A new species of the genus Ptyctolaemus Peters, 1864 (Squamata, Agamidae) from Sagaing, Myanmar

A new species of Ptyctolaemus Peters, 1864 is described from Htamanthi Wildlife Sanctuary, Sagaing Division, Myanmar. The new species differs from P. gularis and Ptyctolaemus aff. gularis from Tibet, China, by having relatively longer limbs and different colorations of the gular region, and it differs from P. collicristatus by having much longer limbs and a less developed nuchal crest in males. Moreover, the new species differs genetically from Ptyctolaemus aff. gularis from Tibet, China, and P. collicristatus by an uncorrected percentage distance of 23.5% and 24.8%, respectively, inferred from mitochondrial NADH dehydrogenase subunit 2 gene sequences. This discovery increases the number of known Ptyctolaemus species to three.

During our fieldwork in northern Myanmar in 2019, four specimens of Ptyctolaemus cf. gularis were collected from Htamanthi Wildlife Sanctuary, Sagaing Division, Myanmar. Morphological and molecular data of these specimens showed obvious differences from the data of the specimens from Tibet, China recorded by Huang (1980) and Che et al. (2020). As no molecular data are available for the type specimen of P. gularis, after comparing the morphological data (Huang 1980;Che et al. 2020) of the specimens from Tibet, China, and the morphological data of our newly collected specimens with the morphological data (Schulte et al. 2004) of the type specimen (holotype ZMB 5004) of P. gularis, we found that the specimens from Tibet, China, largely agree with the type specimen but that the newly collected specimens disagree with the type specimen. Therefore, we consider the specimens from Tibet, China, as Ptyctolaemus aff. gularis and describe the specimens from Sagaing, Myanmar, as a new species herein.

Materials and methods
Field surveys in northern Myanmar were undertaken at the invitation of the Republic of the Union of Myanmar, Ministry of Natural Resources and Environmental Conservation, Forest Department, Forest Research Institute. Specimens were collected by hand at night from Htamanthi Wildlife Sanctuary, Sagaing Division, Myanmar, in December 2019. Photographs were taken to document the color pattern in life prior to euthanasia. Liver tissues were stored in 99% ethanol and lizards were preserved in 75% ethanol. Specimens were deposited in Southeast Asia Biodiversity Research Institute, Chinese Academy of Sciences (abbreviation: SEABRI; address: Yezin, Nay Pyi Taw, Myanmar).
Genomic DNA was extracted from preserved liver tissues using a standard phenol-chloroform extraction protocol (Sambrook et al. 1989). A fragment of the mitochondrial gene NADH dehydrogenase subunit 2 (ND2) gene was amplified and sequenced by using the primers L4437b: AAGCAGTTGGGCCCATACC and H5540: TTTAGGGCTTTGAAGGC (Macey et al. 2000). PCR was conducted using a denaturing step at 94 ℃ for 35 s, annealing at 55 ℃ for 35 s, and extending at 70 ℃ for 150 s. PCR products were electrophoresed in 1.5% agarose gels, visualized with ethidium bromide. The products were purified and sequenced Tsingke Biotechnology (Beijing) Co., Ltd., using the same primers as in PCR. Sequences were edited and manually managed using SeqMan in Lasergene 7.1 (DNASTAR Inc., Madison, WI, USA) and MEGA X (Kumar et al. 2018). All new sequences were deposited in GenBank. Draco blanfordii Boulenger, Japalura tricarinata (Blyth), Calotes calotes (Linnaeus), and Mantheyus phuwuanensis were chosen as outgroups based on Schulte et al. (2004) and Che et al. (2020). Homologous and outgroup sequences were obtained from GenBank (Table 1).
