Zoosyst. Evol. 100 (4) 2024, 1231-1241 | DOI 10.3897/zse.100.128567 Ate BERLIN Tscherskia ningshaanensis: A neglected species based on phylogenetic and taxonomic analysis of 7scherskia and Cansumys (Cricetidae, Rodentia) Haijun Jiang’*, Xuming Wang*, Yaohua Yang’, Xuan Pan*, Shaoying Liu®, Jigi Lut? School of Life Sciences, Zhengzhou University, Zhengzhou 450001, China Institute of Biodiversity and Ecology, Zhengzhou University, Zhengzhou 450001, China Sichuan Academy of Forestry, Chengdu 610081, China Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu 610213, China fF Gi oho https://zoobank. org/45 B95842-70F D-4B6C-895F-18E4DDAEODSF Corresponding authors: Jiqi Lu (lujq@zzu.edu.cn); Shaoying Liu (shaoyliu@163.com) Academic editor: Melissa TR Hawkins # Received 29 May 2024 # Accepted 6 August 2024 Published 2 September 2024 Abstract The greater long-tailed hamster is primarily distributed in North Korea, Siberia (Russia), and central and northern China, while the Gansu hamster is restricted to southern Gansu Province, China. The genera 7scherskia and Cansumys have each been considered monotypic. The taxonomic status of these two genera has long been debated, and the specific status of 7? ningshaanensis has also been contentious. Researchers have variously treated 7’ ningshaanensis as a subspecies of either 7! triton or Can. canus. In this study, we estimated the phylogeny, divergence times, species delimitation, and biogeographical history of 7? ningshaanensis by using one mitochondrial (CYT B) and three nuclear loci (GHR, IRBP, and RAG1) and compared the external and skull morphology variations between T. ningshaanensis and T. triton. The results showed that: 1) The genus Cansumys is a distinct genus in Cricetinae; 2) The notion that the genus 7scherskia is a monotypic genus is unsupported, 7? ningshaanensis and T: triton were identified within this genus; and 3) The formation of 7. ningshaanensis may have been driven by uplift of the Qinling Mountains. We conclude that T. ningshaanensis is a valid species within the subfamily Cricetinae. Key Words Classification, morphology, Mt. Qinling, phylogeny, Rodentia, 7scherskia Introduction The greater long-tailed hamster (7scherskia triton de Win- ton, 1899), family Cricetidae, Order Rodentia, 1s mainly distributed in North Korea, Siberia (Russia), and central and northern China, including the Provinces of Hebei, Shanxi, Shaanxi, Henan, Anhui, Jiangsu, Shandong, Hei- longjiang, Jilin, Liaoning, as well as Inner Mongolia and Being (Smith and Xie 2008; Wilson et al. 2017; Wei et al. 2021). Currently, a single species with five subspecies (7 t. triton, T: t. incanus, T. t. collinus, T. t. fuscipes, and T. t. nestor) has been identified in the genus 7scherskia (Smith and Xie 2008; Wilson et al. 2017). The species-level classification within 7scherskia has been controversial to date. In 1899, Cricetus triton was first described by de Winton (1899) from Shantung (= Shandong) Province, China. Based on morphological differences, geographical distribution, behavioral, and ecological characteristics, Thomas (1907, 1908) pro- posed that the classification of C. triton be revised from the genus Cricetus to Cricetulus, with subsequent studies recognizing one species and one subspecies within the ge- nus Cricetulus. In 1907, Cricetulus nestor was described from Korea (Thomas 1907); and in 1908, Cricetulus tri- ton incanus was described from Ko-lan-chow (= Kelan), Shan-si (= Shanxi), and Yen-an-fu (= Yan’an), Shen-si Copyright Jiang, H. et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 1232 (= Shaanxi) Provinces (Thomas 1908). Ognev (1914) de- scribed 7scherskia albipes from eastern Siberia, noting its larger body size and longer tail. In 1925, two new subspecies were recognized: Cricet- ulus triton collinus from the base of Tai-pei-shan (= Tai- baishan), Tsing-ling (= Qinling) Mountains, Shen-si (= Shaanxi) Province; and Cricetulus triton fuscipes from Peking, Chili (= Beijing, Hebei) Province (Allen 1925). Kishida (1929) described a new genus, Asiocricetus, from Korea, including Asiocricetus bampensis and Asiocricetus yamashinai. According to early taxonomic studies (e.g., Argyropulo 1933; Ellerman 1941), 7scherskia and Asioc- ricetus were regarded as subgenera of the genus Cricetulus. Furthermore, 7: albitpes, A. bampensis, and A. yamashinai were treated as synonyms of C. nestor (Argyropulo 1933). In 1934, Cricetulus triton meihsienensis was described by Ho (1934) from Mei-hsi (= Meixian), Shen-si (= Shaanxi) Province, and was subsequently treated as a synonym of C. t. collinus (Ho 1934; Wilson et al. 2017). In 1985, Cricetulus triton ningshaanensis was described by Song (1985) based on its smaller body size, tail length, and color of tail from Ningshaan, southern Shaanxi Province. Neu- mann et al. (2006) moved C. triton into T. triton and treat- ed 7scherskia as a monotypic genus based on molecular phylogenetic studies, a view accepted by many researchers (Smith and Xie 2008; Wilson et al. 2017; Wei et al. 2021). The classification status of Gansu hamster (Cansumys canus Allen, 1928) and C. t. ningshaanensis has been debated for a long time in China due to the significant morphological (particularly tail length and color) and distributional differences (Wang and Zheng 1973; Chen and Min 1982; Wang and Xu 1992; Gu et al. 2005). Can. canus was first described from Choni (= Zhuon1), south- ern