﻿Integrative taxonomy delimits and diagnoses cryptic arboreal species of the Cyrtodactylusbrevipalmatus group (Squamata, Gekkonidae) with descriptions of four new species from Thailand

﻿Abstract Species delimitation and species diagnosis must remain separate operations to avoid constructing taxonomies comprised of non-monophyletic species based on morphological similarity as opposed to phylogenetic propinquity. This is particularly true for highly specialized species such as the range-restricted upland taxa in the Cyrtodactylusbrevipalmatus group of Indochina where strong selection pressure for an arboreal lifestyle has contributed to morphologically similar but distantly related species. This in turn, has resulted in a history of erroneous taxonomies that have actually obscured rather than revealed the diversity within this group. A Bayesian phylogeny of the C.brevipalmatus group recovered at least 15 putative species-level lineages, at least seven of which are undescribed, and of which four are described herein. A total evidence morphological data set comprised of 16 normalized morphometric, 15 meristic, and seven categorical characters scored across 51 individuals was subjected to a multiple factor analysis (MFA) and an analysis of variance (ANOVA) to diagnose the putative species. These new species descriptions contribute to focusing attention to the unrealized diversity in upland tropical ecosystems, which due to climate change, are some of the most impearled ecosystems on the planet. Thus, it is paramount that taxonomies do not conflate species identities and underrepresent true diversity.


Introduction
A cornerstone of biodiversity conservation is a phylogenetic-based taxonomy where the names of the component species are consistent with the patterns and processes by which they evolved. Taxonomies constructed from paraphyletic or polyphyletic species misrepresent history, thus obscuring true diversity and potentially countermanding the effectiveness of conservation efforts. Before a newly discovered population can be given a new name, or the name of an existing species be successfully challenged, it should be properly delimited and diagnosed. Delimiting and diagnosing species are independent operations used together to construct taxonomies that reflect, and are consistent with evolutionary history. Unfortunately, these two operations are often conflated when analyses to diagnose species, which are most often rooted in morphological similarity, are equated with analyses to delimit species, which are rooted in phylogenetic propinquity. The unfortunate consequence of this is that taxonomies may be constructed using non-monophyletic species, thus obscuring rather than revealing the group's actual diversity. This is especially true for taxonomies comprised of highly specialized cryptic species where, in the absence of a phylogeny, morphological convergence can be mistaken for common ancestry (see Grismer 2008;Grismer et al. 2020Grismer et al. , 2022. Despite methodological improvements over the years that have increased the efficacy of both delimitation and diagnostic procedures, they have had little effect on operational decisions to not conflate them. Convergent morphology results from the specific resource requirements that necessitate a particular functional morphology (Ricklefs 2012;Dehling et al. 2016;Baulechner et al. 2020). In such cases, phylogenetic analysis becomes paramount to disentangle convergence from common ancestry in order to construct proper phylogenetic taxonomies. This is especially true of clades containing highly specialized species such as the Cyrtodactylus brevipalmatus group (sec. Grismer et al. 2021a, b, c). This group currently contains four nominal species that range from northern Vietnam and Laos, southward through Thailand to southern Peninsular Malaysia (Fig. 1). Nearly all members of this adaptive radiation are highly specialized for an arboreal lifestyle and all have a prehensile tail carried in a tightly coiled position, a cryptic color pattern of different shades of brown that closely match the substrate of their microhabitat, and generally slow, deliberate "chameleon-like" movements. Presumably, the selection pressures for such a specialized lifestyle have contributed to morphological convergence within this relatively small group . Thus, as demonstrated by Grismer (2008) and Grismer et al. (2021cGrismer et al. ( , 2022, previous morphology-based operations intending to delimit species resulted in reoccurring taxonomies composed of non-monophyletic species (Smith 1935;Welch et al. 1990;Ulber 1993;Manthey and Grossmann 1997;Stuart 1999;Nabhitabhata et al. 2004;Nabhitabhata and Chan-ard 2005;Pauwels and Chan-ard 2006;Ellis and Pauwels 2012) that underrepresented the group's diversity. Grismer et al. (2021cGrismer et al. ( , 2022 were the first to employ phylogenetic analyses to the brevipalmatus group to delimit species and statistically based morphological analyses to diagnose those species. As such, by treating these analyses as independent operations and their results as corroborating evidence, a revised taxonomy consistent with the group's evolutionary history was disentangled from a convergence-based taxonomy.  Table 1. The phylogenetic-based taxonomy of Grismer et al. (2022) is expanded here with significantly increased morphological and genetic sampling from new localities in Thailand at Ko (= island) Chang (ZMKU R 00945) and Khlong Nakha Wildlife Sanctuary (THNHM 01667), Ranong Province; Hua Hin, Prachuap Khiri Khan Province (ZMKU R 00946-47); Si Sawat (ZMKU R 00958-64) and Thong Pha Phum (ZMKU R 00950-57), Kanchanaburi Province; and an isolated population from a small hilly region at Lan Sak, Uthai Thani Province (ZMKU R 00949) in the Chao Phraya River Basin. In this analysis, we discuss the high degree of morphological similarity throughout the phylogeny despite the relatively high degrees of genetic divergence. The analyses also recovered the existence of 4-10 new populations we hypothesize to be distinct species, four of which are described herein.

Species delimitation
The general lineage concept (GLC: de Queiroz 2007) adopted herein proposes that a species constitutes a population of organisms evolving independently from other such populations owing to a lack of, or limited gene flow. By "independently," it is meant that new mutations arising in one species cannot spread readily into another species (Barraclough et al. 2003;de Queiroz 2007). Under the GLC implemented herein, molecular phylogenies recovered monophyletic mitochondrial lineages of individuals (populations) that were used to develop initial species-level hypotheses, the grouping stage of Hillis (2019). Discrete color pattern data and univariate and multivariate analyses of morphological data were then used to search for characters and morphospatial patterns consistent with the tree-designated species-level hypotheses, the construction of boundaries representing the hypothesis-testing step of Hillis (2019), -thus providing independent diagnoses to complement the molecular analyses. In this way, delimiting (phylogeny) and diagnosing (taxonomy) species are not conflated (Frost and Hillis 1990;Frost and Kluge 1994;Hillis 2019). Species boundaries were cross-checked using a Bayesian Poisson Tree Process for species delimitation (bPTP; Zhang et al. 2013), thus providing an independent framework to complement the empirically based thresholds of the morphological and molecular analyses. This method delimits species from single-locus gene trees with low population samples (Lin et al. 2018) by detecting genetic clustering beyond the expected levels of a null hypothesis which infers that all individuals of a population form a genetically, interacting nexus. A Markov Chain Monte Carlo (MCMC) was run for 10,000 generations on the bPTP web server at https://species.h-its.org/ ptp/ and checked for convergence. Delimitation accuracy was based on the normalized mutual information criterion (NMI) (Vinh et al. 2010). The model relies on the prediction that independent evolution leads to the appearance of distinct genetic clusters, separated by relatively longer internal branches (Barraclough et al. 2003;Acinas et al. 2004). Such groups therefore diverge into discrete clades of genetic variation. These clades are then diagnosed by statistically defensible morphological analyses.

Genetic data
Methods for DNA extraction, sequencing, and editing followed Grismer et al. (2022) and resulted in a 1,386 base pair segment of the mitochondrial NADH dehydrogenase subunit 2 gene (ND2) and adjacent tRNAs. All material examined is listed in Table 1 along with GenBank accession numbers for the new and published genetic materials.

