The specimen deposited at Yibin University (YBU 230088) was collected in Beibeng Town, Medog County, southeastern Xizang, China (29°14′02″N, 95°10′38″E) (Fig. 1) on 14 August 2023 at an elevation of 1,431 m by Xiaoqi Mi. The snake was found on a tree near a road at 23:30 hours. Characters relating to scalation, color pattern, and body proportions were recorded from the preserved specimen in laboratory. Snout–vent length (SVL) and tail length (TL) were measured using a meter ruler to the nearest 0.5 centimeter, while all remaining measurements were taken using digital calipers to the nearest millimeter. Symmetric mensural head characters were taken on the right side unless unavailable (e.g. damaged), while meristic characters were recorded on both sides and reported in left/right order.

Figure 1. 

Map showing currently known localities of Lycodon gammiei.

Comparative data of other specimens of this species were taken from the literature (Blanford 1878; Mistry et al. 2007; Chettri and Bhupathy 2009; Wangyal 2013).

Genomic DNA was extracted from the liver tissue of the newly collected specimen using an Animal Genomic DNA Purification Kit (TIANGEN Bio-tech Co., Ltd, Beijing, China). Subsequently, a fragment of the mitochondrial gene cytochrome b (cyt b) was amplified using primers H14919 (5′-AACCACCGTTGTTATTCAACT-3′) and L16064 (5′-CTTTGGTTTACAAGAACAATGCTTTA-3′) (Burbrink et al. 2000). The polymerase chain reaction (PCR) products were purified and sequenced in both directions by Sangon Biotech Co., Ltd (Chengdu, China). The obtained sequences were manually edited using SeqMan in Lasergene v. 7.1 (DNASTAR, USA), and aligned using the ClustalW algorithm with default parameters in MEGA v. 7.0 (Kumar et al. 2016), followed by a visual inspection for minor manual adjustments. The DNA sequences were translated into amino acid sequences using MEGA v. 7.0 to verify sequence quality and detect any unexpected stop codons (Kumar et al. 2016). Furthermore, 80 additional sequences were downloaded from GenBank (Table 1).

Both Bayesian-inference (BI) and maximum-likelihood (ML) analyses were executed for the final dataset. Prior to analyses, the best-fit model of nucleotide substitution was selected for each partition (codon position) using Akaike Information Criterion (AIC) implemented in PartitionFinder (Lanfear et al. 2012). The BI analyses were conducted using MrBayes v. 3.2.2 (Ronquist et al. 2012). Searches consisted of three independent runs, each involving four Markov chains (three heated chains and one cold chain), with 10 million generations, sampling every 2,000 generations and with 25% of initial samples discarded as burn-in. Convergence was determined via effective sample size (ESS > 200) and likelihood plots against time using Tracer v. 1.7 (Rambaut et al. 2018). The resulting trees were combined to determine the posterior probabilities (PP) for each node based on a 50% majority-rule consensus tree. The ML trees were constructed in IQ-tree (Lam-Tung et al. 2015) using the GTRCAT model and the same partitioning scheme. In total, 1,000 Ultrafast bootstraps (UFB) topological replicates were performed for branch support assessment. Boiga cynodon (Boie, 1827) was selected as the outgroup following previous research (Guo et al. 2013).

Uncorrected genetic distance (p-distance) was calculated in MEGA v. 7.0 (Kumar et al. 2016).

Female, SVL 698 mm and TL 223 mm. Body elongated; head rather flattened; snout blunt. Rostral large, trapezoid; internasals much broader than long; prefrontals 3.0 mm in length, distinctly wider than long, extending beyond both sides and touching preocular and loreal; frontal peltate, 4.6 mm in length and 4.1 mm in width; parietals subrectangular, 7.9 mm in length and 4.2 mm in width. Nasals large, nostril located anteriorly and opening backward; loreal scale 1, long, nearly rectangular, failing to touch eye; preocular 1, postoculars 2; temporals 2+2+3. Supralabials 8, 1^st small, 3^rd, 4^th, and 5^th entering orbit, 6^th highest, 7^th largest; infralabials 10, first pair in contact, 1^st to 5^th in contact with anterior chin shields. Chin shield pairs 2, elongate, anterior pair slightly larger than latter pair. Dorsal scales 17-17-15 rows, scales weakly keeled, except for outermost several rows; scales reduced from 17 to 15 at 143^rd ventral position. Ventrals 228 (+ 1 preventral); cloacal plate entire; subcaudals 106, paired, dorsal scales of the tail reduced from 6 to 4 at 16^th subcaudal position.

