The Sanjiangyuan region of Qinghai-Tibetan Plateau is recognized as a biodiversity hotspot of alpine mammals but a barren area in terms of amphibians and reptiles. Here, we describe a new pit viper species, Gloydius rubromaculatus sp. n. Shi, Li and Liu, 2017 that was discovered in this region, with a brief taxonomic revision of the genus Gloydius. The new species can be distinguished from the other congeneric species by the following characteristics: cardinal crossbands on the back, indistinct canthus rostralis, glossy dorsal scales, colubrid-like oval head shape, irregular small black spots on the head scales, black eyes and high altitude distribution (3300-4770 m above sea level). The mitochondrial phylogenetic reconstruction supported the validity of the new species and furthermore reaffirms that G. intermedius changdaoensis, G. halys cognatus, G. h. caraganus and G. h. stejnegeri should be elevated as full species. Gloydius rubromaculatus sp. n. was found to be insectivorous: preying on moths (Lepidoptera, Noctuidae, Sideridis sp.) in the wild. This unusual diet may be one of the key factors to the survival of this species in such a harsh alpine environment.
The Sanjiangyuan region (the Source of Three Rivers region) lies in the southern part of Qinghai Province, along the eastern part of Qinghai-Tibetan Plateau with an area covering 0.36 million km2. It encompasses the headwaters of the Yellow River, the Yangtze River, and the Mekong River. The Sanjiangyuan region is rich in biodiversity of alpine mammals, such as snow leopards (Uncia uncia), wild yaks (Bos grunniens) and Tibetan antelopes (Pantholops hodgsonii) (Shen and Tan, 2012). However, the herpetological diversity here is quite low due to the harsh conditions for sustaining life (e.g. low temperatures, low oxygen levels, and intense solar radiation). To date, only ten reptile species have been recorded (Li et al., 1989; Zhao et al., 1998), within which are three snake species (Gloydius strauchi, Gloydius cognatus and Elaphe dione). In this study, a new species of Asian pit viper (Gloydius) has been discovered along the Tongtianhe River at the elevation up to 4770 m.
Asian pit vipers are small venomous snakes distributed mainly in Asia. They are widely recognized to be one of the most successfully evolved snake groups, which radiated into various habitats, such as subfrigid forests (G. halys), alps or plateaus (G. strauchi, G. himalayanus and G. monticola), islands (G. shedaoensis and G. changdaoensis) and deserts (G. cognatus) (Hoge and Romano-Hoge, 1981). Asian pit vipers hold the record for the highest altitude distribution within venomous snakes (G. himalayanus, above 4880 m; Sharma et al., 2013) and the highest population density within the suborder of Serpentes (G. shedaoensis, 20 281 snakes within 0.73 km2, express as about 0.028/m2, based on the pit viper population survey conducted by the Snake Island National Nature Reserve; Li et al., 2007). Taking advantage of their heat sensitive pits, most Asian pit vipers tend to prey on small endotherms. As well, some of them are reported to be insectivorous (Gloyd and Conant, 1990; Zhao, 2006).
- 1.Gloydius halys-intermedius complex (G. halys, G. intermedius, G. changdaoensis, G. rickmersi, G. cognatus, G. stejnegeri and G. shedaoensis).
- 2.Gloydius blomhoffii complex (G. blomhoffii, G. brevicaudus, G. tsushimaensis and G. ussuriensis).
- 3.Gloydius strauchi complex (G. strauchi, G. monticola, G. qinlingensis G. liupanensis and G. himalayanus).
The members of Gloydius strauchi complex are generally described as an alpine group with 21 dorsal scale rows (except for 19 rows in G. monticola) and three palatine teeth, distributed along the north of the Hengduanshan Mountains (Zhao and Yang, 1997). The taxonomy on this group is still controversial: some regard G. monticola as a full species (Gloyd and Conant, 1990; Wagner et al., 2016), while others suggest that G. monticola should be attributed to one of the subspecies of G. strauchi and deny the validity of qinlingensis and liupanensis (Zhao, 1998; Zhao, 2006). Xu et al. (2012) conducted the molecular phylogeny of the genus Gloydius and suggested that G. qinlingensis and G. liupanensis should be regarded as full species. However, the topological structures of the maximum likelihood (ML), maximum parsimony (MP) and Bayesian inference (BI) trees differ significantly, with primary differences indicated by the positions of G. qinlingensis, G. strauchi and G. liupanensis. Thus, a further investigation is required to clarify the taxonomic relationship between the different taxa of this complex.