Morphological data
Morphological data included both meristic and morphometric characters. To reduce the degree of researcher bias, data were taken using the protocol of Le et al. (2021) and Grismer et al. (2022) and where possible, double checked by LLG using high resolution digital photographs and/or the actual specimens. All data were taken on the left side of the body (when possible) and measured to the nearest 0.1 mm using digital calipers under a Nikon SMZ745 stereomicroscope. Morphometric data taken were: snout-vent length (SVL), taken from the tip of the snout to the vent; tail length (TL), taken from the vent to the tip of the tail-original or partially regenerated; tail width (TW), taken at the base of the tail immediately posterior to the postcloacal swelling; humeral length (HumL), taken from the proximal end of the humerus at its insertion point in the glenoid fossa to the distal margin of the elbow while flexed 90°; forearm length (ForL), taken on the ventral surface from the posterior margin of the elbow while flexed 90° to the inflection of the flexed wrist; femur length (FemL), taken from the proximal end of the femur at its insertion point in the acetabulum to the distal margin of the knee while flexed 90°; tibia length (TibL), taken on the ventral surface from the posterior margin of the knee while flexed 90° to the base of the heel; axilla to groin length (AG), 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; head length (HL), the distance from the posterior margin of the retroarticular process of the lower jaw to the tip of the snout; head width (HW), measured at the angle of the jaws; head depth (HD), the maximum height of head measured from the occiput to base of the lower jaw posterior to the eyes; eye diameter (ED), the greatest horizontal diameter of the eye-ball; eye to ear distance (EE), measured from the anterior edge of the ear opening to the posterior edge of the bony orbit; eye to snout distance or snout length (ES), measured from anteriormost margin of the bony orbit to the tip of snout; eye to nostril distance (EN), measured from the anterior margin of the bony orbit to the posterior margin of the external nares; interorbital distance (IO), measured between the dorsomedial-most edges of the bony orbits; internarial distance (IN), measured between the external nares across the rostrum; and ear length (EL), greatest oblique length across the auditory meatus. Meristic characters evaluated were the number of supralabial scales (SL), counted from the largest scale at the corner of the mouth or posterior to the eye, to the rostral scale; infralabial scales (IL), counted from termination of enlarged scales at the corner of the mouth to the mental scale; number of paravertebral tubercles (PVT) between the limb insertions counted in a straight line immediately left of the vertebral column; number of longitudinal rows of body tubercles (LRT) counted transversely across the body midway between the limb insertions from one ventrolateral body fold to the other; number of longitudinal rows of ventral scales (VS) counted transversely across the abdomen midway between limb insertions from one ventrolateral fold to the other; number of transverse rows of ventral scales (VSM) counted along the midline of the body from the postmentals to just anterior to the cloacal opening, stopping where the scales become granular; number of expanded subdigital lamellae on the fourth toe proximal to the digital inflection (TL4E) counted from the base of the first phalanx where it contacts the body of the foot to the largest scale on the digital inflection-the large contiguous scales on the palmar and plantar surfaces were not counted; number of small, generally unmodified subdigital lamellae distal to the digital inflection on the fourth toe (TL4U) counted from the digital inflection to the claw including the claw sheath; total number of subdigital lamellae (TL4T) beneath the fourth toe (i.e. TL4E + TL4U = TL4T); number of expanded subdigital lamellae on the fourth finger proximal to the digital inflection (FL4E) counted the same way as with TL4E; number of small generally unmodified subdigital lamellae distal to the digital inflection on the fourth finger (FL4U) counted the same way as with TL4U; total number of subdigital lamellae (FL4T) beneath the fourth toe (i.e. FL4E + FL4U = FL4T); total number of enlarged femoral scales (FS) from each thigh combined as a single metric; number of enlarged precloacal scales (PCS); number of precloacal pores (PP) in males; the number of femoral pores (FP) in males from each thigh combined as a single metric; and the number of dark body bands (BB) between the dark band on the nape and the hind limb insertions on the body. A post-sacral or sacral band when present, was not counted. Categorical characters evaluated were the presence or absence of tubercles on the flanks (FKT); single enlarged, unmodified, medial subcaudal scales (SC2); enlarged medial subcaudals intermittent, medially furrowed, posteriorly emarginated (SC3); slightly enlarged medial subcaudals (SC1); large or small dorsolateral caudal tubercles (DCT) forming a narrow or wide ventrolateral caudal fringe (VLF1); ventrolateral caudal fringe scales generally homogenous or not (VLF2); and the cross-section of the tail round or square (TLcross).

Phylogenetic analyses
An input file implemented in BEAUti (Bayesian Evolutionary Analysis Utility) v.  (Rambaut et al. 2018) to ensure effective sample sizes (ESS) were clearly above 200 for all parameters. A maximum clade credibility tree using mean heights at the nodes was generated using TreeAnnotator v. 1.8.0 (Rambaut and Drummond 2013) with a burn-in of 1,000 trees (10%). Nodes with Bayesian posterior probabilities (BPP) of 0.95 and above were considered strongly supported (Huelsenbeck et al. 2001;Wilcox et al. 2002). Uncorrected pairwise sequence divergences were calculated in MEGA 11 (Tamura et al. 2021) using the complete deletion option to remove gaps and missing data from the alignment prior to analysis.

Statistical analyses
All statistical analyses were conducted using R Core Team (2018). Morphometric characters used in statistical analyses were SVL, AG, HumL, ForL, FemL, TibL, HL, HW, HD, ED, EE, ES, EN, IO, EL, and IN. Tail metrics were not used due to the high degree of incomplete sampling (i.e., regenerated, broken, or missing). In order to most successfully remove the effects of allometry (sec. Chan and Grismer 2022), size was normalized using the following equation: X adj = log(X)-β[log(SVL)-log(SVL mean )], where X adj = adjusted value; X = measured value; β = unstandardized regression coefficient for each population; and SVL mean = overall average SVL of all populations (Thorpe 1975(Thorpe , 1983Turan 1999;Lleonart et al. 2000), accessible in the R package GroupStruct (available at https://github. com/chankinonn/GroupStruct). The morphometrics of each species were normalized separately and then concatenated so as not to conflate potential intra-with interspecific variation (Reist 1986;McCoy et al. 2006). The juvenile Cyrtodactylus ngati (HNUE-R00112) was removed from the data so as not to skew the normalization results. All data were scaled to their standard deviation to ensure they were analyzed on the basis of correlation and not covariance. Meristic characters analyzed were SL, IL, PVT, LRT, VS, VSM, TL4E, TL4U, TL4T, FL4E, FL4U, FL4T, FS, PCS, and BB. Precloacal and femoral pores were omitted from the multivariate analyses due to their absence in females. Categorical characters analyzed were DCT, VLF1, VLF2, TLcross, SC1, SC2, and SC3. Small sample sizes (N = 1 or 2) for some of the species/populations precluded them from statistical analyses. A Levene's test for the normalized morphometric and meristic characters was conducted to test for equal variances across all groups. Characters with equal variances were analyzed with an analysis of variance (ANOVA) and TukeyHSD post hoc test for mean comparisons involving more than three groups. Those with unequal variances were subjected to Welch's F-test and Games-Howell post hoc test to test for mean comparisons involving more than three groups.
Morphospatial clustering and positioning among the species/populations was analyzed using multiple factor analysis (MFA) on a concatenated data set comprised of 15 meristic characters, 16 normalized morphometric characters, and seven categorical characters (Suppl. material 1). For this test, it was not necessary to remove populations represented by small sample sizes. The MFA was implemented using the mfa() command in the R package FactorMineR (Husson et al. 2017) and visualized using the Factoextra package (Kassambara and Mundt 2017). MFA is a global, unsupervised, multivariate analysis that incorporates qualitative and quantitative data (Pagès 2015) simultaneously, making it possible to analyze different data types in a nearly total evidence environment. In an MFA, each individual is described by a different set of variables (i.e., characters) which are structured into different data groups in a global data frame, in this case quantitative data (i.e., meristics and normalized morphometrics) and categorical data (i.e., scale, tubercle, and caudal morphology). In the first phase of the analysis, separate multivariate analyses are carried out for each set of variables: principal component analyses (PCA) for each quantitative data set and a multiple correspondence analysis (MCA) for the categorical data. The data sets are then normalized separately by dividing all their elements by the square root of their first eigenvalue. For the second phase of the analysis, these normalized data sets are concatenated into a single matrix for a final global PCA of the normalized data. Standardizing the data in this manner prevents one data type from overleveraging another. In other words, the normalization of the data in the first phase prevents data types with the greatest number of characters or the greatest amount of variation from outweighing other data types in the second phase. This way, the contributions of each data type to the overall variation in the data set are scaled to define the morphospatial distance between individuals as well as calculating each data type's contribution to the overall variation in the analysis (Pagès 2015;Kassambara and Mundt 2017).
A non-parametric permutation multivariate analysis of variance (PERMANOVA) from the vegan package 2.5-3 in R (Oksanen et al. 2020) was used to determine if the centroid locations and group clustering of each species/population in the MFA were statistically different from one another (Skalski et al. 2018). The analysis was based on the calculation of a Gower (dis)similarity matrix using 50,000 permutations based on the loadings of the first four dimensions of the MFA. A pairwise post hoc test calculates the differences between all combinations of population pairs, generating a p-value, a Bonferroni-adjusted p-value, and a pseudo-F ratio (F statistic). A p < 0.05 is considered significant and larger F statistics indicate more pronounced group separation. A rejection of the null hypothesis (i.e., centroid positions and/or the spread of the data points [i.e. clusters] are no different from random) signifies a statistically significant difference between species/populations.