Sequences were aligned using ClustalW (Thompson et al. 2002) with default parameters in MEGA X (Kumar et al. 2018) with default parameters. The genetic divergence (uncorrected p-distance) between species was calculated in MEGA X (Kumar et al. 2018). The best substitution model GTR+F+I+G4 was selected using the Akaike Information Criterion (AIC) in ModelFinder (Kalyaanamoorthy et al. 2017). Bayesian Inference was performed in MrBayes 3.2.6 (Ronquist et al. 2012). Two runs were performed simultaneously with four Markov chains Monte Carlo (MCMC) simulations, three hot and one cold. The chains were run for 10,000,000 generations and sampled every 1,000 generations. The first 25% of sampled trees were discarded as burn-in. Chain stationarity and run parameter convergence were checked in Tracer 1.7.1 (Rambaut et al. 2018). Maximum Likelihood analysis was performed in raxmlGUI 2.0 (Silvestro and Michalak 2012) and nodal support was estimated by 1,000 rapid bootstrap replicates.
Measurements were taken with a digital caliper to the nearest 0.1 mm, except tail length (TAL) which was measured using a string and a ruler. Morphological terminology followed Schulte et al. (2004) and Che et al. (2020). Paired meristic characters are given as left/right. Morphometric characters included: FLL, forelimb length, from forelimb insertion to end of fourth finger; HL, head length, distance from tip of snout to rear border of right angle of jaw; HLL, hindlimb length, from thigh insertion to end of fourth toe; HW, head width, widest point in the temporal region, anterior to the tympanum; InfL, Infralabials, number of enlarged scales bordering left and right margin of lower lip (not including mental scale); IOD, interorbital distance; MB, number of scales around midbody; NC, number of nuchal crest spines, enlarged mid-dorsal crest scales from posterior portion of the head to the posterior portion of the nape; OD, orbit diameter, measured from the posterior to the anterior edge of the orbit; SEL, snout to eye length, measured from the anterior edge of orbit to the tip of snout; SupL, supralabials,

Results
The generated gene sequences were 1013 bp in length. The topologies derived from Bayesian Inference and Maximum Likelihood analyses were consistent (Fig. 1). The specimens from Sagaing, Myanmar, formed a separate clade sister to a specimen (CAS 221515) from Kachin, Myanmar, with strong support, and they together formed a clade sister to the specimens from Tibet, China, with weak support. The average uncorrected pairwise distance (p-distance) between the sequences of the specimens from Sagaing, Myanmar, and the sequences of the specimens from Tibet, China, is 23.5% (Table 2). The gular regions of both sexes in life are white with long blue or bluish black stripes for the specimens from Tibet, China (Che et al. 2020), however, the gular regions of the males in life are bright yellow with short black stripes for the newly collected specimens from Sagaing, Myanmar ( Fig. 6A-C), and the gular region of the female of the newly collected specimens in life is greyish white with no stripe (Fig. 6D).
Morphological comparisons are presented in Table 3. The specimens from Tibet, China, largely agree with the type specimen (Holotype ZMB 5004) of Ptyctolaemus gularis, there is a clear overlap in morphology of the specimens from Tibet, China, and the type specimen (Holotype ZMB 5004) of P. gularis. However, the ratios of FLL/SVL (0.48-0.52) and HLL/SVL (0.90-0.94) of the   Etymology. The specific epithet refers to the Chindwin River, the new species was discovered in its basin.
Diagnosis. Body size medium, SVL 45.6-83.5 mm, slender, tail long, TAL/SVL 2.24-2.47, limbs long, FLL/ SVL 0.48-0.52, HLL/SVL 0.90-0.94. Tympanum concealed. Nuchal crest undeveloped, low and flat. Dorsal scales heterogeneous. The gular region is bright yellow with two or three short black stripes in males and the gular region is greyish white with no stripe in females.