Material and methods
We examined preserved specimens from Chengdu Institute of Biology (CIB), Northwest Institute of Plateau Biology (NWIPB) and Kunming Institute of Zoology (KIZ). Newly obtained specimens collected were preserved in 75% ethanol and deposited at Institute of Zoology (IOZ), Northwest Institute of Plateau Biology (NWIPB) (table 1).
Details of the molecular samples for this study.
IVPP: Institute of Vertebrate Paleontology and Paleoanthropology; CIB: Chengdu Institute of Biology; IOZ: Institute of Zoology; NWIPB: Northwest Institute of Plateau Biology; SYNU: Shenyang Normal University; KIZ: Kunming Institute of Zoology. (IVPP, CIB, IOZ and NWIPB are belonging to Chinese Academy of Science.)
Measurements were taken with vernier calliper (Guanglu, 0-200 mm, Made in China). Snout-vent length (SVL), tail length (TL) and total length (TTL = SVL + TL) are measured to the nearest 0.1 mm; head length (HL, from the tip of snout to the posterior margin of mandible), head width (HW, from the posterior jaw, which is the widest part of the head), head height (HH, the highest part of the head), eye diametre (ED, horizontal distance), interorbital space (IOS), and internasal space (INS). We took counts of supralabials (SPL), infralabials (IFL), dorsal scales (DS), ventral scales (V) and subcaudal scales (Sc). Dimensions and scale data are listed in table 2.
Dimensions (mm) and scale data of the specimens of Gloydius snakes for this study.
X-ray micro-computerized tomography
The scanning was carried out with the 225 kV micro-computerized tomography (developed by the Institute of High Energy Physics (IHEP), Chinese Academy of Sciences (CAS)) at the Key Laboratory of Vertebrate Evolution and Human Origins, CAS. Specimens were scanned at 140 kV with a flux of 100 μA at a resolution of 42.3 μm per pixel using a 360° rotation with a step size of 0.5° and an unfiltered aluminium reflection target. A total of 720 transmission images were reconstructed into the 2048 × 2048 matrix of 1536 slices using a two-dimensional reconstruction software developed by IHEP, CAS.
Specimens were fixed in 95% ethanol or 10% formalin. Shed skin and scale tissues were preserved in 98% ethanol for molecular study. Genomic DNA was extracted with Miniprep Kit (Axygen). Samples included in this study are listed in table 1.
Molecular phylogenetic analysis
Four fragments of mitochondrial genome are specifically amplified in this study: 12s rRNA (12S), 16s rRNA (16S), cytochrome b (Cytb) and NADH dehydrogenase subunit 4 (ND4). The standard PCR protocol is performed in 20 μl of reactant with at least 20 ng of template DNA and 10 pmol of primers. The PCR conditions: initial denaturation for 3 min at 94°C, followed by 35 cycles: denaturation at 94°C for 30 s, 30 s of annealing at different temperatures (56°C for ND4, 48°C for Cytb, 54°C for 16S, and 52°C for 12s), and then elongation at 72°C for 60 s, then finalized with elongation step of 10 min at 72°C. See online supplementary table S1 for primer sequences and modifications of the standard PCR protocols.
Sequencing is conducted by Beijing Genomics Institute. Sequence data are uploaded to GenBank and are available under accession numbers showed in table 1. Data are aligned by Mega 6.0 (Tamura et al., 2013). A dataset with a total of 3129 basepairs containing 44 specimens is analyzed in this study: 42 of the samples are belonging to the genus Gloydius, while three of which are identified as G. rubromaculatus sp. n. (Y2, Y4 and Y5). Two samples, Deinagkistrodon acutus and Trimeresurus sichuanensis are recognized as outgroups. With respect to the different evolutionary characteristics of each molecular marker, the dataset is split into 8 partitions by gene and codon positions, then combined into 4 ones taking advantage of PartitionFinder 2.1.1 (Lanfear et al., 2012) (online supplementary table S2). General time-reversible (GTR) model, the most probable substitution model for the corrected ND4 p-distance matrix is calculated by PAUP 4.0 (https://people.sc.fsu.edu/~dswofford/paup_test/). Bayesian phylogenetic analyzis is performed with MrBayes 3.1.2 (Ronquist et al., 2011). All searches consist of three heated chains and a single cold chain. Three independent iterations each comprising two runs of 20 000 000 generations are performed, sampling every 1000 generations, parameter estimates are plotted against generation, The first 25 percents of the samples are discarded as burnin. Maximum likelihood analyzis (sharing the same partition as Bayesian phylogenetic analyzis) is conducted in RAxML 7.0.4 (Stamatakis, 2006), with 1000 fast bootstrap repeats (see Part 3 of the supplementary material for the command blocks).