Phylomorphospace
The BEAST phylogeny was projected onto the first two dimensions of the MFA plot using the phylomorphospace() command from the R package phytools (Revell 2012). This allows one to map the history of a group's morphological diversification and infer the magnitude and direction of shape change along branches of the phylogeny (Sidlauskas 2008). Outgroups and species from the phylogenetic analyses not represented in the morphological dataset were excluded. To eliminate a potential bias caused by the highly derived Cyrtodactylus elok (see Grismer et al. 2022), it was not included.
Some of the populations examined had only genetic or morphological data. However, only phylogenetically delimited populations bearing morphological differences from other populations were described as new species. In some cases, populations represented by only morphological data but in close geographic proximity to named species from which they could not be separated morphologically, were considered conferre (cf.) to the named species pending further investigation.

Phylogenetic analysis
The BEAST analysis recovered 8-10 new species and two major clades within the brevipalmatus group: a weakly supported clade (BPP = 0.82) from the southernmost portion of the Thai-Malay Peninsula south of the Isthmus of Kra comprised of Cyrtodactylus elok, C. brevipalmatus, and C. sp.14 and a well-supported clade (1.00) comprised of all other species north of the Isthmus of Kra (Figs 1, 2). Within the latter, a well-supported (1.00) more exclusive clade containing the sister species from Hua Hin, Prachuap Khiri Khan (Cyrtodactylus rivularis sp. nov. [see below]) and C. rukhadeva with the Ko Chang population (Cyrtodactylus kochangensis sp. nov. [see below]) being sister to these. All relationships within this clade have strong nodal support (1.00) and its taxa occur only in the northernmost portion of the Thai-Malay Peninsula (Fig. 1). The population from Si Sawat (Cyrtodactylus fluvicavus sp. nov. [see below]), C. sp.10 from Chao Doi Waterfall, Mae Moei National Park, Tak Province, and C. sp.9 from Thong Pha Phum are sequentially related to a strongly supported (1.00) clade containing the remaining species. Within that latter clade, C. interdigitalis is the strongly supported (1.00) sister species to C. sp.11 from Phu Hin Rong Kla National Park and together they compose the well-supported (0.93) sister lineage to the Uthai Thani population (Cyrtodactylus uthaiensis sp. nov. [see below]). The remaining species all form a strongly supported (1.00) sequentially related clades with C. ngati. Uncorrected pairwise sequence divergences within the brevipalmatus group range from 2.84-22.84% (Table 2). Ranges for those of the new species described below are as follows: Cyrtodactylus rivularis sp. nov. ranges from 3.61-21.81%, being most similar to C. rukhadeva and most distant to C. elok; Cyrtodactylus fluvicavus sp. nov. ranges from 9.55-20.26%, being most similar to the Thong Pha Phum population (C. sp.9) and most distant to C. elok; Cyrtodactylus kochangensis sp. nov. ranges from 12.00-20.90%, being most similar to Cyrtodactylus rivularis sp. nov. and most distant to C. elok; and Cyrtodactylus uthaiensis sp. nov. ranges from 5.81-21.16%, being most similar to C. cf. ngati1 and most distant to C. elok.

Bayesian Poisson Tree Process (bPTP)
The bPTP species delimitation analysis recovered 16 putative species within the brevipalmatus group with varying degrees of support (Table 3). The newly acquired material in this study recovered as distinct species are Cyrtodactylus fluvicavus sp. nov., Cyrtodactylus kochangensis sp. nov., Cyrtodactylus rivularis sp. nov., the holotype of C. rukhadeva (ZMMU R-16851), C. rukhadeva (ZMKU R 00948), Cyrtodactylus uthaiensis sp. nov., C. sp.9 from Thong Pha Phum, and C. sp.14 from Pulau Langkawi. However, Cyrtodactylus rivularis sp. nov. and C. sp.9 were not recovered with strong support (NMI 0.575 and 0.565, respectively) despite them being genetically distinct (Table 2) and morphologically diagnosable (Tables 4-6). The analysis also separated the two specimens of C. rukhadeva from the same locality which only differ by a genetic distance of 1.55% and are similar in morphology. Sukumaran and Knowles (2017) demonstrated that many species delimitation analyses recover clades not species and that a wider range of empirical data are necessary to interpret these boundaries (Coyne and Orr 1998;Feulner et al. 2007;Fontaneto et al. 2007;Knowles and Carstens 2007;Leach et al. 2009) as was done here (see below).

Statistical analyses
The ANOVA and TukeyHSD post hoc analyses of the adjusted morphometric and meristic characters were consistent with the phylogenetic and pairwise distance data in recovering a number of statistically significant differences between the Cyrtodactylus fluvicavus sp. nov. and C. interdigitalis as well as others (Table 6). Cyrtodactylus fluvicavus sp. nov. plotted separately in the MFA, only slightly overlapping with C. brevipalmatus (Fig. 3A). Cyrtodactylus kochangensis sp. nov., Cyrtodactylus rivularis sp. nov. and Cyrtodactylus uthaiensis sp. nov. plotted separately from each other and all other populations. A PERMANOVA analysis recovered several instances of various combinations of statistically significant differences among all the populations/species numbering more than two samples in regards to their clustering and centroid placement ( Table 7). The contributions of each data type are shown in Fig. 3C.  Figure 3. A MFA of the species-level lineages based on the BEAST phylogeny ( Fig. 2) B phylomorphospace projection of the BEAST phylogeny C percent contribution of each data type to Dim 1-4 of the MFA. The percentage score at the top of each graph is the percent contribution of that dimension to the overall variation in the data set. Morphological and phylogenetic discordance The phylomorphospace analysis illustrates that the morphological similarities among the species/populations in the MFA are discordant with their phylogenetic placement in the tree (Fig. 3A, B). In other words, species/populations that are similar in morphology are not necessarily each other's closest relatives. For example, Cyrtodactylus interdigitalis is morphologically more similar to the distantly related C. rukhadeva than it is to C. sp.11 or Cyrtodactylus uthaiensis sp. nov. to which it is closely related. Similarly, C. sp.9 is morphologically similar to the distantly related Cyrtodactylus kochangensis sp. nov. than to the closely related clade containing Cyrtodactylus uthaiensis sp. nov., Other examples of morphological and phylogenetic discordance can be seen by comparing Fig. 3A, B.