Head black, with yellow spots or short lines on some shields. Large, yellow spots on each side of posterior part of head. Conspicuous yellow collar on neck. Supralabials and anterior infralabials light yellow with dusky margins. Body surrounded by alternating dusky and light-yellow rings with very irregular, crooked margins. Yellow rings on body totaling 43, first pale ring clear above, anterior dark patch not continuous across throat, remaining rings encircling body. Lower part of head and neck light yellow. On belly, across anterior part of body, dark rings only about half as broad as light-yellow rings, less difference above, dark rings near head much broader above than white rings. Yellow rings on tail totaling 21 (Fig. 2). Preserved specimen somewhat faded, with no yellow visible (Fig. 3).

Figure 2. 

General view of the studied specimen (YBU 230088) in life and its microhabitat a big tree trunk (by XQ Mi).

Figure 3. 

Views of the studied specimen (YBU 230088) in preservation. General dorsal (A) and ventral (B) views of specimen, dorsal (C), ventral (D) and lateral (E) views of head (by P Guo).

In total, 1,047 bp of sequence data from 84 samples were aligned, with the generated novel sequence deposited in GenBank (Table 1). No deletions, insertions, or stop codons were detected, indicating that unintentional amplification of pseudogenes was unlikely (Zhang and Hewitt 1996). The best-fit evolutionary models of the data were: GTR+I+G for the first codon position, HKY+I+G for the second codon position, and GTR+G for the third codon position.

The mtDNA-based BI and ML analyses depicted relatively consistent topologies, with slight disagreement in several shallow nodes (Fig. 4). Both analyses indicated that all putative species of Lycodon formed a highly supported lineage (100 PP and 84% UFB). The newly collected specimen formed a clade with L. fasciatus, L. gongshan Vogel & Luo, 2011, L. butleri Boulenger, 1900, and L. cavernicolus Grismer, Quah, Anuar, Muin, Wood & Nor, 2014 with high support (100 PP and 97% UFB). Nevertheless, it occupied a basal position in relation to this clade and did not exhibit monophyly with any individual member. Uncorrected p-distances among the species within this clade ranged from 7.2% (L. gongshan and L. fasciatus) to 12.9% (L. gammiei and L. cavernicolus), while genetic distances between L. gammiei and its congeners within this clade ranged from 10.2% to 12.9% (data not shown).

Figure 4. 

Bayesian 50% majority-rule consensus tree of Lycodon inferred from cyt b sequences analyzed using models detailed in the text. Posterior probabilities from BI analysis (>0.50) and Ultrafast bootstraps from ML analysis (>50%) are given adjacent to respective nodes for major nodes. Branch support indices are not given for most nodes to preserve clarity.

The authors have declared that no competing interests exist.

No ethical statement was reported.

This study was supported by the Second Tibetan Plateau Scientific Expedition and Research (STEP) Program (2019QZKK05010105), National Natural Science Foundation of China (32000308 and 32370486), and Yarlung Zangbo Grand Canyon National Nature Reserve Research and Monitor Program (Linzhi Forestry and Grassland Bureau)

Conceptualization: XM. Formal analysis: BL. Methodology: TZ. Resources: KG. Software: YW. Supervision: PG. Validation: LL. Writing – original draft: FS.

Fu Shu https://orcid.org/0000-0002-6082-8112

Bing Lyu https://orcid.org/0000-0001-5594-1543

Keji Guo https://orcid.org/0000-0001-7508-1173

Tong Zhang https://orcid.org/0009-0009-1492-585X

Xiaoqi Mi https://orcid.org/0000-0003-1744-3855

Li Li https://orcid.org/0009-0007-4149-6662

Yayong Wu https://orcid.org/0000-0003-2752-4085

Peng Guo https://orcid.org/0000-0001-5585-292X

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