Viperidae Gray, 1825
Gloydius Hoge and Romano-Hoge, 1981
Gloydius rubromaculatus sp. n. Shi, Li
and Liu, 2017
ZooBank accession: BF478EA7-C2D6-4F0E-B133-F4D624C3EA2A
The specific name of the new species is made up of the Latin word “rubro” (red) and “maculatus” (spot), indicating cardinal crossbands on the body. Its common name is suggested to read “Red-spotted alpine pit viper” or “Tongtianhe pit viper” in English and “Hóng Bān Gāo Shān Fù ()” in Chinese.
Holotype and paratypes
Holotype: IOZ032317, adult male, collected by Jingsong Shi (JS) and Xi’er Chen (XC) from the mid-upper reaches of the Tongtianhe River, Qumarleb, Qinghai Province, on 9 July 2016. Paratypes: NWIPB790058 (allotype, adult female), IOZ 032318 (adult male), JS1607Y3 (subadult female), NWIPB790056 (adult female), NWIPB 630064 (adult female), and NWIPB0512 (adult male). Referred specimens: NWIPB 17092:1, from Jyekundo, Qinghai Province; NWIPB 790 060-790 067, from Zhiduo, Qinghai Province. See table 2 for detailed information. Note: some of the specimens of the old species are labelled with “G. strauchi” which are identified as G. rubromaculatus sp. n. in this study.
The above mentioned specimens were identified as the members of Gloydius judging on their small body size, bilateral pits and divided subcaudal scales (Hoge and Romano-Hoge, 1981), while differ from other congeneric species in the following characteristics: 1. two rows of cardinal crossbands on the back, regularly spaced along the body; 2. glossy dorsal scales, compared to matte scales in other members of Gloydius; 3. colubrid-liked dome shaped head in lateral view and oval shaped in dorsal view, compared to flat-shaped head in lateral view and triangular in dorsal view in other Gloydius; 4. irregular small black spots dispersed on the head scales; 5. inconspicuous canthus rostralis; 6. dark brown eyes with black pupils (figs 1 and 2).
The new species is distinct from species in the Gloydius blomhoffii complex by having three palatine teeth (versus four palatine teeth); from the species in the Gloydius halys-intermedius complex by having 21 rows of mid-body dorsal scales (versus 23-rows). Thus, the new species is suggested to belong to the Gloydius strauchi complex.
Within the Gloydius strauchi complex, G. rubromaculatus sp. n. is distinct from G. monticola by its 21-rows mid-body scales (versus 19-rows mid-body scales and 6 supralabials); from G. himalayanus by its indistinct canthus rostralis (versus very distinct canthus rostralis); from G. qinlingensis and G. liupanensis by its oval head (versus triangular head), regular crossbands (versus irregular crossbands) and the lack of the white line on each side of body (versus obvious white line); from G. strauchi by its brownish black eyes (versus light brown eyes), and two rows of regular round crossbands (versus four irregular longitudinal strips).
See table 3 and fig. S1 for the morphological comparisons within the Gloydius strauchi complex.
The validity of this new species is also supported by molecular genetics. Three samples of G. rubromaculatus sp. n. (Y2, Y4 and Y5) form a strongly supported monophyletic group. In addition, the corrected ND4 p-distance between G. rubromaculatus sp. n. and other species is greater than the ones between most of the other species (8.6-12.4 percent, see online supplementary table S3).
Description of the holotype
Adult male, a small slender pit viper with a total length of 554 mm (body length 473 mm and tail length 81 mm), preserved in 75% ethanol with its bilateral hemipenis extruded.