Taxonomy
Given the phylogenetic delimitation of the Si Sawat, Prachuap Khiri Khan, Uthai Thani, and Ko Chang populations (Figs 2, 3), the statistically significant diagnostic placement of the Si Sawat population in morphospace ( Fig. 3A; Table 7) and its statistically significant diagnostic morphological differences (Table 6), we describe them below as new species. The phylogenetic delimitation and statistically significant morphological differences among C. sp.9 from Thong Pha Phum and C. sp.11 from Phu Hin Rong Kla National Park will require descriptions at a future date. Diagnosis. Cyrtodactylus fluvicavus sp. nov. can be separated from all other species of the brevipalmatus group by the combination of having 11-13 supralabials, 9 or 10 infralabials, 26-30 paravertebral tubercles, 14-18 rows of longitudinally arranged tubercles, 30-39 transverse rows of ventrals, 154-175 longitudinal rows of ventrals, 9-11 expanded subdigital lamellae on the fourth toe, 10-13 unexpanded subdigital lamellae on the fourth toe, 19-22 total subdigital lamellae on the fourth toe; 7-9 expanded subdigital lamellae on the fourth finger, 9-11 unexpanded subdigital lamellae on the fourth finger, 17-19 total subdigital lamellae on the fourth finger; 9-12 total enlarged femoral scales, 8-11 total femoral pores in males; 14 or 15 precloacal pores in males; 14 or 15 enlarged precloacals; enlarged femorals and enlarged precloacals not continuous; proximal femorals less than one-half the size of the distal femorals; small tubercles on forelimbs and flanks; small dorsolateral caudal tubercles and narrow ventrolateral caudal fringe; ventrolateral caudal fringe composed scales of different size; tail circular in cross-section; slightly enlarged unpaired medial subcaudals not posteromedially furrowed; maximum SVL 78.2 mm; three dark transverse body bands (Tables 4-6).