Kansu (= Gansu) Province, China (Allen 1928). Lat- er, Ellerman (1941) treated Can. canus as a subspecies of Cricetulus triton, a view that was accepted by some researchers (Ellerman and Morrison-Scott 1951; Wang and Zheng 1973). However, Can. canus was considered a species placed in Cricetulus based on its geographical distribution and tail haircoat (Chen and Min 1982). Ross (1988) supported the idea that Cansumys was a valid ge- nus and Can. canus a separate species, which was sub- sequently widely accepted (Corbet and Hill 1992; Muss- er and Carleton 1993; Pavlinov et al. 1995a). In 1985, C. t. ningshaanensis was first described as a subspecies of C. triton (Song 1985), but some researchers placed C. t. ningshaanensis within Can. can. ningshaanensis, supporting Cansumys as a valid genus (Wang and Xu 1992; Lu 1997). Yang et al. (2003) claimed Cansumys was a separate genus based on karyotype analyses of Can. can. ningshaanensis from Ningshaan, Shaanxi Province. Furthermore, Gu et al. (2005) analyzed the external morphology of Can. canus and C. triton from Zhuoni, southern Gansu Province, and the results supported that Can. canus and C. triton were two distinct species and that the status of Can. can. ningshaanensis warranted further investigation. However, Liao et al. (2007) treated Can. canus as a synonym of C. triton and C. t. canus as a subspecies of C. triton based on molecular analysis of zse.pensoft.net Jiang, H. et al.: Phylogeny and taxonomy of Tscherskia and Cansumys specimens from Gansu Province and Ningxia. Since then, there have been no arguments regarding the classification status of C. triton and Can. canus (Smith and Xie 2008; Wilson et al. 2017; Wei et al. 2021). However, we know little about why 7: t. ningshaanensis was treated as a syn- onym of 7 ¢. incanus (Wilson et al. 2017; Wei et al. 2021). In this study, we evaluate specimens from 7scherskia and Cansumys collected from Gansu, Henan, Shandong, Shanxi, Shaanxi, Heilongjiang, Beijing, and Inner Mon- golia Provinces. We compared the external and skull mor- phologies of these specimens and conducted a combined analysis of the DNA sequences of one mitochondrial and three nuclear genes. Our aims were to infer: 1) the phy- logenetic relationship and status of the genera 7scherskia and Cansumys;, 2) the molecular phylogeny among the subspecies of 7scherskia;, and 3) the taxonomic status of T. ningshaanensis. Materials and methods Specimen collection and DNA sequencing We collected tissues from 27 specimens (7scherskia and Cansumys) from Gansu, Henan, Shandong, Shanx1, Shaanxi, Heilongjiang Provinces, as well as Inner Mon- golia Autonomous Region and Beijing, China (Fig. 1 and Suppl. material 1: table S1). Voucher specimens and liver or muscle tissue are deposited at the Institute of Biodiver- sity and Ecology (IBE), Zhengzhou University, Sichuan Academy of Forestry Sciences (SAF), Marine College, Shandong University (SDU), and College of Life Sci- ences, Sichuan Normal University (SNU), respectively. The genomic DNA of specimens was extracted from the liver and muscle tissues using a DNA extraction kit (Tian- gen DNA Easy Blood and Tissue Kit, Beijing, China). One complete mitochondrial locus [Cytochrome b (CYT B)| and partial sequences of three nuclear genes [inter- photoreceptor retinoid-binding protein (IRBP), growth hormone receptor (GHR), and recombination activating protein 1(RAG1)] were amplified. Primer pairs were ob- tained from the literature (Teeling et al. 2000; Galewski et al. 2006; He et al. 2010; Cheng et al. 2017) and are shown in Suppl. material 1: table S2. PCR products were sent to Sangon Biotech Co., Ltd. for sequencing. Phylogenetic analyses and molecular dating Phylogenetic analyses Recovered DNA sequences were assembled and aligned in- dividually using MEGA X (Kumar et al. 2018). Additionally, 49 sequences from 10 species were downloaded from Gen- Bank (Suppl. material 1: table S3). Lagurus lagurus was used as the outgroup for all subsequent phylogenetic analyses. We calculated Bayesian Inference (BI) and maxi- mum likelihood (ML) using BEAST v1.7.4 (Drummond et al. 2012) and W-IQ-TREE (Trifinopoulos et al. 2016), Zoosyst. Evol. 100 (4) 2024, 1231-1241 100°0'0"E 110°0’0"E 120°0'0"E 1283 130°0'0"E * 50°0'0"N Figure 1. Collection sites of specimens of 7scherskia and Cansumys in China. respectively, based on CYT B and nuDNA (GHR + IRBP + RAG1). For the BI analyses, the best-fit evolutionary models for CYT B and the three nuclear loci were deter- mined using the Akaike Information Criterion (AIC) 1m- plemented in JMODELTEST v2.1.10 (Suppl. material 1: table S4) (Darriba et al. 2012). We employed a relaxed, uncorrelated lognormal clock model, Yule process tree priors, and the default prior distribution of the program for the model parameters. Each analysis was run for 100 million generations, with samples taken every 5,000 gen- erations (Drummond et al. 2012). TRACER v1.7.0 was used to assess the effective sample size (ESS) values (..e., ESS > 200) (Rambaut et al. 2018). And the first 10% of the trees were treated as burn-in. For the ML analyses, the prior value of the parameter used was the default value for W-IQ-TREE (Trifinopou- los et al. 2016). We employed ultrafast bootstrap analysis with 1,000 bootstrap replicates, the SH-aLRT branch test with 1,000 iterations, the maximum iterations set to 1,000 iterations, and a minimum correlation