL adj . In adult females, the Tarutao Island population was significantly different    Table 5. Additional differences in meristic characters and coloration are discussed in the comparison sections.
Description of holotype. Adult male SVL 94.2 mm; head large, moderate in length (HL/SVL 0.29) and wide (HW/HL 0.61), somewhat flattened (HD/HL 0.38), distinct from neck, and triangular in dorsal profile; lores concave anteriorly, inflated posteriorly; frontal and prefrontal regions deeply concave; canthus rostralis rounded anteriorly; snout elongate (ES/HL 0.39), rounded in dorsal profile, laterally constricted; eye large (ED/HL 0.25); ear opening elliptical, moderate in size (EL/HL 0.09), obliquely oriented; eye to ear distance slightly greater than diameter of eye; rostral rectangular, divided dorsally by an inverted Y-shaped furrow, bordered posteriorly by left and right supranasals and internasal, bordered laterally by first supralabials; external nares bordered anteriorly by rostral, dorsally by a large anterior supranasal, posteriorly by two postnasals, ventrally by first supralabial; 13/14 (left/right) rectangular supralabials extending to just beyond upturn of labial margin, tapering abruptly below midpoint of eye; second supralabial slightly larger than first; 11/11 infralabials tapering in size posteriorly; scales of rostrum and lores slightly raised, larger than granular scales on top of head and occiput, those on posterior portion of canthus rostralis slightly larger; scales on occiput intermixed with small tubercles; large, boney frontal ridges bordering orbit confluent with boney, V-shaped, transverse, parietal ridge; dorsal superciliaries elongate, smooth, largest anteriorly; mental triangular, bordered laterally by first infralabials and posteriorly by left and right, trapezoidal postmentals which contact medially for 50% of their length; one row of slightly enlarged, elongate sublabials extending posteriorly to the seventh (left) and fifth (right) infralabials; small, granular, gular scales grading posteriorly into larger, flat, smooth, imbricate, pectoral and ventral scales.
Body relatively short (AG/SVL 0.46) with well-defined, non-tuberculate, ventrolateral folds; dorsal scales small, granular, interspersed with low, regularly arranged, weakly keeled tubercles, smaller intervening tubercles occasionally present; tubercles extend from occiput to caudal constriction, absent from regenerated portion of tail; tubercles on occiput and nape relatively small, those on body largest; approximately 21 longitudinal rows of tubercles at midbody; 36 paravertebral tubercles; 33 flat imbricate ventral scales between ventrolateral body folds; ventral scales larger than dorsal scales; precloacal scales large, smooth; deep precloacal groove.
Forelimbs moderate in stature, relatively short (FL/SVL 0.16); scales on dorsal surfaces of forelimbs granular intermixed with larger tubercles; scales of ventral surface of forearm flat, subimbricate, tubercles absent; palmar scales small, weakly rounded; digits well-developed, inflected at basal, interphalangeal joints; subdigital lamellae rec-  tangular proximal to joint inflection, only slightly expanded distal to inflection; digits narrower distal to joints; claws well-developed, sheathed by a dorsal and ventral scale; the fifth digit broken on left forearm; hind limbs more robust than forelimbs, moderate in length (TBL/SVL 0.19), larger tubercles on dorsal surface of legs separated by smaller juxtaposed scales; ventral scales of thigh flat, smooth, imbricate, larger than dorsal granular scales; ventral, tibial scales flat, smooth, imbricate; a single row of 34 enlarged femoroprecloacal scales extend nearly from knee to knee through precloacal region where they are continuous with enlarged, pore-bearing precloacal scales; 27 separated pore-bearing femoroprecloacal scales (Fig. 10A), forming an inverted T bearing a deep, precloacal groove; six pore-bearing scales bordering groove (three on each side of groove); postfemoral scales immediately posterior to enlarged scale row small, nearly granular, forming an abrupt union with postfemoral scales on posteroventral margin of thigh; plantar scales weakly rounded to flat; digits well developed, inflected at basal, interphalangeal joints; subdigital lamellae proximal to joint inflection rectangular, only slightly expanded distal to inflection; digits narrower distal to joints; claws well-developed, sheathed by a dorsal and ventral scale; 21/22 subdigital lamellae on the 4 th toe.
Tail 94.8 mm in length, completely regenerated, 9.2 mm in width at base, tapering to a point; regenerated tail covered with small, smooth, rectangular scales dorsally; base of tail bearing hemipenial swellings; one row of 4/4 medium-sized postcloacal tubercles on each hemipenial swelling; postcloacal scales smooth, flat, large, imbricate.
Coloration in life (Fig. 5). Dorsal ground color of head, body, and limbs lightbrownish grey; a wide, dark-brown nuchal band bordered anteriorly and posteriorly by thin, creamy-white lines bearing tubercles that extend from the posterior margin of one eye to the posterior margin of other eye; the color of nuchal band and creamy-white lines is faded above left ear opening; four dark-brown body bands between nuchal loop and hind limb insertions that are also bordered anteriorly and posteriorly by thin, creamy-white lines bearing tubercles, first band terminates at shoulders, second and third bands terminate just dorsal of ventrolateral folds, the fourth band terminates at  femurs; dark body bands slightly larger than light-colored interspaces; creamy-white to light-yellow tubercles scattered on dorsal surfaces of nape, body, and limbs; one additional dark-brown band posterior to hind limbs; light-brown regenerated tail, bearing yellowish pigment on some scales; ventral surfaces of head smudged with brown; abdomen and limbs beige, with slightly darker, lateral regions.
Coloration in preservative (Figs 6, 7). The overall color pattern of head, body, limbs, and tail similar to that in life with some fading. Ground color of head, body, limbs, and dorsum light-brown; dark body bands lighter than in life. Colored tuberculation on dorsum fade to off-white. Tan colored on the ventral surface.