Cyrtodactylus fluvicavus
Description of holotype (Fig. 4). Adult male SVL 72.5 mm; head moderate in length (HL/SVL 0.28), width (HW/HL 0.70), depth (HD/HL 0.42), distinct from neck, triangular in dorsal profile; lores concave slightly anteriorly, weakly inflated posteriorly; prefrontal region concave; canthus rostralis rounded; snout elongate (ES/HL 0.42), rounded in dorsal profile; eye large (ED/HL 0.25); ear opening obliquely elongate, small; eye to ear distance greater than diameter of eye; rostral rectangular, divided by a deep furrow, bordered posteriorly by large left and right supranasals and one small azygous internasal, bordered laterally by first supralabials; external nares bordered anteriorly by rostral, dorsally by large supranasal, posteriorly by two smaller postnasals, bordered ventrally by first supralabial; 12R/12L rectangular supralabials, second through seventh supralabials nearly same size as first, then tapering abruptly below eye; 10R/10L infralabials tapering smoothly to just below and slightly past posterior margin of eye; scales of rostrum and lores flat to domed, larger than granular scales on top of head and occiput; scales of occiput intermixed with distinct, small tubercles; superciliaries subrectangular, largest dorsally; mental triangular, bordered laterally by first infralabials and posteriorly by large left and right trapezoidal postmentals contacting medially for 50% of their length posterior to mental; one row of slightly enlarged, elongate sublabials extending posteriorly to sixth(L) and fifth(R) infralabial; gular and throat scales small, granular, grading posteriorly into slightly larger, flatter, smooth, imbricate, pectoral and ventral scales.
Body relatively short (AG/SVL 0.46) with well-defined ventrolateral folds; dorsal scales small, granular interspersed with larger, conical, semi-regularly arranged, weakly keeled tubercles; tubercles extend from occipital region onto base of tail and slightly beyond as paravertebral rows; smaller tubercles extend anteriorly onto nape and occiput, diminishing in size anteriorly; approximately 17 longitudinal rows of tubercles at midbody; approximately 30 paravertebral tubercles; small tubercles on flanks; 34 longitudinal rows of flat, imbricate, ventral scales much larger than dorsal scales; 155 transverse rows of ventral scales; 15 large, pore-bearing, precloacal scales; no deep precloacal groove or depression; and two rows of post-precloacal scales on midline.
Forelimbs moderate in stature, relatively short (ForL/SVL 0.14); granular scales of forearm slightly larger than those on body, interspersed with large tubercles; palmar scales rounded, slightly raised; digits well-developed, relatively short, inflected at basal interphalangeal joints; digits narrower distal to inflections; subdigital lamellae wide, transversely expanded proximal to joint inflections, narrower transverse lamellae distal to joint inflections; claws well-developed, claw base sheathed by a dorsal and ventral scale; 8R/8L expanded and 10R/10L unexpanded lamellae beneath the fourth finger; hind limbs larger and thicker than forelimbs, moderate in length (TibL/SVL 0.16), covered dorsally by granular scales interspersed with moderately sized, conical tubercles dorsally and posteriorly and anteriorly by flat, slightly larger, subimbricate scales; ventral scales of thigh flat, subimbricate, larger than dorsals; subtibial scales flat, imbricate; one row of 5R/6L enlarged pore-bearing femoral scales not continuous with enlarged pore-bearing precloacal scales, terminating distally at knee; proximal femoral scales smaller than distal femorals, the former forming an abrupt union with much smaller, rounded, ventral scales of posteroventral margin of thigh; plantar scales flat; digits relatively long, well-developed, inflected at basal interphalangeal joints; 9R/9L wide, transversely expanded subdigital lamellae on fourth toe proximal to joint inflection extending onto sole, and 11R/11L unexpanded lamellae beneath the fourth toe; and claws well-developed, claw base sheathed by a dorsal and ventral scale.
Tail original, 97.6 mm long (TL/SVL 1.34), 5.2 mm in width at base, tapering to a point; sub-circular or nearly round in cross-section; dorsal scales flat, square bearing tubercles forming paravertebral rows and small tubercles forming a dorsolateral longitudinal row; slightly larger, posteriorly directed, semi-spinose tubercles forming narrow but distinct ventrolateral caudal fringe; larger scales of ventrolateral fringe occur at regular intervals; medial subcaudals slightly enlarged but not paired, distinctly enlarged single medial subcaudals absent; subcaudals, larger than dorsal caudals; base of tail bearing hemipenial swellings; 3R/2L conical postcloacal tubercles at base of hemipenial swellings; and postcloacal scales flat, imbricate.
Coloration in life (Fig. 5). Ground color of the head, body, limbs, and tail brown; faint, diffuse mottling on the top of the head; thin, dark brown postorbital stripe; ventral portion of lores and suborbital region dark brown; nuchal band faint, bearing two dark-colored posterior projections; paired dark brown paravertebral blotches on nape; three wide faint irregularly shaped body bands edged in dark brown between limb insertions; band interspaces bearing irregularly shaped dark-colored markings; darkcolored speckling on limbs and digits; digits bearing pale-colored bands; eight wide dark-colored caudal bands separated by seven pale-colored bands; first six dark-colored and seven pale-colored caudal bands encircle tail; all ventral surfaces beige, generally immaculate; iris orange-gold in color. Variation (Fig. 5). Individuals of the type series are very similar in overall coloration and pattern. TL and TW of complete original tails (ZMKU R 00959-00960, ZMKU R 00963-00964) are 79.7-97.6 mm (mean 91.4 ± 8.2 mm; N = 4) and 4.0-5.2 mm (mean 4.5 ± 0.5; N = 4), respectively. The posterior one-third of the tail of ZMKU R 00958 is regenerated (TL 72.0 mm, TW 4.9 mm); that of ZMKU R 00962 is missing (TL 83.8 mm, TW 4.4 mm); and that of ZMKU R 00961 is broken one-third of the way passed the base and nearly all of the broken section is regenerated (TL 73.6 mm, TW 4.07 mm). Regenerated tail sections lack a color pattern. ZMKU R 00963-64 have seven dark-colored and six pale-colored caudal bands as opposed to eight and seven bands, respectively, in the holotype. ZMKU R 00958, ZMKU R 00960, ZMKU R 00962, ZMKU R 00964 are slight less boldly marked than the holotype. Meristic and morphometric differences are listed in Table 5.
Distribution. Cyrtodactylus fluvicavus sp. nov. is currently known from the type locality at Tham Than Lot Noi-Tham Than Lot Yai Nature Trail in Chaloem Rattanakosin National Park, Si Sawat District, Kanchanaburi Province, western Thailand (Fig. 1).
Etymology. The specific epithet fluvicavus comes from the Latin fluvius, meaning stream, river, or flow and the Latin cavus, meaning hollow or hole and refers to a landmark cave in the Chaloem Rattanakosin National Park which has a stream that flows through it.
Comparisons. Cyrtodactylus fluvicavus sp. nov. is the sister species to a clade composed of ten lineages in a phylogenetic sequence of C. sp.9, C. sp.10, Cyrtodactylus uthaiensis sp. nov., C. sp.11, C. interdigitalis, C. cf. ngati1, C. cf. ngati2, C. ngati3, and the sister lineages C. ngati4 and C. ngati (Fig. 2). Cyrtodactylus fluvicavus sp. nov. differs from those lineages by an uncorrected pairwise sequence divergence of 9.55-12.13% and from all members of the brevipalmatus group by 11.61-20.26% (Table  2). It differs categorically from C. elok by having as opposed to lacking paravertebral tubercles and femoral pores, and by having 14-18 as opposed to 4-7 longitudinal rows of tubercles. It differs from C. brevipalmatus, C. interdigitalis, C. ngati, C. ngati3, C. rukhadeva, and C. sp.9 in having statistically significant different mean values of the morphometric characters of AG, HumL, ForL, FemL, TibL, HL, HW, HD, EE, ES, EN, EL and IN (Table 6). It differs further from C. brevipalmatus, C. interdigitalis, C. ngati, C. ngati3, C. rukhadeva, and C. sp.9 in having statistically significant different mean values of the meristic characters of SL, PVT, LRT, VS, VSM, TL4E, TL4T, FL4E, FL4U, FL4T, FS, PCS, and BB. Statistically significant and discrete differences between Cyrtodactylus fluvicavus sp. nov. and all other species and populations are presented in Tables 4-6.  Natural history. All individuals were found in karst forests bearing mixed deciduous and dry evergreen trees amidst rocky streams and a nearby waterfall (Fig. 6). This area is surrounded by agricultural fields and residential areas. Specimens (N = 7) were collected at night (1900-2100 h) during the dry season (April) on the tree trunks or palm tree leaves (57.1%; N = 4), twigs of shrubs (14.3%; N = 1), karst walls (14.3%; N = 1), and a wooden bridge (14.3%; N = 1) at 526 m elevation with a temperature of 31.9 °C and relative humidity of 56.9%. The holotype (ZMKU R 00959) and two specimens (ZMKU R 00960, ZMKU R 00962) were found on tree trunks ≤ 100 cm above ground level. One specimen (ZMKU R 00964) was found on a palm tree branch approximately 50 cm above the ground. Another specimen (ZMKU R 00963) was found on the twig of a shrub. Another specimen (ZMKU R 00961) was found on a karst wall approximately 3 m above the ground, and another (ZMKU R 00958) on a wooden bridge over a stream. Given these observations, this species appears to be an arboreal habitat generalist. The new species was found to co-occur with two other species of gekkonid lizards, Cyrtodactylus monilatus Yodthong, Rujirawan, Stuart, Grismer, Aksornneam, Termprayoon, Ampai & Aowphol, 2022 and Dixonius siamensis (Boulenger, 1899). can be separated from all other species of the brevipalmatus group by the combination of having 12 or 13 supralabials, 9-11 infralabials, 33 or 34 paravertebral tubercles, 18-20 rows of longitudinally arranged tubercles, 34-37 transverse rows of ventrals, 160-166 longitudinal rows of ventrals, nine expanded subdigital lamellae on the fourth toe, 12 or 13 unexpanded subdigital lamellae on the fourth toe, 21 or 22 total subdigital lamellae on the fourth toe; eight expanded subdigital lamellae on the fourth finger, 10-12 unexpanded subdigital lamellae on the fourth finger, 18-20 total subdigital lamellae on the fourth finger; 14-16 total enlarged femoral scales; 15 enlarged precloacals; enlarged femorals and enlarged precloacals not continuous, and lacking pores; proximal femorals less than one-half the size of the distal femorals; small tubercles on forelimbs and flanks; large dorsolateral caudal tubercles and wide ventrolateral caudal fringe; ventrolateral caudal fringe composed generally homogeneous scales; tail square in cross-section; single enlarged unpaired medial subcaudals not posteromedially furrowed; maximum SVL 73.9 mm; three or four dark transverse body bands (Tables 4, 5).
Description of holotype (Fig. 7). Adult female SVL 73.9 mm; head moderate in length (HL/SVL 0.27), width (HW/HL 0.73), depth (HD/HL 0.40), distinct from neck, triangular in dorsal profile; lores concave slightly anteriorly, weakly inflated pos- of eye; scales of rostrum and lores flat to domed, larger than granular scales on top of head and occiput; scales of occiput intermixed with distinct, small tubercles; superciliaries subrectangular, largest anteriorly; mental triangular, bordered laterally by first infralabials, posteriorly by large left and right elongate postmentals contacting medially for approximately 40% of their length posterior to mental; one row of two (R) and four (L) slightly enlarged sublabials extending posteriorly to third(L) and second(R) infralabial, subsequent sublabials much smaller; gular and throat scales small, granular, grading posteriorly into slightly larger, flatter, smooth, imbricate, pectoral and ventral scales. Body relatively short (AG/SVL 0.47) with well-defined ventrolateral folds; dorsal scales small, granular, interspersed with larger conical, semi-regularly arranged, weakly keeled tubercles; tubercles extend from occipital region onto base of tail and slightly beyond as paravertebral rows; tubercles of nape and occiput small; approximately 20 longitudinal rows of tubercles at midbody; approximately 34 paravertebral tubercles; tubercles on flanks nearly same size as those on dorsum; 34 longitudinal rows of flat, imbricate, ventral scales much larger than dorsal scales; 160 transverse rows of ventral scales; no pore-bearing, precloacal scales; 15 enlarged precloacal scales; no deep precloacal groove or depression; and three rows of post-precloacal scales on midline.
Forelimbs moderate in stature, relatively short (ForL/SVL 0.13); granular scales of forearm slightly larger than those on body, interspersed with tubercles; palmar scales rounded, slightly raised; digits well-developed, relatively short, inflected at basal interphalangeal joints; digits narrower distal to inflections; subdigital lamellae wide, transversely expanded proximal to joint inflections, narrower transverse lamellae distal to joint inflections; claws well-developed, claw base sheathed by a dorsal and ventral scale; 8R/8L expanded and 11R/11L unexpanded lamellae beneath the fourth finger; hind limbs larger and thicker than forelimbs, moderate in length (TibL/SVL 0.15), covered dorsally by granular scales interspersed with moderately sized, conical tubercles dorsally and posteriorly and anteriorly by flat, slightly larger, subimbricate scales; ventral scales of thigh flat, subimbricate, larger than dorsals; subtibial scales flat, imbricate; no pore-bearing femoral scales; 8R/8L enlarged femoral scales; enlarged femoral scales not contiguous with enlarged precloacal scales, terminating distally at knee; proximal femoral scales smaller than distal femoral scales, the former forming an abrupt union with much smaller, rounded, ventral scales of posteroventral margin of thigh; plantar scales flat; digits relatively long, well-developed, inflected at basal interphalangeal joints; 9R/9L wide, transversely expanded subdigital lamellae on fourth toe proximal to joint inflection that extend onto sole, and 13R/13L unexpanded lamellae beneath the fourth toe; and claws well-developed, sheathed by a dorsal and ventral scale at base.
Posterior one-half of tail regenerated, tail long 91.5 mm (TL/SVL 1.24), 4.8 mm in width at base, tapering to a point; nearly square in cross-section; dorsal scales flat, square bearing large tubercles forming a discontinuous dorsolateral longitudinal row; slightly larger, posteriorly directed, semi-spinose tubercles forming large distinct ventrolateral caudal fringe; scales of ventrolateral fringe generally homogeneous; single medial subcaudals enlarged but not paired; subcaudal scales, larger than dorsal caudal scales; base of tail bearing hemipenial swellings; 2R/2L conical postcloacal tubercles at base of hemipenial swellings; and postcloacal scales flat, imbricate.
Coloration in life (Fig. 8). Ground color of the head, body, limbs, and tail pale brown; faint, diffuse mottling on rostrum; lores darkly colored; wide, distinct, palecolored post-orbital stripe; nuchal band faint, bearing two posterior projections; three very faint, wide irregularly shaped body bands edged in slightly darker brown between limb insertions; band interspaces bearing irregularly shaped, faint, dark-colored markings; dark-colored speckling on limbs and digits; digits bearing pale-colored bands; four wide faint dark-colored caudal bands separated by three pale-colored bands on original portion of tail; all caudal bands encircle tail; all ventral surfaces beige, generally immaculate; iris orangish to coppery in color.
Variation (Fig. 8). The paratype (ZMKU R 00946) closely approximates the holotype in overall coloration and pattern except that it is more boldly marked. It has four dark-colored body bands as opposed to three and a complete original tail (TL 89.0mm, TW 4.1 mm) bearing eight dark-colored and seven pale-colored bands. The pale-colored postorbital stripe is slightly thinner and less distinct. Meristic and morphometric differences are listed in Table 5.
Distribution. Cyrtodactylus rivularis sp. nov. is currently known from the type locality at Pa La-U Waterfall, Kaeng Krachan National Park, Huai Sat Yai Subdistrict, Hua Hin District Prachuap Khiri Khan Province, Thailand.
Etymology. The specific epithet rivularis is derived from the Latin rivus, meaning stream, brook, or creek refers to rocky brook or stream habitat of the new species.
Natural history. The holotype and paratype were collected at night (1900-2055 h) on granite boulders by a rocky stream dry evergreen forest at 368 m in elevation ( Fig. 9) with a temperature of 25.3 °C and relative humidity of 86.9%. The new species was found to co-occur with two other species of gekkonid lizards, Cyrtodactylus oldhami (Theobald, 1876) and Gehyra mutilata (Wiegmann, 1834).
Comments. Cyrtodactylus rivularis sp. nov. and C. cf. rukhadeva occur on separate branches of the phylogeny, thus neither is embedded within one another's branch indicating there is no evidence of gene flow between them. Both species are reported to occur within the boundaries of Kaeng Krachan National Park . Kaeng Krachan National Park is the largest national park in Thailand, encompassing 2,914.7 km 2 across five districts through the provinces of Phetchaburi and Prachuap Khiri Khan and extending approximately 300 km in latitude. As such, it harbors a wide range of habitats from lowland forests near sea level to cloud forests at its highest peak of 1,513 m in elevation. The seven specimens of C. cf. rukhadeva (THNHM 01807, 24622, 24838, 03251-54) were obtained by various collectors between 11 March 1991 and 25 October 2015 with no specific collection data other than Phetchaburi Province. The locality reported by Grismer et al. (2021c) was an estimate obtained from Ulber (1993) from the central portion of the park in Phetchaburi Province who referred to a specimen he examined (THNHM 24838) as C. brevipalmatus. Cyrtodactylus rivularis sp. nov. however, is known only from the Park's southernmost limit in Prachuap Khiri Khan Province. Additional collecting and DNA sequencing will be required to establish the approximate distribution limits of each population.

