This is the accepted version of the journal article: Jianzuo, Qigao; Li, Lu; Madurell-Malapeira, Joan; [et al.]. «The diversification of the lynx lineage during the Plio-Pleistocene-evidence from a new small Lynx from Longdan, Gansu Province, China». Biological Journal of the Linnean Society, Núm. blac054 (2022). DOI 10.1093/biolinnean/blac054 This version is available at https://ddd.uab.cat/record/259564 under the terms of the license 1 The diversification of lynx lineage during the Plio-Pleistocene—— 2 evidence from a new small Lynx from Longdan, Gansu Province, 3 China 4 5 Running title: A new small Lynx from China 6 Qigao Jianzuo a,b,c*, Lu Li b,c,d, Joan Madurell‑Malapeira e, Shiqi Wang b,c, Shijie Li b,c,d, Jiao 7 Fu b,c,d, Shanqin Chen f 8 9 10 a Key Laboratory of Orogenic Belts and Crustal Evolution, School of Earth and Space 11 Sciences, Peking University, 5 Yiheyuan Road, Beijing, 100871 12 b Key Laboratory of Vertebrate Evolution and Human Origins of Chinese Academy of 13 Sciences, Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of 14 Sciences, Beijing, 10044, China 15 c CAS Center for Excellence in Life and Paleoenvironment, Beijing, 100044, China; 16 d University of Chinese Academy of Sciences, Beijing, 100049, China 17 e Institut Català de Paleontologia, Universitat Autònoma de Barcelona, Cerdanyola del 18 Vallès, 08193. Barcelona. Spain. 19 f Hezheng Paleozoological Museum, Hezheng, 731200, China 20 21 22 23 Abstract: A new small-sized lynx from Longdan, Gansu Province, Lynx hei sp. nov., is 24 described in this study. The new species displays the characteristic Lynx generic traits, e.g., 25 distinct buccal grooves in the upper canine, presence of anterior groove in the upper canine, 26 absence of the P2, and moderately developed mastoid process, but it is markedly smaller than 27 the previously described L. issiodorensis specimens from the same site, and is also overall 28 smaller than most living species, comparable to Lynx rufus in size. The new species has a 29 relatively wide and deep zygomatic arch, similar to that of living L. lynx, L. pardinus, L. 30 canadensis but wider than that of L. rufus. Our phylogenetic analyses suggest that L. hei falls 31 within the crown group Lynx, being the sister group to L. rufus, or less probably to be a sister 32 group to L. issiodorensis + three other living species of Lynx. The Plio-Pleistocene L. 33 issiodorensis is supported as the ancestor of L. lynx, L. pardinus and L. canadensis. Our 34 phylogenetic study suggests that Lynx diversification over the Plio-Pleistocene is first 35 achieved by body size differentiation putatively forced by intraspecific competition with 36 other carnivorans, and later followed by morphological divergence. 37 38 39 Key Word: Lynx, diversification, Pleistocene, eastern Asia 40 41 42 43 Introduction 44 The genus Lynx includes several medium-sized cat species with Holarctic distribution 45 (Sunquist and Sunquist, 2002; Bellani, 2019). Four modern species are recognized in the 46 genus, i.e., Eurasian lynx Lynx lynx, Iberian lynx Lynx pardinus, Canadian lynx Lynx 47 canadensis, and bobcat Lynx rufus. The first three species are adaptive to dense forest in the 48 temperate and subarctic regions (Tumlison, 1987; Heptner, 1992; Sunquist and Sunquist, 49 2002; Poole, 2003; Bellani, 2019), whereas L. rufus is more like a generalist, although 50 preferring forest environment when available (Lovallo and Anderson, 1996; Sunquist and 51 Sunquist, 2002; Chamberlain et al., 2003). L. rufus is different from other species in both 52 morphology and habitat, whereas the rest three species are less well distinguished (Werdelin, 53 1981; Tumlison, 1987). Recent phylogeny of multilocus data and genomic scale (Johnson et 54 al., 2006; Li et al., 2016), both supported the first divergence of L. rufus during the middle 55 part of the Pliocene, followed by L. canadensis during the middle Early Pleistocene, and 56 lastly L. pardinus and L. lynx during the late Early Pleistocene. 57 As a largely temperate lineage, Lynx has one of the best fossil records among Felinae 58 (Werdelin, 1981). The discovery of the fossil material of Lynx can be dated back to the early 59 19th century, i.e. Felis issiodorensis (Croizet and Jobert, 1828). More specimens assigned to 60 the Lynx genus were later described in Asia (Teilhard de Chardin and Leroy, 1945), North 61 America (Schultz and Martin, 1972), and even Africa (Hendey, 1974; Geraads, 1980, 2016). 