coefficient of 0.99. Species delimitation Firstly, we calculated the Kimura-2-parameter (K2P) dis- tance between specimens of Cricetinae in this study based on the CYT B gene using MEGA X (Kumar et al. 2018). Species trees (CYT B + nuDNA combined) were calcu- lated using the *BEAST model in BEAST v1.7.4 (Heled and Drummond 2010; Drummond et al. 2012). Eight lin- eages were treated as species in the *BEAST based on the results of the K2P distance and phylogenetic analyses. We used the Yule speciation model and the strict clock model for tree construction. Other parameters followed BI settings. Each analysis was run for 100 million gen- erations, with samples taken every 5,000 generations (Drummond et al. 2012). Secondly, another species delimitation analysis was conducted using the program BPP v3.1 (Camargo et al. 2012; Yang and Rannala 2014). The BPP analyses were performed using dataset] (CYT B + nuDNA combined) and dataset2 (nuDNA combined), respectively. And the best tree topology recovered by BEAST v1.7.4 was used as the guide tree. The validity of our assignment of Cricet- inae species was tested in BPP v3.1. The species delimi- tation analysis only included individuals who possessed both mtDNA and nuDNA data. Two reversible jump Mar- kov chain Monte Carlo (r,; MCMC) algorithms for species delimitation (algorithms 0 and 1) were used, respective- ly. Each rfMCMC was run for 100,000 generations, with sampling every 100 generations following a pre-burn-in of 10,000 generations as determined by TRACER v1.7 (Rambaut et al. 2018). Divergence-time analyses Divergence times were estimated based on the three nu- clear loci combined (IRBP + GHR + RAG!). The diver- gence time was estimated using BEAST v1.7.4. The prior for the age of the tree root was based on the results by Steppan et al. (2004) (mean = 19 ma, standard deviation = 1.5), as referenced in Lebedev et al. (2018). We used the Yule Process speciation model and the uncorrelated re- laxed clock model for tree construction. The substitution rate model is set according to Bayesian trees. Each anal- ysis was run for 100 million generations, with samples taken every 5,000 generations (Drummond et al. 2012). zse.pensoft.net 1234 Analyses of external morphological and skull features The external morphological characteristics of specimens (TZ. ningshaanensis and T: triton) were compared based on specimens and data from a previous study. Follow- ing the original description by Song (1985), we examined T. ningshaanensis and compared it with all subspecies of T. triton. A total of 15 specimens (T. triton: 7 and T: ningshaan- ensis: 8) were collected. For these specimens, we exam- ined and measured several parameters, including external morphology and 11 craniodental measurements (Yang et al. 2005). External morphological data (including W: weight; HBL: head and body length; TL: tail length; HL: hindfoot length; EL: ear length) were measured by a digital scale (0.1 g) and measured (1 mm) from the original specimens; craniodental measurements (includ- ing PL: Profile length; BL: Basal length; SUCL: Short upper cranium length; ZB: Zygomatic breadth; IOB: In- terorbital breadth; CH: Cranial height; TBL: tympanic bulla length; UMRL: Upper molar row length; LMRL: Lower molar row length; ML: Mandibular length; CL: Condyle length) were taken with digital calipers (0.01 mm). We compared specimens of 7scherskia based on measurements of external and skull morphology. Over- all similarities between external morphology and skulls were assessed first through principal component analyses (PCA). The PCA was conducted at OriginLab (OriginLab Corporation, version 2024, USA). Results Sequence characteristics We obtained ~3573 bp of sequence for most specimens, partitioned into 1140 bp of mitochondrial sequence (CYT B [1140 bp]) and 2433 bp of nuclear sequence (IRBP [895 bp], GHR [810 bp], and RAG/ [728 bp]). All new sequences have been deposited in GenBank (accession numbers: CYT B PP975895—PP975921, GHR PP975932—PP975950, RAGI PP975951—PP975969, IRBP PP975970—PP975985). Phylogenetic analyses The concatenated BI and ML recovered the identical to- pology; therefore, only the BI tree is presented (Fig. 2). Most of the nodes were strongly supported [1.e., BEAST posterior probabilities (PP) > 0.95, SH-aLRT values (SH) > 80, ultrafast bootstrap values (UBS) (Huelsenbeck and Rannala 2004; Guindon et al. 2010; Minh et al. 2013)], with few exceptions based on combined CY7 B and nuclear loci (Fig. 2a, b). The BI and ML results strongly supported zse.pensoft.net Jiang, H. et al.: Phylogeny and taxonomy of Tscherskia and Cansumys sister relationships between 7! ningshaanensis and T. tri- ton, and both should be treated as single species, respec- tively [T ningshaanensis (CYT B: PP = 1, SH=99.9, UBS = 100; nuDNA: PP = 1, SH = 96.6, UBS = 97); T triton (CYT B: PP = 1, SH = 100, UBS = 100; nuDNA: PP = 1, SH = 99.9, UBS = 100)] (Fig. 2a, b). The genus Cansumys was strongly supported as monophyletic based on CYT B (PP = 1, SH = 100, UBS = 100) (Fig. 2a). The BI and ML analyses based on CYT B indicate that 7’ ¢. triton is differ- entiated from the other subspecies, whereas 7° t. incanus and 7? t. fuscipes do not show distinct separation (Fig. 2a). The BI and ML analyses based on nuDNA results do not support the classification as a subspecies of 7: triton (Fig. 2b). In addition, the species Phodopus roborovskii