Liver samples taken from 69 specimens of Cyrtodactylus consobrinus throughout its range in southern Thailand and Peninsular Malaysia were stored in 95% ethanol. Genomic DNA was isolated from liver or skeletal muscle specimens stored in 95% ethanol using the Qiagen DNeasy animal tissue kit (Valencia, CA, USA). The mitochondrial NADH dehydrogenase subunit 2 gene (ND2) was amplified using a double-stranded polymerase chain reaction (PCR) under the following conditions: 1.0 μL genomic DNA (10–30 μg), 1.0 μL light strand primer (concentration 10 μM), 1.0 μL heavy strand primer (concentration 10 μM), 1.0 μL dinucleotide pairs (1.5 μM), 2.0 μL 5 × buffer (1.5 μM), MgCl 10 × buffer (1.5 μM), 0.1 μL Taq polymerase (5 U/μL) and 6.4 μL ultra-pure H2O. PCR reactions were executed on an Eppendorf Mastercycler gradient thermocycler under the following conditions: initial denaturation at 95 °C for 2 min, followed by a second denaturation at 95 °C for 35 s, annealing at 48–50 °C for 35 s, followed by a cycle extension at 72 °C for 35 s, for 31 cycles. All PCR products were visualized on a 1.0% agarose gel electrophoresis. Successful PCR products were vacuum purified using MANU 30 PCR plates (Millipore) and purified products were resuspended in ultra-pure water. Purified PCR products were sequenced using the ABI Big-Dye Terminator v3.1 Cycle Sequencing Kit in an ABI GeneAmp PCR 9700 thermal cycler. Cycle sequencing reactions were purified with Sephadex G-50 Fine (GE Healthcare) and sequenced on an ABI 3730xl DNA Analyzer at the BYU (Brigham Young University) DNA Sequencing Centre (DNASC). Primers used for amplification are L4437b Macey et al. (1997) External 5′-AAGCAGTTGGGCCCATACC-3′ and H5934 (Macey et al. 1997) External 5′-AGRGTGCCAATGTCTTTGTGRTT-3′ and we also used these for sequencing along with two internal primers CyrtintF1 Siler et al. (2010) Internal 5′-TAGCCYTCTCYTCYATYGCCC-3′ CyrtintR1 Siler et al. (2010) Internal 5′-ATTGTKAGDGTRGCYAGGSTKGG-3′. To this, four samples of C. limajalur Davis, Bauer, Jackman, Nashriq & Das, 2019, four samples of C. malayanus (de Rooij, 1915), one sample of C. kapitensis Davis, Nashriq, Woytek, Wikramanayake, Bauer, Karin, Brennan, Iskandar & Das, 2023, three samples of C. hantu Davis, Das, Leaché, Karin, Brennan, Jackman, Nashriq, Chan & Bauer, 2021, and one sample of an unnamed species of the malayanus group were downloaded from GenBank (Table 1). The protein-coding region of the ND2 sequence was aligned manually. Mesquite (Maddison and Maddison 2015) was used to calculate the correct amino acid reading frame and to confirm the lack of premature stop codons. The GenBank accession numbers for all specimens are listed in Table 1.

Maximum likelihood (ML) and Bayesian inference (BI) were used to estimate the phylogenetic relationships based on aligned sequences. The ML phylogeny was estimated using the IQ-TREE webserver (Nguyen et al. 2015; Trifinopoulos et al. 2016) preceded by the selection of the best substitution models using the Bayesian Information Criterion (BIC) in ModelFinder (Kalyaanamoorthy et al. 2017), which supported K2P+I+G4 for the non-coding position, TIM3+F+I+G4 for codon position1, TN+F+I+G4 for codon position 2, and TVM+F+T+G4 for codon position 3. Ten-thousand bootstrap pseudoreplicates via the ultrafast bootstrap (UFB; Hoang et al. 2018) approximation algorithm were employed and nodes having UFB values of 95 and above were considered strongly supported (Minh et al. 2013). The Bayesian inference (BI) analysis was carried out in MrBayes 3.2.3. (Ronquist et al. 2012) on XSEDE using the CIPRES Science Gateway (Cyberinfrastructure for Phylogenetic Research; Miller et al. 2010) employing bModelTest for codon partitioning. Two independent Markov chain Monte Carlo (MCMC) simulations were performed each with four chains, three hot and one cold. We ran the MCMC simulation for 50 million generations, sampled every 5,000 generations, and discarded the first 10% of each run as burn-in. Convergence and stationarity of all parameters from both runs were checked in Tracer v1.6 (Rambaut and Drummond 2013) to ensure effective sample sizes (ESS) were above 200. Post-burn-in sampled trees from both runs were combined using the sumt function and a 50% majority rule consensus tree was constructed. Nodes with Bayesian posterior probabilities (BPP) of 0.95 and above were considered strongly supported (Huelsenbeck et al. 2001; Wilcox et al. 2002). Retaining only ingroup taxa, uncorrected pairwise sequence divergences were calculated in MEGA X (Tamura et al. 2021) using the pairwise deletion option.