The head is spoon-shaped in dorsal view and dome-shaped in lateral view, 24.6 mm in length, 15.4 mm in width and 7.4 mm in height. Rostral scale slightly turns up to the upper side of the head. Seven bilateral supralabials are presented: the second one smallest, not reaching the pit; the third and fourth largest, with the former extending to the bottom of orbit. Three preoculars, two postoculars, and two rows of temporals (2 + 4). Ten infralabials are on the left while nine on the right, of which the first pair contact behind the mental, the second and third ones meet the chin shield. Mental groove is made up of paired parallelogram chin shields, which extend to the mental. The canthus rostralis are not distinct. The head has a clear border with the neck. The eyes are brownish black, with vertical spindly, light brown margined black pupil. A thick, black bordered yellowish red cheek stripe extends from the posterior side of each eye (separated from the orbit by the inferior postocular) to the first pair of neck crossbands, throughout the inferior temporal and the last two supralabials. Lots of irregular, uneven-sized black spots dispersed on most of the head scales, except for the temporals (figs 1 and 2). Mouth lining is pink in life. The anterior part of the tongue is black while the base is pink.
Detailed comparison between G. rubromaculatus sp. n. and the remaining species of the Gloydius strauchi complex.
The body colour is light greyish yellow, with a column of complete regular round cardinal crossbands on each side of the back, black bordered and shallow inside, in pairs or interfaces, about three to four scales in length, and four to six scale rows in width, separated by blank areas one or two scales in width, extending down to lateral one or two ventral scales. Crossbands range from the neck to the tip of the tail, 49/46 on body and 12/16 on the tail (left/right). A range of triangular or irregular black ventrolateral blotches (irregular one is made up of the adjacent two or three ones) range on the boundary between back scales and ventrals on each side of the body, divided by one or two ventral scales (fig. 1). Ventral scales are opalescent, with mottled irregular black blotches. The black blotches are concentrated on the middle of scales. The edge and the junction of bilateral subcaudals are trimmed in black thread. One semicircular black blotch is present on the lateral edge of each subcaudal scale. A black stripe is present on each side of the boundary of the two bilateral ventrals. The tip of the tail is black and bony.
Dorsal scales are in 21-21-15 rows (reduce from 21 to 15 rows beginning with the ventral 92/93), keeled (excluding the ones bordering the ventral scales) and glossy. Ventrals 158 (excluding three preventral scales). Anal plate is complete. Subcaudals divided in pairs (43 pairs).
See table 2 for the detailed measurements of the specimens examined.
The frontal of G. rubromaculatus sp. n. is inverted triangular in dorsal view, contrast to round in G. strauchi, and “T”-shaped in G. shedaoensis. The lateral margin of nasal is rounded, without any process, while in G. strauchi the lateral process is distinct. The fang is quite short, approximately one third length of the ectoptery (versus approximately a half-length of the ectoptery in G. halys-intermedius complex and G. blomhoffii complex; Gloyd and Conant, 1990). Six replacement fangs behind each primary fang, three palatine teeth, 12 pterygoid teeth and 11 alvenlus teeth on each side. The upper edge of postfrontal is in contact with the posterior-lateral process of the frontal, versus separated in G. strauchi (Gloyd and Conant, 1990). The quadrate is quite slender, about 1.2 times as long as the squamosal, versus almost equal to the squamosal in length in G. strauchi (fig. 3).
The hemipenes of G. rubromaculatus sp. n. are generally similar to those of G. strauchi but differ by the podgier spines. Eight subcaudals in length, forked for two subcaudals. Small stubby spines range from the basal to the distal side of the organ, without any enlarged spines (versus three to five enlarged spines on the base in Gloydius halys complex; Gloyd and Conant, 1990). The spines gradually increase in size distally. More spines occur on the asulcate side on than the sulcus side. Spines and sulcus merge with the calyces on the distal side. No giant spines present on the basal of the hemipenis as Gloydius halys-intermedius complex (fig. 4).