and Urocricetus kamensis were placed at the base of Cricetinae in both the analysis of CYT B and nuDNA results (CYT B: PP = 1, SH = 83, UBS = 85; nuDNA: PP = 0.99, SH = 97.3, UBS = 99) (Fig. 2a, b). Species delimitation Calculated K2P distances based on CYT B were as fol- lows: between 7! ningshaanensis and Can. canus (25.5%), between 7. ningshaanensis and T: triton (15.1%), and be- tween Can. canus and T: triton (23.8%) (Table 1). Addi- tionally, BPP analysis results based on dataset] and data- set2 supported 7? ningshaanensis and T. triton as separate species (PP = 1.00), respectively. The BEAST tree analy- ses recovered the same topology as the BI and ML trees, with sister relationships between 7: ningshaanensis and T. triton also strongly supported (PP = 0.99) (Fig. 3a). Molecular divergence time Our phylogenetic analyses based on nuDNA revealed highly concordant divergence time estimates (Fig. 3b). The species Phodopus roborovskii and Urocricetus kamensis were placed at the base of Cricetinae, with the divergence time result estimated to be in the latest Mid- dle Miocene (12.73 Ma). Apart from the split between T: triton and T. ningshaanensis (3.88 Ma), intra-generic di- vergence events primarily occurred in the latest Pliocene. Morphological and skull comparison All external and skull measurements are provided in Table 2. The mean values of most measurements for T. ningshaanensis are smaller than those for T. triton, with significant differences in W, HBL, PL, BL, SUCL, ZB, CH, TBL, ML, and CL. However, the sizes of HFL (23.50- 27.00, 24.79+1.15 vs 20.00—26.00, 23.24+2.17) and LMRL (5.28—-5.41, 5.34+0.05 vs 5.13—-5.77, 5.45+0.20) of T ning- shaanensis are bigger than those for T triton (Table 2). Zoosyst. Evol. 100 (4) 2024, 1231-1241 Lagurus lagurus — Phodopus roborovskii Urocricetus kamensis CSD3982 il Cricetulus griseus Cricetus cricetus | Cricetus cricetus 2 Cricetus cricetus 3 170 M11356 M11372 S0843@ SI24@ CSD3942 ® IBE00052@ BJ20001 9% BI20002% IBE00513@ @ F ningshanensis Wi Can. canus i Wi 7. t triton IBE00533@ W Tt. fuscipes IBE00644@ @ Ft incaus HEE @ uncertain IBE00262 IBE00271 TBE00236 E00643 TBE00249' 03 IBE00690@ 5 00244 A:CYTB IBEDIGC® i IBE00664@ Cricetulus Tongicaudatus 70 Cricetulus longicaudatus 449106 Tscherskia triton | @ Tscherskia triton 2 @ IBE00248@ 1235 Lagurus lagurus V2 0.99/97.3/99 Urocricetus kamensis 449099 Phodopus roborovskii 138 Allocricetulus curtatus 16 Cricetus cricetus Cc1 Cricetulus longicaudatus 70 Cricetulus longicaudatus 449106 Cricetulus barabensis 131 Cricetlus sokolovi Cs M11356 CSD3942@ M11372 BI20001& B20002% Tscherskia triton Tt} S1524@ S083 @ $4170 IBE00249@ IBE00248@ QL19006@ QL19005@ IBE00245@ IBE00644@ IBE00690@ IBE00236@ TBE00643@ IBE00237@ IBE00271@ U. kamensis P. roborovskii P. roborovskii U. kamensis 1/87.1/93 Can. canus A, curtatus Cri. cricetus C. griseus 1/93.4/95 C. longicaudatus C. longicaudatus C. barabensis _C. sokolovi Cri. cricetus 0.94/99.9/100 1/99.9/100 T. triton T. triton 1/100/100 1/96.6/97 T. ningshanensis T. ningshanensis 0.007 B: nuDNA Figure 2. Maximum likelihood and Bayesian inference analysis results based on CYT B (A) and nuDNA (B). Left: BI posterior probabilities; middle: SH-aLRT values; right: ultrafast bootstrap values. Table 1. K2P distances between species of Cricetinae based on the CYT B gene. Tsc nin Cri gri Tsc nin Cri gri 03231 Lag lag 0.269 0.224 Tsc tri O15] 0.229 Cri cri 0.204 0.188 Pho rob 0.257 0.225 Cri long 0.236 0.143 Uro kam 0.225 0.237 Can can 0.255 0.222 Lag lag 0.258 0.213 0.222 0.236 0.220 0.221 Tse tri Cri cri Pho rob Cri long Uro kam 0.209 0.263 0.231 0.223 0.208 0.240 0.238 0.201 0.198 0.244 0.238 0.208 0.212 O22 0.217 Table 2. Some external and skull measurements (mm) used in PCA analyses of 7. ningshaanensis and T: triton. Measurement (min, max, mean + SD) W HBL Tk HFL EL PL BL SUCL ZB lOB CH TBL UMRL LMRL ML CL: 66.00 — 92.00, 80.67 + 10.16 146.00 - 175.00, 128.00 + 16.38 65.00 — 90.00, 77.43 + 8.28 20.00 - 26.00, 23.24 + 2.17 18.00 - 22.00, 20.26 + 1.49 33.43 - 37.68, 35.79 + 1.56 30.42 - 35.30, 32.13 + 1.79 33.43 - 38.94, 35.97 + 1.86 16.87 — 20.22, 17.99 + 1.06 4.88 - 5.54, 5.18 + 0.22 12.83 - 13.38, 13.11 + 0.20 8.62 - 10.94, 10.06 + 0.79 9:03: = 5:97, 527 +O: l fF 5.13 - 5.77, 5.45 + 0.20 22.46 - 24.70, 23.69 + 0.74 19.08 - 20.70, 19.71 + 0.64 T. triton T. ningshaanensis 37.64 - 93.10, 52.44 + 19.81 112.00 - 155.00, 129.29 + 16.15 76.00 - 114.00, 89.71 + 12.62 23.50 - 27.00, 24.79 + 1.15 20.00 — 23.00, 21.71 + 1.04 29.33 - 37.07, 31.49 + 2.65 26.99 — 34.47, 29.03 + 2.68 30.24 - 37.93, 32.43 + 2.74 15.54 - 19.02, 16.30 + 1.34 4.66 —- 5.47, 5.03 + 0.26 11.90 - 12.98, 12.31 + 0.40 7.98 - 10.29, 8.63 + 0.89 4.94 —- 5.34, 5.13 + 0.16 5.28 — 5.41, 5.34 + 0.05 19.23 - 25.06, 20.84 + 2.02 16.19 - 19.83, 17.18 + 1.28 Note: W: weight; HBL: head and body length; TL: tail length; HL: hindfoot length; EL: ear length; PL: profile length; BL: basal length; SUCL: short upper cranium length; ZB: zygomatic breadth; IOB: interorbital breadth; CH: cranial height; TBL: tympanic bulla length; UMRL: upper molar row length; LMRL: lower molar row length; ML: mandibular length; CL: condyle length. zse.pensoft.net 1236 18.64 8.32 T. ningshaanensis i C. barabensis A. curtatus Jiang, H. et al.: Phylogeny and taxonomy of Tscherskia and Cansumys Lagurus lagurus V2 P roborovskii U. kamensis Cri. cricetus A, curtatus Allocricetulus curtatus 16 