Description of the holotype. Adult male, habitus slender, slightly compressed laterally, SVL 83.5 mm, TAL 206 mm, TAL/SVL 2.47. Rostral scale crescent, width approximately 3.1 times of height, five scales in contact with rostral edge including the first supralabials; nasal octagonal, separated from rostral by one scale, seven scales in contact with nasal on each side; canthus rostralis sharp, composed of 12 enlarged scales on each side; scales on snout irregular in shape and size; an inverted Y-shaped pattern on the middle of the snout; ten supralabials left and eleven right; orbit 7.2 mm in horizontal diameter; distance from anterior edge of orbit to nostril 5.9 mm, and 9.6 mm to tip of rostral scale; tympanum concealed, covered with smooth, slightly imbricate scales, equal in size to adjacent scales; three enlarged scales posterior and horizontal to orbit, keeled and elevated; temporal area with three enlarged, pointing scales. Mental scale triangular, wider than long, slightly narrower than rostral; mental followed by an infralabial on either side and two postmentals in contact with first infralabials; posterior to postmentals are four chin shields on each side that run parallel to infralabials, anterior portion of first chin shield touching first infralabial, remaining portion of first chin shield and following two chin shields sepa-rated from infralabials by one scale row; nine infralabials left and ten right. Gular scales anterior to gular pouch small, rounded, imbricate and slightly mucronate; scales of gular pouch slightly keeled, becoming larger toward center. Nuchal crest poorly developed, composed of 29 conical scales. Scales from angle of jaw to shoulder with feebly keeled, imbricate scales, interspersed with three large pointed scales; small oblique curved fold in front of shoulder. Dorsal scales keeled, imbricate, pointing backwards; mid-dorsal scale row strongly keeled and equal in size to bordering dorsal scales, one row of discontinuous, strongly keeled, enlarged, scales net to the vertebral ridge, separated from the mid-dorsal scale row by three scale rows; lateral scales heterogeneous, majority of scales much smaller than dorsals, feebly keeled, imbricate, interspersed with enlarged strongly keeled scales; lateral scales pointing backwards and downwards. Ventral scales larger than lateral scales, approximately equal in size to largest dorsal scales, strongly keeled, imbricate, mucronate, pointing backwards. Limbs slender, covered dorsally with strongly keeled, imbricate, mucronate scales; ventral surface of limbs with smaller feebly keeled, imbricate, scales; relative length of digits: IV > III > II > V > I, relative length of toes: IV> III > V > II> I; 33/35 subdigital lamellae under fourth toes. Tail slightly compressed laterally, covered with homogenous, strongly keeled, imbricate, scales.
Color of holotype in life. Dorsal head light brownish grey, several indistinct transverse brown stripes on the anterior portion dorsally, two black spots on the top of the head; lateral head grayish brown, many indistinct radial black stripes around the eye, the two below and posterior to the eye extended to the corner of the mouth; ventral head grayish brown, gular region bright yellow with two horizontal parallel black stripes on the center of each side of the gular pouch, the lengths of the two stripes almost equal and no more than half length of the gular pouch, the two stripes slightly connected up and down, separated by a indistinct thin yellow stripe. Dorsal neck and body light brownish grey with some indistinct slightly symmetrical brown patches beside the vertebral ridge; lateral neck and body flank brownish grey with indistinct purple brown reticulated pattern; the middle of the ventral body white, ventrolateral region purple brown; dorsal side of the limbs brownish grey with some indistinct brown bands; ventral side of the limbs scattered with white and brown. Anterior portion of the tail brownish grey, middle portion of the tail with black and white rings, posterior portion of the tail almost uniform black.
Color of holotype in preservative. Dorsal head, neck, body, and limbs light brown, transverse stripes on dorsal head became more indistinct, the patches and bands on dorsal body and limbs almost invisible; lateral head light brown, the radial stripes around the eye still visible; lateral side of the neck grey; lateral side of the body brown, the reticulated pattern almost invisible; ventral side of the head, body, and limbs yellowish white, bright yellow on gular region faded; brown and white rings on the tail.  Variations. The paratypes resemble the holotype in morphometric characteristics, the paratypes merely having smaller body sizes than the holotype. For coloration, the male paratype (SEABRI 2019120031) has an almost uniform light brownish grey dorsal color with no stripes or patches, and the gular region has three black stripes, of which the upper two are parallel and the lower two intersect; the female paratype (SEABRI 2019120046) has more obvious patches on the back and a more obvious reticulated pattern on the body flanks, and the gular region is greyish white with only some tiny black spots but no stripe.