Infraspecific morphological variation
Some morphological variations are present within the specimens for this study: some of the G. rubromaculatus sp. n. have large, complete scarlet crossbands, while other have small irregular ones (e.g., Y4), or large brown regular crossbands (e.g., Y7) (fig. 1). Body colour is mostly grayish white, sometimes brownish yellow (Y9). Spots are mostly present on the head scales, but some have none or only a few small ones. Most have seven supralabials (rarely eight) and 10 infralabials (rarely 9 or 11, table 2). Ventrals range from 146 to 158 in males (mean 152.3, ) while 153 to 163 in females (mean 159.5, ). Subcaudals range from 41 to 43 in males (mean 42.3, ), and 35 to 43 in females (mean 40, ). See table 2 for the detailed measurements of the specimens examined.
Phylogenetic and phylogeographic analysis
The validity of the new species is supported by phylogenetic analyzis, the topological structure of the maximum likelihood (ML) and Bayesian inference (BI) trees are approximately identical except for the clade of G. stejnegeri, G. rickmersi and G. caraganus (fig. 5). All members of Gloydius perform as a monophyletic group. The clade of the new species from along the Tongtianhe river (Y2, Y4 and Y5) performs as a strongly supported monophyletic group (Clade A, in red), and constitutes sister groups with the clade of G. monticola from Yunnan (Clade B). Despite the geographic proximity with G. strauchi, G. rubromaculatus’s group does not perform as sister groups with G. strauchi (Clade C). Thus, the new species is more closely related to G. monticola than to G. strauchi. The samples of G. qinlingensis (Clade D) and G. liupanensis (Clade E) from near the Yellow River do not present as sister groups, moreover, the p-distance between them is relatively greater (7.2 percents for ND4) than the ones between other congeneric species, thus they should be regarded as distinct species. In general, the samples of the Gloydius strauchi complex, including qinlingensis (Shaanxi), liupanensis (Ningxia), monticola (Yunnan), strauchi (Sichuan) and rubromaculatus sp. n. (Qinghai) do not completely constitute a monophyletic group as the Gloydius halys-intermedius complex (Clade F) and Gloydius blomhoffii complex (Glade G).
The validity of G. rickmersi Wagner, Tiutenko, Mazepa, Simonov, Borkin, 2016 is reconfirmed in this study. On the other hand, the populations distributed in Jiaodong Peninsular (Shandong Province) had long been regard as a subspecies of G. intermedius (G. i. changdaoensis Li, 1999), however, the molecular phylogeny shows a significant distance between G. i. changdaoensis and G. intermedius (p-distance: 5.4 percent), the clade of changdaoensis (Z1 and C1) does not constitute as sister groups with G. intermedius (SX1, 22, Q4 and QS002), but firstly separates from the remaining taxa of the G. halys-intermedius complex. Therefore, G. i. changdaoensis should be regarded as a full species, but not a subspecies of G. intermedius. Additionally, the samples of the different subspecies of G. halys (e.g. caraganus, stejnegeri and cognatus) do not performed as a monophyletic group, and none of them performed as sister groups with G. h. halys, allowing for the high p-distance between the different subspecies, this study fully shares the taxonomy of Shi et al. (2016), suggesting that G. changdaoensis, G. caraganus, G. stejnegeri and G. cognatus should be elevated as full species. (See Orlov and Barabannov, 1999 and Shi et al., 2016 for the detailed comparisons between different subspecies of G. halys.)
We checked the faeces samples of four snakes (Y1, Y2, Y5 and Y6). The faeces contain almost entirely the debris of moths, such as undigested bodies, siphoning mouthparts, wings, periopods and arms of the furcas (supplementary fig. S2). One of the moths could be identified as Sideridis sp. (female). No hairs, bones or feathers of birds or mammals can be found in the faeces yet. Two undigested neonate zokors (Eospalax fontanierii) are found in another specimen’s stomach (NWIPB 630064). Two juvenile pit vipers are observed to prey on moths and pink mice in captivity (Y1 and Y5).
Two hypotheses can account for the moth debris: they may be from the moths preyed and excreted directly by snakes; or alternatively, these items were secondarily ingested along with the primary prey items, frogs and lizards, which are more routinely found to be insectivores compared to snakes. However, one of the snakes (Sample Y3) was observed to vomit a whole undigested moth, which could confirm the former hypothesis.