All statistical analyses were conducted using R Core Team (2024). A Levene’s test for meristic and size-corrected morphometric characters was conducted to test for equal variances across all groups. To remove potential effects of allometry in the morphometric characters (see Chan and Grismer 2022), size was adjusted 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, 1983; Turan 1999; Lleonart et al. 2000). The statistical analyses employed on all data sets included combinations of multiple factor analysis (MFA) from the R package FactorMineR (Husson et al. 2017) and visualized using the Factoextra package (Kassambara and Mundt 2017); principle component analysis (PCA) and discriminant analysis of principle components (DAPC) from the adegent package 2.1.5 in R (Jombart 2022; PCAtest from Camargo (2022); non-parametric permutation multivariate analysis of variance (PERMANOVA) from the vegan package 2.5-3 in R (Oksanen et al. 2020); and analysis of variance (ANOVA) followed by a TukeyHSD post hoc test.

The general lineage concept (GLC: de Queiroz 2007) adopted herein proposes that a species constitutes a metapopulation of organisms evolving independently from other such populations usually owing to a lack of 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 were used to recover monophyletic mitochondrial lineages of individual(s) (i.e., populations) in order to develop initial species-level hypotheses, the grouping stage of Hillis (2019). Univariate, multivariate, and discrete color pattern and morphological data were used to search for statistically significant unique characters and patterns and compare their consistency with the previous species-level hypotheses designations—the construction of boundaries representing the hypothesis-testing step of Hillis (2019)—thus providing lineage diagnoses independent of the molecular analyses. In this way, the inherent errors of simultaneously delimiting (phylogeny) and diagnosing (taxonomy) species are avoided (Frost and Hillis 1990; Frost and Kluge 1994; Hillis 2019).

The phylogenetic relationships within the malayanus group indicate that peninsular Cyrtodactylus consobrinus and Bornean C. consobrinus form distinct monophyletic mitochondrial lineages separated by a sequence divergence of 4.5% (Fig. 1). The PERMANOVAs of the MFAs show that peninsular C. consobrinus plots significantly different from all other species in the malayanus group (Figs 2C, 3C) and the ANOVAs recovered peninsular C. consobrinus as significantly different from Bornean C. consobrinus in nine morphometric and six meristic characters (Tables 2, 3). Based on these data and their allopatry, clearly precluding any possibility of contemporary gene flow, we consider peninsular C. consobrinus to be an unnamed species that we describe below.

The well-supported phylogeographic structure within peninsular C. consobrinus is discordant with the morphospatial similarity among its lineages, nor is it strongly corroborated in the ANOVA analyses, save for the eastern lineage. Thus, at this juncture, we consider peninsular C. consobrinus to be a single new species. Given that these vagile geckos are common inhabitants of both primary and secondary lowland and upland forests (Grismer 2011), the possibility of admixture among the lineages will remain unresolved in the absence of a genomic dataset. Therefore, we elect not to recognize any of the lineages as distinct species. Cyrtodactylus consobrinus sensu stricto is restricted to Borneo given that the type locality is “Sarawak, [Borneo]” (Peters 1871: 569).

 Cyrtodactylus peninsularis sp. nov.

https://zoobank.org/AEE782A0-E37B-4815-88E0-9768947965EA

Figs 7, 8, 9

Gymnodactylus consobrinus — Boulenger 1912: 37 (part); Smith 1930: 13 (part); Bourret 2009: 123 (part).

Cyrtodactylus consobrinus — Grandison 1972: 81 (part); Dring 1979: 181 (part); Denzer and Manthey 1991: 314 (part); Kluge 2001: 8 (part); Cox et al. 1998: 88 (part); Chan-ard et al. 1999: 1058, 1064–65 (part), 2015: 49, 51 (part); Grismer 2008: 30 (part), 2011: 386 (part); Rösler 2016: 12 (part); Grismer and Quah 2019: 233; Sumarli et al. 2015: 5, 19 (part); Hong et al. 2021:799, 800 & 807; Quah et al. 2021:241 & 246; Davis et al. 2023: 3 (part); Poyarkov et al. 2023: 299, 301 (part).

Cyrtodactylus consubrinus — Manthey and Grossmann 1997: 221 (part; unjustified subsequent spelling).

Cyrtodactylus cf. consobrinus — Figueroa et al. 2023: 100.