R-16492
Sex    Sex  Sex cf. ngati1 C.

cf. ngati2
Institutional catalog number

Cyrtodactylus rivularis
sp. nov.   Sex  Diagnosis (based on the holotype). Cyrtodactylus kochangensis sp. nov. can be separated from all other species of the brevipalmatus group by the combination of having 12 or 13 supralabials, nine infralabials, 34 paravertebral tubercles, 14 rows of longitudinally arranged tubercles, 35 transverse rows of ventrals, 172 longitudinal rows of ventrals, 8 or 9 expanded subdigital lamellae on the fourth toe, 11 or 12 unexpanded subdigital lamellae on the fourth toe, 19-21 total subdigital lamellae on the fourth toe; eight expanded subdigital lamellae on the fourth finger, ten unexpanded subdigital lamellae on the fourth finger, 18 total subdigital lamellae on the fourth finger; 12 total enlarged femoral scales; 12 enlarged precloacal scales; enlarged femoral and enlarged precloacal scales not continuous; proximal femoral scales less than one-half the size of the distal femorals; small tubercles on forelimbs and flanks; large dorsolateral caudal tubercles and a wide ventrolateral caudal fringe not composed homogeneous scales; tail square in cross-section; slightly enlarged paired medial subcaudals not posteromedially furrowed; maximum SVL 60.1 mm; five dark transverse body bands (Tables 4,5).