Distribution and habitat
Gloydius rubromaculatus sp. n. is distributed mainly along the Tongtianhe River, in the Sanjiangyuan region of Qinghai Province. Additionally, G. rubromaculatus sp. n. is also found in Tibet (Tongpu village, Jiangda Country) and Sichuan (Shiqu Country) (table 1 and supplementary fig. S3). The distribute altitude ranges from 3300 to 4770 m. G. rubromaculatus sp. n. holds the highest snake distribution report within Chinese venomous snakes, and the second highest one all over the world, next to G. himalayanus, which can occur up to 4880 m (Sharma et al., 2013). Gloydius rubromaculatus sp. n. tends to stay in the mountain passes, sandy riversides, sunny slopes, bushes and shales (fig. 6).
Gloydius rubromaculatus sp. n. is protected under the conservation regulations of the Sanjiangyuan National Reserve. Traditional Tibetan culture also offers alternative knowledge and perspectives that facilitate the environmental conservation throughout the region (Shen and Tan, 2012). Due to the faith of native Tibetans, the animals there, including snakes, are fully respected and well protected.
In the past decades, the pit vipers with 21-rows mid-back scales from Qinghai-Tibet Plateau have been identified as G. strauchi (Zhao and Yang, 1997; Zhao, 2006), while according to Pope’s study (1935), the type locality of G. strauchi is restricted to Tungngolo (between Litang and Kangting, near Xindu, Sichuan) but redefined as Da-Tsian-lu (Kangding, Sichuan, locality of the lectotype ZISP 8534) by Orlov and Barabanov (2000). Thus, we recognize sample G3 as topotype of G. strauchi. However, there are both morphological and genetic differences between G. rubromaculatus sp. n. and G. strauchi (p-distance: 9.0 percent). The two species did not recover as sister groups in the phylogenetic trees. Thus, G. rubromaculatus sp. n. is recognized as a new taxon in this study.
This study confirmed that G. rubromaculatus sp. n. is able to prey on moths in the wild, and has the potential to prey on small mammals. More attention should be paid to its unusual diet, which may be attributed to the lack of prey in the alpine habitat. It is still unknown whether the moths-preying behaviour is seasonal. More data remain to be obtained over a longer duration of time (i.e., a proper diet study over an entire year), including adults, juveniles, and prey availability for the habitat. The reason why the pit vipers have a special preference for moths is unknown. A possible assumption would be that they are attracted by the smell of pheromone given off by the moths for sex attraction (Groot et al., 2006). These questions need further investigation.
Body article and figures: Jingsong Shi. Field work: Jingsong Shi and Xi’er Chen. Molecular experiments: Jingsong Shi, Gang Wang and Yihao Fang. Specimen’s measurement: Jingsong Shi, Li Ding and Gang Wang. Data analysis: Jingsong Shi, Mian Hou, Li Ding and Song Huang. Naming: Jingsong Shi, Pipeng Li and Jun Liu. All authors gave the final approval for publication.
This study is conducted with appropriate permissions (letter voucher: IHSYNU  08) for field survey and specimens’ collecting, and guidelines from the responsible authority, the Forest Department, Ministry of Forest and Environment, the People’s Republic of China. All tissues for DNA extracting are from ventral scales, shed skins or road-killed dead bodies in this study; no snakes were killed or vivisected.
This study is supported by Ministry of Science and Technology of China (2014FY210200), and National Natural Scientific Foundation of China (130204, 130201). We are grateful to Songchang Guo, Wenjing Li, Xiaocheng Chen (NWIPB), Dajie Gong, Gang Liang, Xiang Zhao and Thupten Gyaltsen for the assistances in specimen collection and examination; to Kevin Messenger, Yulong Li, Jinzhong Fu, David Cundall, Yunke Wu, Jiasheng Hao, Liping Dong, Zhiyong Yuan, Xuankun Li and Liqun Hao for the kindly language revisions and professional advices; to Tingting Zhang, Bin Wang (CIB), Yemao Hou and Yong Xu (IVPP) for the help with picture drawings and software operations; to Ding Ding (IOZ), Xiaoping Wang (Liaoning Snake Island National Nature Reserve), Jianfang Gao, Xiaoyu Zhu, Wei Xue, Jincheng Liu and Xinlei Huang for important samples; to Chaodong Zhu and Qingyan Dai for moth identification; to Vivek Sharma, Upadhyay, Deepak CK (Indiansnakes. org), Jiansheng Peng and Zhiyuan Tang for nice photographs.