Cyrtodactylus (Cyrtodactylus) consobrinus — Rösler 2000: 65 (part).

Type material.

Holotype • Adult male (LSUHC 10202) collected from the base of Gunung Belumut, Johor State, Peninsular Malaysia (2.065817°N, 103.526119°E at 245 m) on 8 September 2009 by Evan S. H. Quah, Perry L. Wood, Jr., Jesse L. Grismer, L. Lee Grismer, Shahrul Anuar, and Mohamed A. Muin. Paratypes. • Adult females: LSUHC 9333 northwestern lineage collected from Sungai Sedim, Kedah State, Peninsular Malaysia (5.414063°N, 100.779803°E at 129 m) on 16 March 2009; LSUHC 10048 northeastern lineage collected from Hutan Lipur Sekayu, Terengganu State, Peninsular Malaysia (4.980644°N, 102.934645°E at 441 m) on 27 March 2009; LSUHC 11136–37 north-central lineage collected from Gunung Stong, Kelantan State, Peninsular Malaysia (5.321880°N, 101.964944°E at 703 m) on 27 June 2009; LSUHC 11152 north-central lineage collected from Hutan Lipur Jelawang, Kelantan State, Peninsular Malaysia (5.340351°N, 101.969544°E at 699 m) on 27 June 2009; LSUHC 11979 northeastern lineage collected from Sungai Bubu, Terengganu State, Peninsular Malaysia (5.0120261°N, 102.952963°E at 77 m) on 1 September 2009; and LSUHC 12386 eastern lineage collected from Hutan Lipur Chemerong, Terengganu State, Peninsular Malaysia (4.660664°N, 103.001320°E at 129 m) on 17 August 2009. Adult males: LSUHC 10230 southern lineage collected from the base of Gunung Belumut, Johor State, Peninsular Malaysia (2.045596°N, 103.530185°E at 244 m) on 9 September 2009 by Evan S. H. Quah, Perry L. Wood, Jr., Jesse L. Grismer, L. Lee Grismer, Shahrul Anuar, and Mohamed A. Muin; and LSUHC 11267 northwestern lineage collected from trail 2, Sungai Enam, Belum, Perak State, Peninsular Malaysia (5.46768°N,101.28961°E on 6 October 2009.

Additional specimens examine

(n = 52). See Suppl. material 5.

Diagnosis based on type series.

Cyrtodactylus peninsularis sp. nov. can be separated from all other species of the malayanus group by the combination of having a maximum SVL of 128.7 mm (female); 8–10 supralabials; 10–12 infralabials; 25–30 paravertebral tubercles; 15–20 rows of longitudinally arranged tubercles; 40–62 longitudinal rows of ventrals; 243–299 transverse rows of ventrals; 7–9 expanded subdigital lamellae on the fourth toe; 13–16 unmodified subdigital lamellae on the fourth toe; 21–25 total subdigital lamellae on the fourth toe; 21–25 total number of enlarged femorals; 2–9 total number of femoral pores in males, no femoral pores in females; 10–12 enlarged precloacals; nine or ten precloacal pores in males (n = 3), precloacal pores in some females (three of seven); two or three rows of large post-precloacals; two postcloacal tubercles (spines) on each side; dorsal pattern extremely variable, dark dorsal bands very wide reducing the pale dorsal interspaces to 2–4 thin lines; seven or eight dark and pale caudal bands (n = 3); large moderately keeled body tubercles; caudal tubercles extend beyond base of tail; subcaudals transversely expanded but not extending high up onto side of tail; enlarged distal femorals and enlarged precloacals not contiguous; no enlarged proximal femorals; top of head overlain with reticulating white network of thin lines; dark caudal bands wider than pale caudal bands; dark markings usually within pale caudal bands in adults (Tables 7, 8, Figs 5, 6, Suppl. material 2).

Table 7.

Download as

CSV

XLSX

Meristic and raw morphometric characters of the type series of Cyrtodactylus peninsularis sp. nov. na = data unavailable, R = regenerated.