Variation. Cyrtodactylus stellatus sp. nov. usually varies in coloration and banding pattern Tables 6,7). All specimens possess a clear dark-brown nuchal band which is less clearly defined in ZMKU R 00903 and the holotype. In adult females, precloacal pores are present but they lack the precloacal groove (Fig. 10B). Four specimens (ZMKU R 00903, ZMKU R 00907, ZMKU R 00911, and ZMKU R 00913) have prominent light-yellow tubercles scattered on the dorsum and limbs. Male paratype (ZMKU R 00907) has continuous pore-bearing femoroprecloacal scales. Original tails (ZMKU R 00899, ZMKU R 00901-00902, ZMKU R 00910, ZMKU R 00912, and ZMKU R 00916) have 10-12 dark caudal bands (Fig. 11A, B), light bands diffused with dark pigment in adults (immaculate in immature and juvenile), subcaudal scales transversely enlarged and shallow caudal furrows. Male paratypes have a single row of 3-4L/2-4R postcloacal tubercles on each hemipenial swelling except ZMKU R 00907 which has two rows of 4/5 on each side. This character in female paratypes is very small, and a single row of 2-4/2-4 on each side at the base of tail.
In life, the juvenile (ZMKU R 00917; SVL 43.1 mm) had a body pattern similar to the adults but with less prominent tuberculation, brownish yellow ground color of body, dark body bands are bordered by yellow lines, some bearing tubercles, the original tail has approximately 10 dark caudal bands, the posterior portion of tail is white, and light caudal bands are immaculate (Fig. 11C).  Distribution. Cyrtodactylus stellatus sp. nov. is currently known only from Tarutao Island, Satun Province, Thailand (Figs 1, 12A).
Natural history. All specimens of C. stellatus sp. nov. were collected from a karst forest at night (1950-2100 h) with temperatures between 27.1-32.2 °C and relative humidity between 71.4-93.0%. The specimens were found on karst walls, within karst crevices and on nearby karst boulders. Some specimens occurred on tree trunks or    Table 8. Diagnostic characters of Cyrtodactylus stellatus sp. nov. and its related species within the C. pulchellus group. W = weak; P = prominent; / = data unavailable.
Some information was collected from the following literature (Grismer et al. , 2014(Grismer et al. , 2016Quah et al. 2019, Wood et al. 2020, and Termprayoon et al. 2021    Tubercles on ventral surface of forelimb groove in males; 11-15 precloacal pores in females; scattered pattern of white, cream or light-yellow tubercles on dorsum; 10-12 dark caudal bands on original tail; white caudal bands on original tail infused with dark pigmentation in adults; and juveniles with white tail tip. Additional comparisons between C. stellatus sp. nov. and other species in the C. pulchellus group are in Table 8. Based on phylogenetic tree, C. stellatus sp. nov. is embedded in Clade A along with C. astrum, C. dayangbuntingensis, C. langkawiensis, and C. lekaguli. It can be distinguished from all four species by having smaller maximum SVL of 96.1 mm (vs. 108.3 mm in C. astrum, 99.0 mm in C. dayangbuntingensis, 99.8 mm in C. langkawiensis, and 108.3 in C. lekaguli); 24-29 femoroprecloacal pores in males (vs. 28-38 in C. astrum, 30 in C. langkawiensis, and 30-40 in C. lekaguli); 11-15 precloacal pores in females (vs. absent in C. astrum, C. dayangbuntingensis, C. langkawiensis, and C. lekaguli); scattered pattern of white, cream or light-yellow tubercles on dorsum (vs. absent in C. langkawiensis, and C. lekaguli); the ratio of dark body bands to the light color interspaces 0.92-1.68 (vs. 0.75 in C. dayangbuntingensis); 10-12 dark caudal bands (vs. 13 or 14 in C. astrum).

Discussion
The discovery of C. stellatus sp. nov. brings the total number of species in the C. pulchellus group to 17, of which four have been reported from Thailand. This new species is only known from karst habitats on Tarutao Island and seems to have a narrow geographic distribution (endemic to Tarutao Island). Molecular analyses recovered it as the sister taxon to C. astrum and is closely related to C. dayangbuntingensis, C. langkawiensis, and C. lekaguli. Although C. stellatus sp. nov. showed a similar morphological pattern to its sister species, morphological analyses and comparisons of meristic characters revealed that this new species is clearly different from its congeners species of Cyrtodactylus. Among Cyrtodactylus, most useful diagnostic characters are associated with the femoral and precloacal pores (Harvey et al. 2015). These characters are easily detected in males, but those in females are superficial and only found in some species (e.g., C. marmoratus Gray, 1831;C. psarops Harvey et al., 2015;C. sworderi Smith, 1925). We found differences in pore-bearing scales between C. stellatus sp. nov. and other species in the C. pulchellus group, that proved to be useful in distinguishing among species. Members of the C. pulchellus group mostly possess a continuous series of enlarged, pore-bearing femoroprecloacal scales in males, but C. stellatus sp. nov. presents a discontinuous row of femoroprecloacal pores except one individual (ZMKU R 00907), which has a continuous series. Moreover, the presence of precloacal pores were found in females of C. stellatus sp. nov., which has not been reported in the other species Quah et al. 2019;Wood et al. 2020, Termprayoon et al. 2021).
The discovery of this new species suggests that undiscovered species of the C. pulchellus group may still occur in southern Thailand where there are still numerous unexplored karst areas. Additional surveys are needed to determine the extent of the geographic range of C. stellatus sp. nov. and the C. pulchellus group in as a whole in the region.