Cricetulus longicaudatus 449106 Cricetulus longicaudatus 70 C. longicaudatus C. barabensis C. sokolovi Cricetulus sokolovi Cs1 T. triton T. ningshanensis A 20 15 10 5 0 Ma Figure 3. Divergence times estimated (A) and species delimitation (B) results in this study. a: black stars represents BPP species definition results; numbers of each node represent posterior probabilities (under); b: numbers of each node represent posterior prob- abilities (upper) and divergence times (under). The PCA, based on 16 measurements [including ex- ternal morphology (5) and skull (11) measurements], produced two axes (PC1: 10.20 and PC2: 2.69) with ei- genvalues > 2.0, explaining 60.00% and 15.85% of the variance (75.85% total) (Table 3). PC1 was positively correlated with all variables (Table 3). PC2 was strongly correlated with TL and EL, loading > 0.5. The PCA re- sults showed that most specimens of 7) ningshaanensis and 7! triton could be distinguished from each other based on 16 log, transformed variables (Fig. 4). Discussion The classification status of Can. canus and T. ningshaan- ensis had been extensively discussed in previous studies (Yang et al. 2003; Gu et al. 2005; Liao et al. 2007). The results from this study provide molecular evidence into the classification status of Can. canus and T. ningshaan- ensis, encompassing almost all subspecies of 7° triton found in China (except for 7 t collinus). Our phyloge- netic and morphological results indicated that the ge- nus Cansumys should be treated as a distinct genus, and T. ningshaanensis 1s a distinct species. The genetic dis- tance values among three species based on CYT B indicat- ed that 7. ningshaanensis, T: triton, and Can. canus are all distinct species (> 11%) (Bradley and Baker 2001). The zse.pensoft.net Table 3. Character loadings, eigenvalues, and percent variance explained on the first two components of a PCA of 7° triton and T. ningshaanensis. Variables PCl PC2 W 0.31 0.02 HBL 0.26 0.00 TL 0.06 0.55 HFL 0.01 0.43 EL 0.02 0.52 PL 0.81 0.03 BL 0.30 0.11 SUCL 0.31 0.10 ZB 0.30 0.05 |OB 0.18 0.21 CH 0.29 -0.14 TBL 0.22 -0.11 UMRL 0.19 0.02 LMRL 0.14 -0.16 ML 0.31 0.03 Gl. 0.31 -0.05 Eigenvalues 10.20 2.69 Variance explained (%) 60.00 15.85 large genetic distance (20.8% — 25.5%) and phylogenetic analyses based on CYT B strongly supported the classi- fication status of Cansumys as a distinct genus (PP = 1, SH = 100, UBS = 100, Table 1), which was consistent Zoosyst. Evol. 100 (4) 2024, 1231-1241 PC2 (15.8%) 5 PC1 (60.0%) Figure 4. Results of principal component analysis (PCA) of T. ningshaanensis and T. triton. with previous research (Smith and Xie 2008; Wilson et al. 2017; Wei et al. 2021). The calculated K2P distances based on CYT B of T. triton compared with other species of Cricetinae in this study ranged from 15.1% (T° triton) to 25.7% (P. roborovskii). The results from phylogenetic analyses based on CYT B and nuDNaA loci strongly sup- ported 7? ningshaanensis as a separate species (CYT B: PP = 1, SH = 99.9, UBS = 100; nuDNA: PP = 1, SH = 96.6, UBS = 97); T. triton (CYT B: PP = 1, SH = 100, UBS = 100; nuDNA: PP =1, SH = 99.9, UBS = 100)] (Fig. 2a, b). The genus Cansumys was strongly supported as monophyletic based on CYT B (PP = 1, SH = 100, UBS = 100. Fig. 2a, b). However, the fine-scale subdi- vision of subspecies of 7: triton indicates that additional studies are warranted to clarify the status of the described subspecies (Fig. 2a, b). The growth of the Tibetan Plateau led to the uplift of the Qinling Mountains during the late Miocene to Plio- cene (8—4 Ma) (Wang et al. 2011). This rapid uplift con- tributed to the biodiversity within the Qinling Moun- tains (Dong et al. 2022). The divergence time between T. ningshaanensis and T: triton was estimated to be approximately 3.88 million years ago. This divergence time suggested that the formation of 7! ningshaanensis and T. triton was influenced by the uplift of the Qinling Mountains. In Europe, the earliest species of 7scherskia (T. europaeus and T: janossyi) was found at Csarnota 2 (MN 15, ca. 5—3.5 Ma) in Hungary (Hir 1994; Venc- zel and Gardner 2005). The earliest known species of 7scherskia (T: sp.) was found in the Late Pliocene (2.58-3.60 Ma) from the Youhe Formation (ca. 3.40— 2.59 Ma) (Yue and Xue 1996; Xie et al. 2021) in Linwei District, Weinan, Shaanxi Province, China. These find- ings suggested that the species of 7scherskia underwent rapid diversification during the late Pliocene (2.58—3.60 Ma). Fossils of 7. ¢t. varians were found from the late Middle Pleistocene to the Early Pleistocene in China (0.129 Ma—2.58 Ma) (Zheng 1984a, 1984b, 1993; Jin et al. 2009; Xie et al. 2023). This suggested that 7° triton L237 underwent diversification during the early Pleistocene, which is consistent with the divergence time of 7° triton estimated in this study. In addition, we compared the distribution range, ex- ternal morphology, and skull morphology of 7! ning- shaanensis, T. triton, and Can. canus. The fact that both T. triton and Can. canus were found at Muer of Zhuoni, Gansu Province, assisted the conclusion that 7. triton and Can. canus are distinct species. The results of ex- ternal morphology analyses showed 7. ningshaanensis and 7: triton could be distinguished from each other with many distinguishable features. 7. t. triton (de Win- ton 1899) has a dorsal coloration that is uniformly drab, with whitish underparts (Suppl. material 2). 