Catalog number	LSUHC 10202	LSUHC 09633	LSUHC 10048	LSUHC 10230	LSUHC 11136	LSUHC 11137	LSUHC 11152	LSUHC 11267	LSUHC 11979	LSUHC 12386
	holotype	paratype	paratype	paratype	paratype	paratype	paratype	paratype	paratype	paratype
sex	M	F	F	M	F	F	F	M	F	F
Meristic characters
supralabials (SL)	9	8	8	8	9	10	9	9	9	10
infralabials (IL)	11	12	11	11	11	11	10	10	11	11
internasals (INA)	3	2	4	5	3	3	2	2	4	2
paravertebral tubercles (PVT)	30	30	28	29	25	25	25	29	29	29
longitudinal tubercle rows (LRT)	16	17	18	20	16	16	15	19	17	17
longitudinal ventral scale rows (LVS)	47	41	62	56	40	54	40	41	52	56
transverse ventral scale rows (TVS)	252	265	281	251	279	279	274	269	243	299
total enlarged femoral scales (FS)	12	12	12	13	10	12	na	12	12	11
femoral pores-right	4	na	na	5	na	na	na	2	na	na
femoral pores-left	3	na	na	4	na	na	na	3	na	na
total femoral pores (FP)	7	na	na	9	na	na	na	5	na	na
enlarged precloacals (PS)	12	10	11	12	12	12	10	11	10	11
precloacal pores (PP)	9	na	10	10	10	na	na	9	na	10
post-precloacal scales (PPS)	3	2	3	2	3	2	3	3	3	3
postcloacal tubercles (CT)	2	2	2	2	2	2	2	2	2	2
expanded 4th toe lamellae (E4TL)	7	8	9	8	8	9	9	8	8	8
unexpanded 4th toe lamellae (U4TL)	16	16	16	16	16	14	16	13	14	16
total 4th toe lamellae (T4TL)	23	24	25	24	24	23	25	21	22	24
dark body bands (BB)	3	2	3	4	3	3	3	2	3	2
dark colored caudal bands (DCB)	7	6R	na	B	8	R	8	na	B	B
pale colored body bands (LCB)	7	5R	na	B	7	R	8	na	B	B
Morphometric characters (mm)
snout-vent length (SVL)	113.10	121.1	115.6	112.4	121.9	128.7	126.2	125.0	122.6	112.0
tail length (TL)	136.8	136.7	120.0	131.0R	132.45	108.8	139.9	118.1R	116.8R	86.6R
tail width (TW)	8.6	7.7	10.0	8.3	7.6	9.4	8.3	8.9	8.1	8.5
brachial length (BracL)	19.0	19.9	19.6	19.2	21.2	19.8	21.0	21.0	18.2	18.2
forearm length (ForeL)	17.6	18.8	18.0	18.2	20.0	19.2	20.4	17.34	17.9	18.8
femoral length (FemL)	22.7	23.0	22.1	23.1	25.7	25.2	24.8	23.1	22.8	20.7
tibia length (TibL)	20.2	19.3	20.9	19.4	20.6	22.2	21.8	18.9	20.2	19.7
axilla-groin length (AG)	52.6	54.2	49.7	54.6	53.2	58.9	54.8	58.2	51.7	52.4
head length (HL)	33.1	35.4	32.7	33.1	33.8	35.6	35.1	35.4	34.5	31.8
head width (HW)	20.4	21.9	21.7	21.7	22.6	24.8	23.6	24.5	22.7	20.4
head depth (HD)	12.3	12.1	12.0	12.8	13.5	13.9	13.7	13.9	12.2	11.8
eye diameter (ED)	7.0	7.4	6.9	7.0	7.3	8.1	7.5	7.1	7.3	7.0
eye to ear distance (EE)	7.4	8.9	9.1	8.8	9.3	7.5	10.5	9.1	9.4	8.2
snout length (SN)	12.7	13.4	12.9	12.4	13.7	14.6	14.5	13.7	13.8	12.3
eye to nostril distance (EN)	9.8	10.1	10.3	10.0	10.7	9.1	11.5	10.7	10.8	9.6
interorbital distance (IO)	10.1	9.0	10.7	10.8	11.6	12.2	12.34	12.1	9.4	9.5
ear length (EarL)	2.2	2.7	2.6	1.8	1.6	1.5	1.6	1.9	1.8	1.9
internarial distance (IN)	3.8	3.6	3.6	3.9	3.9	5.0	4.8	4.2	4.3	3.6

Table 8.

Download as

CSV

XLSX

Summary statistics of the meristic characters Cyrtodactylus peninsularis sp. nov. Abbreviations are in Materials and methods.