Cyrtodactylus
Description of holotype (Fig. 10). Adult female SVL 60.1 mm; head moderate in length (HL/SVL 0.29), width (HW/HL 0.67), depth (HD/HL 0.38), distinct from neck, triangular in dorsal profile; lores flat anteriorly, weakly inflated posteriorly; prefrontal region slightly concave; canthus rostralis rounded; snout elongate (ES/HL 0.40), rounded in dorsal profile; eye large (ED/HL 0.24); ear opening subcircular, small; eye to ear distance greater than diameter of eye; rostral rectangular, furrowed dorsally, bordered posteriorly by large left and right supranasals and one slightly smaller azygous internasal, bordered laterally by first supralabials; external nares bordered anteriorly by rostral, dorsally by large supranasal, posteriorly by two slightly smaller postnasals, bordered ventrally by first supralabial; 12R/13L rectangular supralabials tapering smoothly to below eye; 9R/9L infralabials tapering smoothly to below eye; scales of rostrum and lores domed, slightly larger than granular scales on top of head and occiput; scales of occiput intermixed with distinct, small tubercles; superciliaries subrectangular, largest anteriorly; mental triangular, bordered laterally by first infralabials and posteriorly by large left and right trapazoidal postmentals contacting medially for approximately 40% of their length posterior to mental; one row of six (R,L) slightly enlarged sublabials extending posteriorly fifth infralabials, subsequent sublabials much smaller; gular and throat scales small, granular, grading posteriorly into slightly larger, flatter, smooth, imbricate, pectoral and ventral scales.
Body relatively short (AG/SVL 0.48) with well-defined ventrolateral folds; dorsal scales small, granular, interspersed with larger conical, semi-regularly arranged, weakly keeled tubercles; tubercles extend from occipital region onto base of tail and slightly beyond as paravertebral rows; tubercles of nape and occiput smaller than those on body; approximately 14 longitudinal rows of tubercles at midbody; approximately 34 paravertebral tubercles; tubercles on upper flanks smaller than those on dorsum; approximately 35 longitudinal rows of flat, imbricate, ventral scales much larger than dorsal scales; approximately 172 transverse rows of ventral scales; no pore-bearing, precloacal scales; 12 enlarged precloacal scales; no deep precloacal groove or depression; and approximately three rows of post-precloacal scales on midline.
Forelimbs moderate in stature, relatively short (ForL/SVL 0.13); granular scales of forearm slightly larger than those on body, interspersed with tubercles; palmar scales rounded, slightly raised; digits well-developed, relatively short, inflected at basal interphalangeal joints; digits narrower distal to inflections; subdigital lamellae wide, transversely expanded proximal to joint inflections, narrower transverse lamellae distal to joint inflections; claws well-developed, claw base sheathed by a dorsal and ventral scale; 8R/8L expanded and 10R/10L unexpanded lamellae beneath the fourth finger; hind limbs larger and thicker than forelimbs, moderate in length (TibL/SVL 0.14), covered dorsally by granular scales interspersed with moderately sized, conical tubercles dorsally and posteriorly and anteriorly by flat, slightly larger, subimbricate scales; ventral scales of thigh flat, subimbricate, larger than dorsals; subtibial scales flat, imbricate; no pore-bearing femoral scales; 6R/6L enlarged femoral scales; enlarged femoral scales not contiguous with enlarged precloacal scales, terminating distally at knee; proximal femoral scales smaller than distal femorals, the former forming an abrupt union with much smaller, rounded, ventral scales of posteroventral margin of thigh; plantar scales flat; digits relatively long, well-developed, inflected at basal interphalangeal joints; 9R/8L wide, transversely expanded subdigital lamellae on the fourth toe proximal to joint inflection that extend onto sole, and 12R/11L unexpanded lamellae beneath the fourth toe; and claws well-developed, sheathed by a dorsal and ventral scale at base.
Tail original (but in two pieces), long 80.3 mm (TL/SVL 1.34), 4.0 mm in width at base, tapering to a point; nearly square in cross-section; dorsal scales flat, square bearing tubercles forming paravertebral rows and large tubercles forming a dorsolateral longitudinal row; slightly larger, posteriorly directed, semi-spinose tubercles forming wide distinct ventrolateral caudal fringe; scales of ventrolateral fringe generally interspersed at regular intervals with larger spinose scales; medial subcaudal scales paired, slightly enlarged; subcaudals, larger than dorsal caudal scales; base of tail bearing hemipenial swellings; one conical postcloacal tubercle at base of hemipenial swellings; and postcloacal scales flat, imbricate.
Coloration in life (Fig. 10). Ground color of the head, body, limbs, and tail pale brown; faint, diffuse mottling on rostrum; lores darkly colored; wide, distinct, pale-colored postorbital stripe; nuchal band faint, bearing two posterior projections; three very faint, wide, irregularly shaped body bands between limb insertions edged in slightly darker brown; band interspaces bearing irregularly shaped, faint, dark-colored markings; dark-colored speckling on limbs and digits; digits bearing pale-colored bands; four wide, faint, dark-colored caudal bands separated by three pale-colored bands on original portion of tail; all caudal bands encircle tail; all ventral surfaces beige, generally immaculate; and iris orangish to coppery in color.
Variation. The additional specimen (THNHM 01667) closely approximates the holotype in overall coloration and pattern except that it is more boldly marked. It has four dark-colored body bands as opposed to three and a complete original tail bearing eight dark-colored and seven pale-colored bands. The pale-colored postorbital stripe is slightly thinner and less distinct. Meristic and morphometric differences are listed in Table 5. Given its overall morphological and color pattern similarities and close geographic proximity to the holotype (~ 40 km), we consider this individual as C. cf. kochangensis sp. nov. pending genetic data.
Distribution. Cyrtodactylus kochangensis sp. nov. is currently known only from the type locality at Ko Chang Island, Ko Phayam Subdistrict, Mueang Ranong District, Ranong Province, Thailand. The additional population of pending species status occurs in the Khlong Naka Wildlife Sanctuary, Suk Samran District Ranong Province.
Etymology. The specific epithet kochangensis is in reference to the type locality, Ko Chang, Ranong Province, Thailand Comparisons (based on the holotype). Cyrtodactylus kochangensis sp. nov. forms a clade with the sister species Cyrtodactylus rivularis sp. nov. and C. rukhadeva (Fig. 2) from which it differs by an uncorrected pairwise sequence divergence of 12.00-12.52% and 12.52-13.68%, respectively (Table 2) and it and C. cf. kochangensis sp. nov. are separated from C. rukhadeva and Cyrtodactylus rivularis sp. nov. by geographic distance of no less than ~280-470 km (Fig. 1). The small sample size (N = 1) precludes it from statistical analyses, however at this point, it differs from C. rukhadeva and C.  Tables 4, 5.
Natural history. The holotype (ZMKU R 00945) was collected at night (2107 h) among branches of a small tree approximately 100 cm above the ground at 36 m elevation with a temperature of 28.6 °C and relative humidity of 83.9%. The surrounding habitat was dry evergreen forest with a rocky stream nearby (Fig. 11). The new species was found to co-occur with two other species of gekkonid lizards, Cyrtodactylus oldhami (Theobald, 1876) and Gekko tokehos (Grismer, Wood, Grismer, Quah, Thy, Phimmachak, Sivongxay, Seateun, Stuart, Siler, Mulcahy, Anamza & Brown, 2019). Diagnosis. Cyrtodactylus uthaiensis sp. nov. can be separated from all other species of the brevipalmatus group by the combination of having 13-15 supralabials, 10-11 infralabials, 33 paravertebral tubercles, 17 rows of longitudinally arranged tubercles, 36 transverse rows of ventrals, 159 longitudinal rows of ventrals, eight expanded subdigital lamellae on the fourth toe, 12 unexpanded subdigital lamellae on the fourth toe, 20 total subdigital lamellae on the fourth toe; seven expanded subdigital lamellae on the fourth finger, 11 unexpanded subdigital lamellae on the fourth finger, 18 total subdigital lamellae on the fourth finger; 16 total enlarged femoral scales, 12 total femoral pores; 14 enlarged pore-bearing precloacals; enlarged femorals and enlarged precloacals not continuous; proximal femorals less than one-half the size of the distal femorals; small tubercles on forelimbs and flanks; large dorsolateral caudal tubercles and wide ventrolateral caudal fringe; ventrolateral caudal fringe composed scales of different size; tail circular in cross-section; slightly enlarged medial subcaudals intermittent, medially furrowed, posteriorly emarginated; maximum SVL 58.1 mm; and six dark transverse body bands (Tables 5, 6).
Description of holotype (Fig. 12). Adult male SVL 58.1 mm; head moderate in length (HL/SVL 0.28), width (HW/HL 0.68), depth (HD/HL 0.39), distinct from neck, triangular in dorsal profile; lores concave slightly anteriorly, weakly inflated posteriorly; prefrontal region slightly concave; canthus rostralis rounded; snout elongate (ES/ HL 0.40), rounded in dorsal profile; eye large (ED/HL 0.29); ear opening elliptical, small; eye to ear distance greater than diameter of eye; rostral rectangular, dorsally furrowed, bordered posteriorly by large left and right supranasals, bordered laterally by first supralabials; external nares bordered anteriorly by rostral, dorsally by large supranasal, posteriorly by two smaller postnasals, ventrally by first supralabial; 13R/15L rectangular supralabials tapering smoothly to posterior margin of eye; 10R/11L infralabials tapering smoothly to posterior margin of eye; scales of rostrum and lores flat to domed, slightly larger than granular scales on top of head and occiput; scales of occiput intermixed with distinct, small tubercles; superciliaries subrectangular, largest dorsally and anteriorly; mental triangular, bordered laterally by first infralabials and posteriorly by large left and right trapezoidal postmentals contacting medially for approximately 40% of their length posterior to mental; one row of slightly enlarged, elongate sublabials extending posteriorly to fifth(L) and seventh(R) infralabial; gular and throat scales small, granular, grading posteriorly into slightly larger, flatter, smooth, imbricate, pectoral and ventral scales. Body relatively short (AG/SVL 0.46) with well-defined ventrolateral folds; dorsal scales small, granular, interspersed with larger, conical, semi-regularly arranged, weakly keeled tubercles; tubercles extend from occipital region onto base of tail and slightly beyond as paravertebral rows; smaller tubercles extend anteriorly onto nape and occiput, diminishing in size anteriorly; approximately 17 longitudinal rows of tubercles at midbody; approximately 33 paravertebral tubercles; small tubercles on flanks; 36 longitudinal rows of flat, imbricate, ventral scales much larger than dorsal scales; 159 transverse rows of ventral scales; 16 total large femoral scales; 12 total femoral pores; 14 enlarged pore-bearing precloacals; no deep precloacal groove or depression; and two rows of post-precloacal scales on midline.
Forelimbs moderate in stature, relatively short (ForL/SVL 0.14); granular scales of forearm slightly larger than those on body, interspersed with large tubercles; palmar scales rounded, slightly raised; digits well-developed, relatively short, inflected at basal interphalangeal joints; digits narrower distal to inflections; subdigital lamellae wide, transversely expanded proximal to joint inflections, narrower transverse lamellae distal to joint inflections; claws well-developed, claw base sheathed by a dorsal and ventral scale; 7R/7L expanded and 11R/11L unexpanded lamellae beneath the fourth finger; hind limbs larger and thicker than forelimbs, moderate in length (TibL/SVL 0.14), covered dorsally by granular scales interspersed with moderately sized, conical tubercles dorsally and posteriorly and anteriorly by flat, slightly larger, subimbricate scales; ventral scales of thigh flat, subimbricate, larger than dorsals; subtibial scales flat, imbricate; one row of 6R/6L of enlarged pore-bearing femoral scales not continuous with enlarged pore bearing precloacal scales, terminating distally at knee; 8R/8L enlarged femoral scales; proximal femoral scales smaller than distal femorals, the former forming an abrupt union with much smaller, rounded, ventral scales of posteroventral margin of thigh; plantar scales flat; digits relatively long, well-developed, inflected at basal interphalangeal joints; 8R/(broken)L wide, transversely expanded subdigital lamellae on fourth toe proximal to joint inflection that extend onto sole, 12R/(broken)L unexpanded lamellae beneath first toe; and claws well-developed, sheathed by a dorsal and ventral scale at base.
Tail original, long 76.7 mm (TL/SVL 1.32), 4.1 mm in width at base, tapering to a point; sub-circular or nearly round in cross-section; dorsal scales flat, square bearing tubercles forming paravertebral rows and large tubercles forming a dorsolateral longitudinal row; slightly larger, posteriorly directed, semi-spinose tubercles forming small but distinct ventrolateral caudal fringe; larger scales of ventrolateral fringe occur at regular intervals; slightly enlarged medial subcaudals intermittent, medially furrowed, posteriorly; single enlarged medial subcaudals absent; subcaudal scales, larger than dorsal caudal scales; base of tail bearing hemipenial swellings; 3R/3L conical postcloacal tubercles at base of hemipenial swellings; and postcloacal scales flat, imbricate.
Coloration in life (Fig. 12). Ground color of the head, body, limbs, and tail palebrown; dark, diffuse mottling on interorbital region and snout; dark blotch on top of head; wide, pale-colored postorbital stripe irregularly edged in dark brown extends from posterior margin of one eye across nape to posterior margin of other eye; ventral portion of lores, suborbital region, and supralabials darkly mottled; wide, dark brown nuchal band, bearing two posterior projections; six irregularly shaped darkly edged body bands extending between forelimb and hind limb insertions followed by one dark sacral band; paired dark brown paravertebral blotches on nape; band interspaces bearing irregularly shaped, dark-colored markings; dark-colored speckling on limbs and digits; digits bearing pale-colored bands; seven wide dark-colored caudal bands separated by six pale-colored bands; caudal bands encircle tail resulting in heavily mottled subcaudal region; all other ventral surfaces beige, generally immaculate; and iris gold in color.
Distribution. Cyrtodactylus uthaiensis sp. nov. is currently known from the type locality at Thung Na Ngam Subdistrict, Lan Sak District, Uthai Thani Province, Thailand.
Etymology. The specific epithet uthaiensis refers to the type locality, Uthai Thani Province, Thailand.
Natural history. Cyrtodactylus uthaiensis sp. nov. is the only species of the brevipalmatus group that occurs in an isolated hilly area within the Chao Phraya River Basin (Fig. 1). The holotype (ZMKU R 00949) was collected at night (2055 h) on a bamboo twig approximately 170 cm above ground level at 106 m elevation. The habitat was isolated karst formations within a mixed deciduous forest. This area was surrounded by agricultural fields (plantations and rice fields) and human residential areas (Fig. 13). The new species was found to co-occur with a gekkonid lizard, Dixonius siamensis (Boulenger, 1899).