Folmer O., Black M., Hoeh W., Lutz R., Vrijenhoek R. (1994): DNA primers for amplification of mitochondrial cytochromec oxidase subunit I from diverse metazoan invertebrates. Mol. Mar. Biol. Biotechnol. 3 (5): 294.
Gloyd H.K., Conant R. (1990): Snakes of the Agkistrodon Complex. A Morphologic Review. Contributions to Herpetology, No. 6. Society for the study of Amphibians and Reptiles, Oxford, OH, vi + 614 pp.; 52 pl.
Groot A.T., Horovitz J.L., Hamilton J., Santangelo R.G., Schal C., Gould F. (2006): Experimental evidence for interspecific directional selection on moth pheromone communication. PNAS 103 (15): 5858-5863.
Hoge A., Romano-Hoge S. (1981): Poisonous snakes of the world. Part 1: checklist of the pit vipers, Viperoidea, Viperidae, Crotalinae. Mem. Inst. Butantan 42 (43): 179-309.
Lanfear R., Calcott B., Ho S.Y.W. et al., (2012): PartitionFinder: combined selection of partitioning schemes and substitution models for phylogenetic analyzes. MBE 29 (6): 1695-1701.
Li J., Sun L., Wang X. et al., (2007): Influence of population distribution pattern of Gloydius shedaoensis Zhao on predatory rate. Journal of Snake 19 (1): 12-16.
Orlov N.L., Barabanov A.V. (1999): Analysis of nomenclature, classification, and distribution of the Agkistrodon halys-Agkistrodon intermedius complexes: a critical review. Russ. J. Herpetol. 6 (3): 167-192.
Orlov N.L., Barabanov A.V. (2000): About type localities for some species of the genus Gloydius Hoge et Romano-Hoge, 1981 (Crotalinae: Viperidae: Serpentes). Russ. J. Herpetol. 7 (2): 159-160.
Ronquist F., Huelsenbeck J., Teslenko M. (2011): Draft MrBayes version 3.2 manual: tutorials and model summaries. Distributed with the software from mrbayes. sourceforge.net/mb3.2_manual.pdf.
Sharma S., Pandey D., Shah K., Tillack F., Chappuis F., Thapa C., Alirol E., Kuch U. (2013): Venomous Snakes of Nepal. A Photographic Guide, 1st Edition, Feb.
Shen X., Tan J. (2012): Ecological conservation, cultural preservation, and a bridge between: the journey of Shanshui Conservation Center in the Sanjiangyuan region, Qinghai-Tibetan Plateau, China. Eco. Soc. 17 (4): 38.
Shi J., Yang D., Zhang W., Ding L. (2016): Distribution and infraspecies taxonomy of Gloydius halys-Gloydius intermedius complex in China (Serpentes: Crotalinae). C. J. Z. 51 (5).
Stamatakis A. (2006): RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinf. 22 (21): 2688-2690.
Swofford D.L. (2003): PAUP. Phylogenetic Analysis Using Parsimony (and Other Methods). Version 4. Sinauer Associates, Sunderland, MA, USA.
Tamura K., Stecher G., Peterson D., Filipski A., Kumar S. (2013): MEGA6: molecular evolutionary genetics analysis version 6.0. M. B. E. 30 (12): 2725-2729.
Wagner P., Tiutenko A., Mazepa G., Simonov E., Borkin L.J. (2016): Alai! Alai! – a new species of the Gloydius halys (Pallas, 1776) complex (Viperidae, Crotalinae), including a brief review of the complex. Amp. Rep. 37 (1): 15-31.
Xu Y., Liu Q., Myers E.A., Wang L., Huang S., He Y., Peng P., Guo P. (2012): Molecular phylogeny of the genus Gloydius (Serpentes: Crotalinae). A. H. R. 3 (2): 127-132.
Zhao E., Yang D. (1997): Amphibians and Reptiles of the Hengduan Mountains Region. The Comprehensive Scientific Expedition to the Qinghai-Xizang Plateau. Chinese Academy of Science, Beijing.
Associate Editor: Sylvain Ursenbacher.