	SL	IL	PVT	LRT	LVS	TVS	FS	PS	PPS	E4TL	U4TL	T4TL	INA	BB	FP	PP	LCB	DCB
mean	8.9	10.9	27.9	17.1	48.9	269.2	11.8	11.1	2.7	8.2	15.3	23.5	3	2.7	7.0	9.7	7.7	7.3
±sd	0.74	0.57	2.08	1.52	8.13	16.91	0.79	0.88	0.48	0.63	1.16	1.27	1.05	0.67	2.99	0.52	0.58	0.58
range	8–10	10–12	25–30	15–20	40–62	243–299	10–13	10–12	2 or 3	7–9	13–16	21–25	2–5	2–4	2–9	9 or 10	7 or 8	7 or 8
N	10	10	10	10	10	10	10	10	10	10	10	10	10	10	3	6	3	3

Description of holotype

(Fig. 7, Table 7). Adult male SVL 113.1 mm; head moderate in length (HL/SVL 0.29), width (HW/HL 0.62), somewhat flattened (HD/HL 0.37), distinct from neck, triangular in dorsal profile; lores concave anteriorly, inflated posteriorly; prefrontal deeply concave; canthus rostralis rounded; snout elongate (SN/HL 0.38), flat, rounded in dorsal profile; eye large (ED/HL 0.21); ear opening elliptical, obliquely oriented, moderate in size; eye to ear distance slightly greater than diameter of eye; rostral rectangular, partially divided dorsally, bordered posteriorly by large left and right supranasals and slightly smaller internasal, bordered laterally by first supralabials; external nares directed posterolaterally, bordered anteriorly by rostral, dorsally by large supranasal, posteriorly by six small postnasals, ventrally by first supralabial; nine (R,L) rectangular supralabials tapering to below posterior margin of eye, first six supralabials largest; 11 (R,L) infralabials tapering smoothly to slightly past termination of enlarged supralabials; scales of rostrum and lores raised, much larger than granular scales on top of head and occiput; scales of occiput intermixed with small, rounded, tubercles; superciliaries flat, elongate, largest dorsally; mental triangular, bordered laterally by first infralabials and posteriorly by large left and right trapezoidal postmentals contacting medially for ~ 40% of their length posterior to mental; one row of enlarged, sublabials extending posteriorly to fifth infralabials (R,L); gular and throat scales small, granular, grading posteriorly into slightly larger, flatter, smooth, subimbricate pectoral and ventral scales.

Figure 7. 

A, C dorsal and ventral views of adult male holotype of Cyrtodactylus peninsularis sp. nov. LSUDPC 10202, respectively B gular region, throat, pectoral region, and ventral view of forelimbs D precloacal region, ventral view of hind limbs, and anterior subcaudal region.

Body relatively long (AG/SVL 0.47) with well-defined ventrolateral folds; dorsal scales small, granular, interspersed with large, moderately keeled, semi-regularly arranged tubercles extending from occiput to beyond base of tail; ~ 16 longitudinal rows of tubercles at midbody; ~ 30 paravertebral tubercles; ~ 47 flat, imbricate, ventral scales much larger than dorsal scales; 12 enlarged precloacal scales not separated medially by poreless scales; no deep precloacal groove or depression; and three rows of large post-precloacal scales on midline.

Forelimbs moderate in length and stature (ForeL/SVL 0.16); granular scales of forelimbs slightly larger than those on body, large spinose tubercles on dorsal surface of forearms; palmar scales slightly rounded, juxtaposed; digits well-developed, inflected at basal interphalangeal joints, slightly narrower distal to inflections; subdigital lamellae transversely expanded, those proximal to joint inflections much wider than lamellae distal to inflections; claws well-developed, sheathed by a dorsal and ventral scale; hind limbs robust, wider and longer than forelimbs (TibL/SVL 0.18), covered dorsally by granular scales interspersed with large pointed tubercles; anterior scales of thigh slightly larger and flatter than dorsal scales of thigh; ventral scales of thighs rounded, subimbricate, slightly larger than dorsals; distal subtibials large, flat, subimbricate; one row of six (R, L) distal enlarged femoral scales, four on right bearing pores and three on left bearing pores, no other enlarged femoral scales; proximal femorals not forming an abrupt union with granular posteroventral scales of thigh; plantar scales rounded, juxtaposed; digits well-developed, inflected at basal interphalangeal joints; claws well-developed, sheathed by a dorsal and ventral scale at base; seven (R, L) wide subdigital lamellae on fourth toe proximal to joint inflection, 16 (R, L) narrower lamellae distal to joint inflection, 23 total subdigital lamellae.

Tail long (TL/SVL 1.21), original, tapering to a point; dorsal caudal scales small, generally square, juxtaposed; median row of subcaudals significantly larger than dorsal caudals, transversely expanded, not extending high up dorsally onto lateral side of tail; body tubercles extending beyond base of tail; hemipenial swellings at base of tail, two large postcloacal tubercles on both sides; and postcloacal scales flat, imbricate.