7? t. colli- nus (Allen 1925) is similar to 7! ¢. triton but is much darker with a slightly longer tail. Its dorsal coloration is between drab and mouse gray, with warm buff sides to the head and body. The chin, feet, wrists, and a small median spot on the throat have clear white hairs. The tail is blackish-brown and thinly covered with short, appressed hairs, with many whitish hairs on the low- er side. T. ¢. fuscipes (Allen 1925) is similar in general appearance to 7! ¢. triton but has ankles and a basal part of the metatarsals that are dusky. Its entire dorsal area is nearly uniformly buffy, with hairs that are entirely black or have a fine black tip. The tail is thinly covered with hairs, dusky above and whitish below. 7! ¢. incanus (Thomas 1908) is similar in general appearance to T. t. triton, with a white dorsum pedis and dorsal and belly hairs that are pale. Can. canus (Allen 1928) has a dorsal surface of the body and tail that is generally gray. The middle of its back has slaty-gray hairs with short whitish tips, interspersed with numerous all-black hairs, and there is a faint wash of buffy color on the sides of the body, while its tail is thickly covered with fine hairs. In the original description, Song (1985) presented several morphological characters to distinguish it from other subspecies of 7 triton: 1) smaller body size com- pared to other subspecies of 7 triton; 2) entire dorsal surface dark grayish—brown, covered with long black hair, while the ventral grayish—white with the medial part of the body hair being gray and the distal part being whit; 3) center of the chest and feet white, ankles taupe covered with thick haired; 4) long tail, with the tail length nearly 66% of the body length; 5) the tail appears bicolored, with the basal part being grayish-brown and the distal part white, covered with dense hair; 6) white tail percentage, with the white length nearly 40%—60% of the tail; 7) intumescent tail base; and 8) less devel- oped supraorbital ridge (Figs 5, 6). 7? ningshaanensis and 7: triton (including four subspecies) can be clearly distinguished from each other based on morphological characteristics (body grayish-brown, long tail with two color rears, 40-60% being white). We believed that the aforementioned evidence supports the conclusion that T. ningshaanensis should not be treated as a synonym of 7. t. incanus. zse.pensoft.net 1238 Jiang, H. et al.: Phylogeny and taxonomy of Tscherskia and Cansumys , Sis of > a Tail Besa g stot TS TEENS OS eee ee * O | ll el ee = 9 i e s : ete See me me ee ere — Tail BE Ps ood eee ee ee ee ee *e — 4 T. triton T. ningshaanensis Figure 6. Dorsal, ventral, and lateral views of the skull and lateral views of the mandible, 7’ ningshaanensis and T. t. triton. zse.pensoft.net Zoosyst. Evol. 100 (4) 2024, 1231-1241 Conclusions In this study, we used morphology and molecular phy- logeny to investigate the taxonomy, phylogenetic rela- tionships, and evolutionary history of the genera 7scher- Skia and Cansumys. The results supported the following conclusions: 1) the genus Cansumys 1s valid and distinct, possibly monotypic as only Can. canus is currently de- scribed; 2) the genus 7scherskia consists of T. ningshaan- ensis and T: triton, and T. ningshaanensis is not a syn- onym of 7! ¢ incanus, and 3) the uplift of the Qinling Mountains likely facilitated the geographical isolation of ancestral species, further promoting the speciation of T. ningshaanensis. Acknowledgments Field surveys and collections of specimens follow rel- evant regulations in China. We thank Prof. Yuchun L1, Marine College, Shandong University, for providing ma- terials for T. triton analysis. This research was funded by the Zhengzhou Science and Technology Talent Team Construction Plan (No. 131PLJRC654). References Allen GM (1925) Hamsters collected by the American Museum Asiatic Expeditions. American Museum novitates; no. 179. American Mu- seum of Natural History, New York City. Allen GM (1928) A New Cricetine Genus from China. Journal of Mam- malogy 9: 244-246. https://doi.org/10.2307/1373275 Argyropulo A (1933) Die Gattungen und Arten der Hamster (Cricet- inae Murray, 1866) der Palaarktik. Zeitschrift fir Saugetierkunde 8: 129-149, Bradley RD, Baker RJ (2001) A Test of the Genetic Species Concept: Cytochrome-b Sequences and Mammals. Journal of Mammalogy 82: 960-973. https://doi.org/10.1644/1545-1542(2001)082<0960:A- TOTGS>2.0.CO;2 Camargo A, Morando M, Avila LJ, Sites JW (2012) Species delimita- tion with abc and other coalescent-based methods: a test of accura- cy with simulations and an empirical example with lizards of the Liolaemus Darwinii complex (Squamata: Liolaemidae). Evolution- ary Anthropology 66: 2834-2849. https://doi.org/10.1111/j.1558- 5646.2012.01640.x Chen FG, Min ZL (1982) Taxonomic review of sevenal speceis in Ron- dents. Zoological research 3: 369-374. Cheng F, He K, Chen ZZ, Zhang B, Wan T, Li JT, Zhang BW, Jiang XL (2017) Phylogeny and systematic revision of the genus 7iphlomys (Rodentia, Platacanthomyidae), with description of a new species. Journal of Mammalogy 98: 731-743. https://doi.org/10.1093/jmam- mal/gyx016 Corbet GB, Hill. JE (1992) Mammals of the Indomalayan region. A sys- tematic review. Oxford University Press, Oxford. Darriba D, Taboada GL, Doallo R, Posada D (2012) jModelTest 2: more models, new heuristics and parallel computing. Nature Methods 9: 772-772. https://doi.org/10.1038/nmeth.2109 1239 de Winton W (1899) On Chinese mammals, principally from West- ern Sechuen, with notes on Chinese squirrels. Proceedings of the Zoological Society of London 67: 572-578. https://doi. org/10.1111/j.1469-7998.1899.tb06875.x Dong Y, Shi X, Sun S, Sun J, Hui B, He D, Chong F, Yang Z (2022) Co-evolution of the Cenozoic tectonics, geomorphology, environ- ment and ecosystem in the Qinling Mountains and adjacent ar- eas, Central China. Geosystems and Geoenvironment 1: 100032. https://doi.org/10.1016/j.geogeo.2022. 