Discussion
A taxonomy consistent with evolutionary history is paramount to any downstream comparative analyses employed to address the evolution of other features of the group be they ecology, behavior, or habitat preference. By delimiting new and existing spe- Figure 14. Comparison of a morphology-based taxonomy pre-2020 and the current phylogenetic-based taxonomy illustrating the misrepresentation of the diversity of the Cyrtodactylus brevipalmatus group by the former. Colored lines connecting species on the left to species on the right indicate under which name the former were masquerading as the latter.
cies using a phylogenetic analysis and then diagnosing those species using univariate and multivariate statistical analyses of their morphological data, phylogenetic history will not conflated with convergent evolution (Fig. 14). This is extremely important for conservation and natural resource management programs so that efforts can be vectored towards all recognized species without missing the species that were previously masquerading among the synonymies of an erroneous taxonomy. This is especially true for range-restricted, highly specialized species, whose general morphological similarity often does not align with their phylogenetic history as exemplified in the brevipalmatus group (Fig. 3). Given that upland tropical regions are currently some of the most vulnerable to climate change (Gehrig-Fasel et al. 2007;Myers-Smith et al. 2011;Scridel et al. 2018), this argument becomes especially germane for the entire brevipalmatus group-the majority of whose species are upland populations and potentially range-restricted. The description of the four new species, Cyrtodactylus fluvicavus sp. nov., Cyrtodactylus kochangensis sp. nov., Cyrtodactylus rivularis sp. nov., and Cyrtodactylus uthaiensis sp. nov. adds to the rapidly growing list of new species of Cyrtodactylus from Thailand. Many of the poorly sampled populations reported here such as C. sp.9 from Thong Pha Phum National Park, Kanchanaburi Province; C. sp.10 from Chao Doi Waterfall, Tak Province; C. sp.11 from Phu Hin Rong Kla National Park, Phitsanulok Province; C. sp.13 from Ban Saphan Lao, Kanchanaburi Province and Thung Yai Naresuan Wildlife Sanctuary, Tak Provinces; C. sp.14 from Langkawi Island, Peninsular Malaysia; C. cf. ngati1 and C. cf. ngati2 from Xaignabouli and Vientiane Provinces, respectively, Laos (Figs 1, 2), will likely be described as new species when new material becomes available. Additionally, many populations listed in various nontechnical websites as C. interdigitalis or C. brevipalmatus based on overall similarity or general distribution, may also be recovered as new species or delegated to some of the more recently described species. This research continues to highlight the unrealized biodiversity in climatically imperiled upland ecosystems throughout Southeast Asia.