Color and pattern in preservative

(Fig. 7). No photograph of the living holotype was available. Ground color of top of head, limbs, and dorsum brown; top of head overlain with a reticulating network of thin white lines; snout bearing irregularly shaped white blotches and lines; thin, white, transverse line (i.e., interspace) on occiput and another immediately anterior to shoulders; a thin, white anterior dorsal line bifurcates paravertebrally forming two thin lines along anterior of flanks; a second thin white line occurs just posterior to midway between the limb insertions; another thin white line occurs at level of groin; all white lines bordered by large white tubercles and are thickly edged in dark brown; center of the brown regions between the thin white lines (i.e., body bands) bear irregularly shaped central pale brown areas; a thin white sacral line followed by seven widely spaced white caudal bands bearing darkened markings, separated by dark caudal bands nearly three times width of pale caudal bands; limbs dark brown to pale brown overlain with thin, white broken lines and irregularly shaped markings.

Etymology.

The species name peninsularis is in reference to the distribution of this species which is restricted to the Thai-Malay Peninsula of southern Thailand, Peninsular Malaysia, and Singapore.

Distribution.

Cyrtodactylus peninsularis sp. nov. ranges from extreme southern Thailand southward through nearly all habitats in Peninsular Malaysia to Singapore (Grismer 2011) (Fig. 1). The Pulau Singkep population of Indonesia has not been investigated.

Variation.

Color pattern varies so extensively within Cyrtodactylus peninsularis sp. nov. it essentially defies a concise meaningful description (e.g., Fig. 8A, C). The type series was chosen to not only cover the geographic range of the species but to cover a great deal of its color pattern variation as well (Fig. 9). This remarkable variation, however, is paradoxically coupled to considerable conservatism in its morphological variation (Figs 4, 5, 6, Tables 4, 5).

Figure 8. 

Color and banding pattern variation among the lineages of Cyrtodactylus peninsularis sp. nov. A NEL— adult male, Lata Kekabu, Setiu, Terengganu, La Sierra University Digital Photograph Collection (LSUDPC) 13548, photo by Evan S. H. Quah B EL—juvenile, Endau-Rompin National Park, Johor, LSUHC 2585, photo by L. Lee Grismer C NWL—adult male, Sungai Enam, Perak LSUDPC 13549 (paratype LSUHC 11267), photo by Evan S. H. Quah D WL—adult female, Gunung Ledang, Johor LSUDPC 13550, photo by Evan S. H. Quah E NCL—adult female, Gunung Tebu, Terengganu LSUDPC 7997, photo by L. Lee Grismer F WL—adult female, Gunung Korbu, Perak, LSUDCP 13548, photo by Kin Onn Chan G NEL—adult male, Hutan Lipur Sekayu, Terengganu, LSUDPC 5951, photo by L. Lee Grismer H SL—adult female, Gunung Pulai, Johor, LSUDPC 13552, photo by Evan S. H. Quah.

Figure 9. 

Paratypes of Cyrtodactylus peninsularis sp. nov. showing the wide variation in dorsal banding pattern.

Comparisons.

Cyrtodactylus peninsularis sp. nov. may be the sister species of C. consobrinus from which it differs morphometrically in having a statistically shorter snout-vent length; axilla-groin length; snout length; tibial and forelimb length; head length, width, and depth, and femur length (Table 3). It differs from C. consobrinus meristically by having statistically more infralabials; longitudinal rows of tubercles; enlarged precloacal scales; and fewer supralabial scales, longitudinal ventral scales, and paravertebral tubercles (Table 3). Hatchling C. peninsularis sp. nov. nearly always have thin white interspaces whereas they are yellow in C. consobrinus (Fig. 10). From the closely related C. hutan, C. peninsularis sp. nov. differs morphometrically in having a statistically shorter AG, SN, TibL, and ForeL, and shorter HL and HW. It differs meristically in having fewer SL, LVS, PS, and PVT and more IL and LRT (Table 3).

Figure 10. 

Cyrtodactylus consobrinus from Gua Angin, Sarawak, East Malaysia (Borneo) A adult female LSUDPC 10767 B adult male LSUDPC 10764 C hatchling LSUDPC 4928 D Juvenile LSUDPC 10769.

Natural history.

Much of the following is adapted from Grismer (2011). Cyrtodactylus peninsularis sp. nov. is a vagile, nocturnal, scansorial species that ranges throughout extreme southern Thailand, Peninsular Malaysia, and forested areas of Bukit Timah in Singapore up to ca 800 m in elevation (Chan et al. 2019). Cyrtodactylus peninsularis sp. nov. is a relatively common inhabitant of primary and secondary dipterocarp forests and is occasionally found in peat swamps. Lizards are usually seen climbing on tree trunks, branches, exposed roots, and fallen logs where there are nearby crevices and holes into which they can quickly retreat when threatened. It is not uncommon to find multiple individuals on the same large tree up to 5 m above ground. Lizards are less commonly found among large boulders, taking refuge in cracks or in holes at their base. Anecdotally, this species seems to be more abundant in riparian areas. The holotype was found during the early evening at the base of Gunung Belumut at 245 m in elevation and 2 m above the ground on the trunk of a dipterocarp tree in primary forest (Fig. 11).