100032 Drummond AJ, Suchard MA, Xie D, Rambaut A (2012) Bayesian Phy- logenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution 29: 1969-1973. https://doi.org/10.1093/molbev/mss075 Ellerman JR (1941) The families and genera of living rodents. Volume II. Muridae. Vol. 2,British Museum, London. Ellerman JR, Morrison-Scott TCS (1951) Checklist of Palaearctic and Indian mammals, 1758-1946. https://www.biodiversitylibrary.org/ bibliography/8602 Galewski T, Tilak M, Sanchez S, Chevret P, Paradis E, Douzery EJP (2006) The evolutionary radiation of Arvicolinae rodents (voles and lemmings): relative contribution of nuclear and mitochondrial DNA phylogenies. BMC Evolutionary Biology 6: 80. https://doi. org/10.1186/1471-2148-6-80 Gu Y, Ma Y, Sun YH (2005) A Rediscussion on the Taxonomic Status of Cansumys canus. Chinese Jounal ef Zoology 40: 116-120. Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W, Gascuel O (2010) New algorithms and methods to estimate maximum-like- lihood phylogenies: assessing the performance of PhyML 3.0. Sys- tematic Biology 59: 307-321. https://doi.org/10.1093/sysbio/syq010 He K, Li YJ, Brandley MC, Lin LK, Wang YX, Zhang YP, Jiang XL (2010) A multi-locus phylogeny of Nectogalini shrews and influ- ences of the paleoclimate on speciation and evolution. Molecular Phylogenetics and Evolution 56: 734-746. https://doi.org/10.1016/j. ympev.2010.03.039 Heled J, Drummond AJ (2010) Bayesian inference of species trees from multilocus data. Molecular Biology and Evolution 27: 570-580. https://doi.org/10.1093/molbev/msp274 Hir J (1994) Cricetinus beremendensis sp. n. (Rodentia, Mammalia) from the Pliocene fauna of Beremend 15. (S Hungary). Fragmenta Mineralogica et Paleontologica 17: 71-89. Ho H (1934) A new subspecies of Cricetidae from Shansi. Contributions from the Biological Laboratory of Science Society of China, Naking Zoological Series 10: 288-291. Huelsenbeck JP, Rannala B (2004) Frequentist Properties of Bayesian Posterior Probabilities of Phylogenetic Trees Under Simple and Complex Substitution Models. Systematic Biology 53: 904-913. https://doi.org/10.1080/10635 150490522629 Jin CZ, Zhang YQ, Wei GB, Cui N, Wang Y (2009) Rodentia. In: Jin CZ, Liu JY, eds. Paleolithic Site—The Renzidong Cave, Fanchang, Anhui Province. Science Press, Beijing. Kishida K (1929) A synopsis of Corean hamsters. In., Lansania, Tokyo 1-160. Kumar S, Stecher G, Li M, Knyaz C, Tamura K (2018) MEGA X: Molecular Evolutionary Genetics Analysis across Computing Plat- forms. Molecular Biology and Evolution 35: 1547-1549. https://doi. org/10.1093/molbev/msy096 Lebedev VS, Bannikova AA, Neumann K, Ushakova MV, Ivanova NV, Surov AV (2018) Molecular phylogenetics and taxonomy of dwarf hamsters Milne-Edwards, 1867 (Cricetidae, Rodentia): description zse.pensoft.net 1240 of a new genus and reinstatement of another. Zootaxa 4387: 331-— 349. https://https://doi.org/10.11646/zootaxa.4387.2.5 Liao J, Xiao Z, Dong Y, Zhang Z, Liu N, Li J (2007) Taxonomic status of Cansumys canus (Allen,1928). Acta Zoologica Sinica 53: 44-53. Lu JQ (1997) The Study on the Geocraphical Division ofGlires in Henan Province. Journal of Henan Normal University (Natural Science) 25: 68-71. https://doi.org/10.16366/j.cnki. 1000-2367.1997.02.017 Minh BQ, Nguyen MA, von Haeseler A (2013) Ultrafast approximation for phylogenetic bootstrap. Molecular Biology and Evolution 30: 1188-1195. https://doi.org/10.1093/molbev/mst024 Musser GG, Carleton. MD (1993) Family Muridae. Pp. 501-755, in: Mammal species of the world, a taxonomic and geographic refer- ence, Second ed. (D. E. Wilson and D. M. Reeder, eds.). Smithso- nian Institution Press, Washington D. C xviii. Neumann K, Michaux J, Lebedev V, Yigit N, Colak E, Ivanova N, Pol- toraus A, Surov A, Markov G, Maak S, Neumann S, Gattermann R (2006) Molecular phylogeny of the Cricetinae subfamily based on the mitochondrial cytochrome and 12S rRNA genes and the nuclear vWF gene. Molecular Phylogenetics and Evolution 39: 135-148. https://doi.org/10.1016/j.ympev.2006.01.010 Ognev SI (1914) Die saugetiere aus dem sudlichen Ussurigebiet (the mammals of the southern Ussuri). Journal de la Section Zoologique de la Societe des Amis des Sciences Naturelles, d’Anthropologie et d’ethnographie 2: 101-128. Pavlinov IY, Yakhontov. EL, Agadzhanyan AK (1995a) Mammals of Eurasia. I. Rodentia. Taxonomic and geographic guide. Vol. 32, Ar- chives of the Zoological Museum, Moscow State University. Rambaut A, Drummond AJ, Xie D, Baele G, Suchard MA (2018) Pos- terior Summarization in Bayesian Phylogenetics Using Tracer 1.7. Systematic Biology 67: 901-904. https://doi.org/10.1093/sysbio/ syy032 Ross P (1988) The taxonomic status of Cansumys canus. Abstracts, Symposium of Asian Pacific Mammalogy. Huirou, Beijing,Peoples Republic of China, 245 pp. Smith AT, Xie Y (2008) A Guide to the Mammals of China. Princeton