Natural history. All specimens were collected on withered leaves on the sides of small roads in forests at night while they were asleep. The collection sites are near a tributary of the Chindwin River, but there was no water body within hundreds of meters of the collection sites. There are both primary and secondary forests around the collection sites, and the woods are dense. Other reptiles also found at the site included Bungarus fasciatus (Schneider), Cyrtodactylus mombergi Grismer, Wood, Quah, Thura, Herr & Lin, C. russelli Bauer, Psammodynastes pulverulentus (Boie), and Ptyas korros (Schlegel).
Ptyctolaemus chindwinensis sp. nov. can change its body color within a certain range like most other draconine lizards. When they rest on withered leaves at night, their bodies are pale colored with almost no stripes or patches; when they active are on the ground during the day, their bodies become darker with obvious stripes or patches (Fig. 7). The colouration provides camouflage in their preferred habitat.
Comparisons. Ptyctolaemus chindwinensis sp. nov. can be distinguished from P. gularis and Ptyctolaemus aff. gularis from Tibet, China, by having relatively longer limbs (FLL/SVL 0.48-0.52 vs 0.35-0.45 in P. gularis and Ptyctolaemus aff. gularis, HLL/SVL 0.90-0.94 vs 0.71-0.78 in P. gularis and Ptyctolaemus aff. gularis), and live colorations of gular region (bright yellow with two or three short black stripes on the center of each side of the gular pouch in males and greyish white with no stripe in females vs. white with three or four long blue or bluish black stripes which occupy most portions of the gular pouch in both sexes in Ptyctolaemus aff. gularis).

Discussion
In Schulte et al. (2004), there is one specimen (CAS 225592) which was identified as Ptyctolaemus gularis, and its collection site is very close to our new collection site, both in Htamanthi Wildlife Sanctuary, Sagaing, Myanmar. According to the morphological data of this specimen in Schulte et al. (2004), this specimen closely resembles Ptyctolaemus chindwinensis sp. nov. So, we consider this specimen to be conspecific with Ptyctolaemus chindwinensis sp. nov.
In Schulte et al. (2004), there are two specimens (USNM 123425 and BMNH 1974.846) which were also identified as Ptyctolaemus gularis, and their morphological characteristics are very close to P. gularis (Table 3). Geographically, the first specimen (USNM 123425) was collected in northeastern India, and the collection site of the second specimen (BMNH 1974.846) was given as "Myanmar, N Chengyang", but its exact location was uncertain. Through querying the website of the Natural History Museum (British Museum), we found that the voucher 1974.846 corresponds to two different collection sites, namely N'Changyong and N'Chongyong respectively. Therefore, the collection site of this specimen is uncertain. Since there are no molecular data of these two specimens, we temporarily regard these two specimens as Ptyctolaemus cf. gularis.
For the other specimens identified as Ptyctolaemus gularis in Schulte et al. (2004), most of which were collected from Kachin state, Myanmar. The morphological data of this specimens are close to Ptyctolaemus chindwinensis sp. nov., but the numbers of nuchal crest spines of these specimens are fewer than Ptyctolaemus chindwinensis sp. nov. (Table 3). Genetically, one sequence of a specimen (CAS 221515) from Putao, Kachin, Myanmar, and the sequences of Ptyctolaemus chindwinensis sp. nov. formed sister groups with strong supports (Figure 1). The average uncorrected pairwise distance (p-distance) between the sequence of the specimen (CAS 221515) from Putao, Kachin, Myanmar, and the sequences of Ptyctolaemus chindwinensis sp. nov. is 8.1% (Table 2). Therefore, we consider that this population in Kachin, Myanmar, represents an unnamed species.