Figure 11. 

Habitat of the type locality at the base of Gunung Belumut, Johor State.

The discordance between the morphological variation and the geographically structured genetic variation in Cyrtodactylus peninsularis sp. nov. is unlike that seen in other species of Cyrtodactylus sharing the same distribution across Peninsular Malaysia. Johnson et al. (2012) constructed an ND2 phylogeny of C. quadrivirgatus (n = 77), a vagile forest generalist, across its entire range in Peninsular Malaysia and designated 14 monophyletic lineages (“clades” using their rhetoric) from different geographic regions. However, there was little to no support for relationships among the lineages which bore extremely short internode branch lengths and an uncorrected pairwise sequence divergence of only 0.14–3.3%. The distribution of some C. quadrivirgatus lineages partially matched that of some of the lineages within C. peninsularis sp. nov. but many others showed intraspecific geographic discordance (i.e., closely related lineages were not necessarily geographically close). There was also no discrete morphological differentiation among any of the lineages and a mismatch distribution analysis indicated that C. quadrivirgatus is undergoing range expansion.

Grismer et al. (2014) constructed an ND2 phylogeny of the pulchellus group species across its range Peninsular Malaysia (n = 78) and recovered 11 well supported clades from distinctive geographic regions. These clades were closely aligned to microhabitat preference (i.e., granite, karst, forest, generalist [Grismer et al. 2021]) and bore little interspecific phylogeographic concordance (Grismer et al. 2014: fig. 1) or concordance with the lineages of C. peninsularis sp. nov. Each clade was morphologically distinct and given its own species designation. The uncorrected pairwise sequence divergence among the species ranged from 5.5–14.5%.

Surprisingly, there is reasonably strong phylogeographic concordance among peninsular populations of Amolops, a fast-flowing rocky stream-adapted ranid frog, and the lineages of Cyrtodactylus peninsularis sp. nov. Chan et al. (2017, 2018) recovered and East 1 clade matching the distribution of the NCL of C. peninsularis sp. nov., including a disjunct population on an eastward projecting arm in the northern portion of the Banjaran Titiwangsa. A closely related East 2 clade matched the distribution of the southern portion of the EL and a Larutensis clade matched the distribution of the southwestern arm of the NWL. Lastly, their West 1–4 clades were similar to the WL of C. peninsularis sp. nov. and its distribution through the Banjaran Titiwangsa. The East 1 and 2 clades were given their own species designation as Amolops gerutu and A. australis, respectively (Chan et al. 2018). These patterns are also congruent with important biogeographic regions inferred by Chan and Grismer (2021) that were based on distribution data from 70 species of frogs and 85 species of lizards. In that study, Chan and Grismer (2021) characterized four significant biogeographic regions: (1) the Eastern Region, which encompasses the distribution of the A. gerutu and NCL of C. peninsularis sp. nov.; (2) the Southern Region, encompassing the distribution of A. australis and EL of C. peninsularis sp. nov.; and (3) Bintang Range and the surrounding lowlands, which includes the distribution of A. larutensis and NWL of C. peninsularis sp. nov.. Interestingly, Chan and Grismer (2021) also identified a Northwestern Region as a separate bioregion that includes the northwestern states of Penang, Kedah, and Perlis. Cyrtodactylus peninsularis sp. nov. occurs in Kedah and Penang (Grismer 2011) but we were unable to include those populations in this study. Based on the congruence in the distribution patterns of C. peninsularis sp. nov. and bioregions identified by Chan and Grismer (2021), it is highly likely that the population genetic structure of C. peninsularis sp. nov. from the northwestern states of Kedah and Penang will reflect a similar regionalized separation. These data suggest that even though these sympatric species presumably shared the same or similar environmental history on the Thai-Malay Peninsula, differences in their phylogeographic structure are most likely due to their inherent biological differences, some being vagile forest dwellers, others being stream dwellers, and others being microhabitat specialists.