University Press, Princeton, New Jersey, 135-144. Song S (1985) A new subspecies of Cricetulus triton from Shananxi, China. Acta Theriological Sinica 5: 137-139. Steppan SJ, Storz BL, Hoffmann RS (2004) Nuclear DNA phylogeny of the squirrels (Mammalia: Rodentia) and the evolution of arboreality from c-myc and RAG. Molecular Phylogenetics and Evolution 30: 703-719. https://doi.org/10.1016/S1055-7903(03)00204-5 Teeling EC, Scally M, Kao DJ, Romagnoli ML, Springer MS, Stan- hope MJ (2000) Molecular evidence regarding the origin of echo- location and flight in bats. Nature 403: 188-192. https://doi. org/10.1038/35003188 Thomas O (1907) The Duke of Bedford’s zoological expedition in eastern Asia. V. Second list of mammals from Korea. Proceed- ings of Zoological Society of London 32: 462-466. https://doi. org/10.1111/j.1469-7998. 1907 .tb06941.x Thomas O (1908) Cricetulus triton incanus subsp. n. In: Woodward H. Abstract of the proceedings of the zoological society of london 63: 43-46. zse.pensoft.net Jiang, H. et al.: Phylogeny and taxonomy of Tscherskia and Cansumys Trifinopoulos J, Nguyen L-T, von Haeseler A, Minh BQ (2016) W-IQ- TREE: a fast online phylogenetic tool for maximum likelihood anal- ysis. Nucleic Acids Research 44: 232-235. https://doi.org/10.1093/ nar/gkw256 Venczel M, Gardner JD (2005) The geologically youngest albanerpeton- tid amphibian, from the lower Pliocene of Hungary. Palaeontology 48: 1273-1300. https://doi.org/10.1111/j.1475-4983.2005.00512.x Wang TZ, Xu WX (1992) Glires (Rondentia and Lagomorpha) fauna of Shaanxi Province. Shaanxi Normal University Press, Xian. Wang S, Zheng CL (1973) Notes on Chinese Hamsters(Cricetinae). Acta Zoologica Sinica 19: 61-68. Wang X, Zattin M, Li J, Song C, Peng T, Liu S, Liu B (2011) Eocene to Pliocene exhumation history of the Tianshui-Huicheng region determined by Apatite fission track thermochronology: Implications for evolution of the northeastern Tibetan Plateau margin. Journal of Asian Earth Sciences 42: 97-110. https://doi.org/10.1016/).jse- aes.2011.04.012 Wei F, Yang Q, Wu Y, Jiang X, Liu S, LiB, YANG G, Li M, Zhou J, Li S, Hu Y, Ge D, Li S, YU W, Chen B, Zhang Z, Zhou C, Wu S, Zhang L, Chen Z, Chen S, Deng H, Jiang T, Zhang L, Shi H, Lu X, Li Q, Liu Z, Cui Y, Li Y (2021) Catalogue of mammals in China(2021). Acta Theriological Sinica 41: 487-501. https://doi.org/10.16829/). slxb.150595 Wilson DE, Mittermeier, RA, Lacher TE (2017) Handbook of the Mammals of the World, Vol. 7: Rodents I. Vol. 7, Lynx Edicions, Barcelona, Spain. Xie K, Zhang YX, Li YX (2021) Revision to from the Houhecun Fauna and a New Discovery of (Cricetidae, Rodentia) from the Youhe Fauna of Weinan, Shaanxi Province, China. Acta Geologica Sinica-English Edition 95: 1073-1079. https://doi.org/10.1111/1755-6724. 14776 Xie K, Zhang YX, Li YX (2023) Large-sized fossil hamsters from the late Middle Pleistocene Locality 2 of Shanyangzhai, China, and discussion on the validity of Cricetinus and C. varians (Rodentia: Cricetidae). Peerj 11. https://doi.org/10.7717/peerj.15604 Yang ZH, Rannala B (2014) Unguided species delimitation using DNA sequence data from multiple Loci. Molecular Biology and Evolution 31: 3125-3135. https://doi.org/doi:10.1093/molbev/msu279 Yang L, Chen X, Zhao X, Wang J (2003) Karyotype and Classif ication Status of Cansumys canus (Cricetidea, Rodentia). Acta Theriologi- cal Sinica 23: 235-238. https://doi.org/10.16829/j.sIxb.2003.03.010 Yang QS, Xia L, Ma Y, Feng ZJ, Quan GQ (2005) A guide to the mea- surement of mammal skull I: Basic measurement. Chinese Journal of Zoology 40: 50-56. https://do1.org/10.13859/).cjz.2005.03.011 Yue LP, Xue XX (1996) Palaeomagnetism of Chinese Loess. Geology Press, Beijing. Zheng SH (1984a) Revised determination of the fossil Cricetine (Roden- tia, Mammalia) of Choukoutien district. Vertebrata PalAsiatica 22: 179-197. https://doi.org/10.19615/j.cnki. 1000-3118.1984.03.002 Zheng SH (1984b) A new species of Kowalskia (Rodentia, Mammalia) of Yinan, Shandong. Vertebrata PalAsiatica 22: 251—260. https://doi. org/10.19615/j.cnki.1000-3118.1984.04.001 Zheng SH (1993) Quaternary rodents of Sichuan-Guizhou area, China. Science Press., Beijing. Zoosyst. Evol. 100 (4) 2024, 1231-1241 Supplementary material | Supplamentary information Authors: Haijun Jiang, Xuming Wang, Yaohua Yang, Xuan Pan, Shaoying Liu, Jiqi Lu Data type: docx Explanation note: table S1. Sampling localities of gen- era Tscherskia and Cansumys in this study. table S2. Primers used for PCR and sequencing in this study.ta- ble S3. Sampling information including localities and GenBank accession numbers for species used in this study. table S4. The best molecular evolution model according to the Akaike Information Criterion (AIC) used in phylogenetic reconstructions based on jMod- eltest v2.10 Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons. org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow us- ers to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://do1.org/10.3897/zse.100.128567.suppl1 1241 Supplementary material 2 Dorsal and ventral views of 7: triton Authors: Haijun Jiang, Xuming Wang, Yaohua Yang, Xuan Pan, Shaoying Liu, Jiqi Lu Data type: psd Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons. org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow us- ers to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://doi.org/10.3897/zse.100.128567.suppl2 zse.pensoft.net