Davis et al. (2023) stated that “many species in the genus [Cyrtodactylus] are described by primarily relying on mitochondrial pairwise distance cut-offs that have not been rigorously tested…” First, a pairwise distance cut-off would not be used to “describe” a species but could be inappropriately used to delimit a species. Although their statement is likely to be true in a diagnostic sense, we know of no such case, nor were any citations provided. They also noted that “mitochondrial datasets can mislead delimitation decisions” and may only “represent population-level diversity”. It has been well-established that phylogenetic structure does not equal species level differences, and that statistically well-supported clades from any data set need evidence from the animals’ natural history (morphology, ecology, behavior, etc.), something not considered in Davis et al. (2024), to indicate these clades may be on separate evolutionary trajectories (Sukumaran and Knowles 2017). The mtDNA gene tree in Davis et al. (2023: Suppl. material 1) showed no discordance with their genomic tree and recovered the same major lineages (Fig. 2). However, these data are not corroborated with robust statistically defensible morphological analyses and lead the authors to naïvely posit that some morphological characters (they listed five) are also problematic because of interspecific variation. They went on to unwisely overstate that this variation would be the same for “many morphological characters” for “many other species complexes across the distribution of the genus”—which extends from South Asia to Melanesia (Grismer et al. 2021)—but provided no citations. The fact is that all morphological characters vary, but how this variation is treated is what matters. By not subjecting character variation in their datasets to basic univariate statistical analyses (e.g., student t-tests or ANOVAs) which can separate signal from noise, character variation may seem uninformative. Additionally, it matters which characters are used because one dataset does not fit all—especially within a genus as diverse (350+ species) as Cyrtodactylus. For example, many characters used to diagnose (not delimit) species among the karst-dwelling taxa of the chauquangensis group (Grismer et al. 2024) would be uninformative for diagnosing the terrestrial species in the triedrus group and vice versa (Amarasinghe et al. 2022; Agarwal et al. 2023). We emphasize here that by recognizing statistically significant morphological differences that correlate with strongly supported monophyletic geographically isolated mitochondrial (or nuclear) lineages, continues to move science forward by providing robustly corroborated testable hypotheses to be evaluated later with additional data sets and analyses (see Grismer et al. 2024a, 2025; Chhin et al. 2024; Neang et al 2024; and dozens of recent citations therein).

As with many other species of lizards, the northern distribution of Cyrtodactylus peninsularis sp. nov. terminates in the vicinity of the Kangar-Pattani Line (Grismer 2011; Quah and Anuar 2018; Quah et al. 2023 and references therein). Given that this is a major faunal, floristic, and climatic transition zone from aseasonal to seasonal evergreen tropical forest, tolerances of each species to different abiotic factors may affect the extent of their distribution. Because temperature and moisture levels affect the floral composition in this area (Baltzer et al. 2007), forest adapted species such as C. peninsularis sp. nov. and others may be more susceptible to changes in these variables across this transition zone than would rock-adapted species that usually do not depend on the associated forest for refuge or resources and rock habitats tend to harbor higher levels of humidity (e.g., Grismer 2011; Grismer et al. 2018).

The authors have declared that no competing interests exist.

No ethical statement was reported.

The work of CS and PP was supported by the Thailand Science Research and Innovation Fund and the University of Phayao (Unit of Excellence 2025 on Aquatic animals biodiversity assessment (Phase I)) to C. Suwannapoom. The work of NAP was was carried out with financial support from the Russian Science Foundation (RSF grant N° 22-14-00037).

Conceptualization: L.L. Grismer, Perry L. Wood Jr. Formal analyses: L.L. Grismer, Perry L. Wood Jr., A. Kaatz. Investigation – field work: L.L. Grismer, J.L. Grismer, E.S.H. Quah, K. O. Chan, S, Anuar, M.A. Muin, N. Poyarkov, C. Suwannapoom, laboratory work: Perry L. Wood Jr., Matthew L. Murdoch, Jeren J. Gregory, N. Poyarkov, Eddie Nguyen. Writing – Original draft: L.L. Grismer. Review and Editing: All authors.

L. Lee Grismer https://orcid.org/0000-0001-8422-3698

Amanda Kaatz https://orcid.org/0009-0005-3444-4560

Jesse L. Grismer https://orcid.org/0000-0002-2542-5430

Eddie Nguyen https://orcid.org/0009-0003-0790-6817

Jeren J. Grergory https://orcid.org/0009-0003-3022-1621

Perry L. Wood Jr. https://orcid.org/0000-0003-3767-5274

Matthew L. Murdoch https://orcid.org/0000-0001-5914-6408

Shahrul Anuar https://orcid.org/0000-0003-0648-7318

Chan Kin Onn https://orcid.org/0000-0001-6270-0983

Muhamad A. Muin https://orcid.org/0000-0001-6712-265X

Parinya Pawangkhanant https://orcid.org/0000-0002-0947-5729

Chatmongkon Suwannapoom https://orcid.org/0000-0002-3342-1464

Nikolay A. Poyarkov https://orcid.org/0000-0002-7576-2283

Evan S. H. Quah https://orcid.org/0000-0002-5357-1953

All of the data that support the findings of this study are available in the main text or Supplementary Information.