62 Lynx issiodorensis represents the best-recorded and widely spread fossil Lynx from the 63 latest Pliocene to Early Pleistocene (Werdelin, 1981; Kurtén and Werdelin, 1984). In Europe, 64 L. issiodorensis gradually evolved to the now endangered Iberian species Lynx pardinus 65 (Werdelin, 1981; Boscaini et al., 2016; Mecozzi et al., 2021), whereas Lynx lynx probably 66 evolved in Asia population of L. issiodorensis, and did not appear in Europe until the Late 67 Pleistocene (Werdelin, 1981; Mecozzi et al., 2021). However, the records of Lynx in Eastern 68 Asia Early Pleistocene were found in a few localities, best represented the material from 69 Longdan, Linxia Basin (Qiu et al., 2004). Materials from Longdan were referred to Lynx 70 shansius in previous works (Qiu et al., 2004), here L. shansius is considered a junior 71 subjective synonym of L. issiodorensis following the interpretation of Kurtén and Werdelin 72 (1984). 73 In this study, we study a new cranium from Longdan, Linxia Basin (Fig. 1, for detailed 74 geological background, see Qiu et al., 2004). This new cranium is distinctly smaller than 75 those of L. issiodorensis from the same locality, and the same taxon from Europe, further 76 displaying characteristic Lynx traits, representing a previously unknown form and provide 77 key information to understand the early diversification of this iconic Holarctic felid genus. 78 Materials and methods 79 The fossil material described in this study is housed in Hezheng Paleozoological 80 Museum (HM). Fossil and living Felidae specimens (including living species of Felidae, as 81 well as fossil L. issiodorensis from both Europe and Asia) housed in AMNH, USNM, IGF, 82 IVPP, IOZ, KIZ have been observed for comparison. 83 The anatomical terms of the cranial anatomical terms of Jayne (1898) on the cat are 84 followed with additional modification as in Qiu et al. (2014). For the craniodental 85 measurement elements, see Fig. 2 for detail. 86 Our phylogenetic analysis is based on a new morphological matrix, partially from that of 87 Salles (1992) and Werdelin and Flink (2018), including 87 characters for 23 taxa. The matrix 88 is edited using software Mesquite 3.4 (Maddison and Maddison, 2021). The phylogenetic 89 analyses of Maximum Parsimony using software TNT 1.5 (Goloboff et al., 2008; Goloboff 90 and Catalano, 2016) and Bayes Inference using software Mrbayes 3.2.7 (Huelsenbeck and 91 Ronquist, 2001; Ronquist et al., 2012b) were performed. All analyses were set for 4 chains, 92 including three cold chains and one hot chain. Two independent runs were performed, each 93 10 million generations. Constraint on the lineage of the living species using the backbone 94 from phylogeny inferred from genomic data (Johnson et al., 2006; Li et al., 2016) was used. 95 See the Comparison and Discussion section for more detailed settings. 96 97 Institutional Abbreviations—AMNH FM fossil mammal collection of the American 98 Museum of Natural History, New York, USA; AMNH M mammalian collection of the 99 American Museum of Natural History, New York, USA; HMV Hezheng Paleozoological 100 Museum, Hezheng, China; IGF, Natural History Museum of Firenze, Italy; IOZ Institute of 101 Zoology, Chinese Academy of Science, Beijing, China; IVPP Institute of Vertebrate 102 Paleontology and Paleoanthropology, Chinese Academy of Sciences, Beijing, China; KIZ 103 Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, China; USNM 104 Smithsonian National Museum of Natural History, Washington DC, USA 105 Anatomical Abbreviations—BW blade width of the P4 (width across the paracone posterior 106 to the protocone); H height; L length; L1 skull total length; L2 skull condylobasal length; L3 107 skull basilar length; LDP diastema between the C-P3; LT toothrow length from C-P4; M/m 108 upper/lower molar; Lme metacone length; P/p upper/lower premolar; Lpa parastyle length; 109 W width; W1 zygomatic width; W2 rostrum width across P4/M1; W3 rostrum width across 110 the canine. 111 Results 112 Order Carnivora Bowdich (1821) 113 Family Felidae Batsch (1788) 114 Subfamily Felinae Batsch (1788) 115 Lynx Kerr (1792) 116 Lynx hei sp. nov. 117 (Fig. 3) 118 Referred specimen—Type specimen HMV2010, almost complete cranium. 119 Measurements—Table 1 120 Diagnosis —small-sized lynx with similar overall size to that of living Lynx rufus. 121 Rostrum relatively narrow compared with most Lynx species. Postorbital process of frontal 122 small. Sagittal crest anteroposteriorly long and weak. Palate narrow. Basisphenoid- 123 basioccipital bone narrow. Upper canine with distinct anterior groove. P2 absent. P3 main 124 cusp high in buccal view. P4 with moderately sized protocone, without buccal groove. 125 Etymology—In honor to Mr. Wen He, the former curator of Hezheng Museum, who has 126 contributed to the fossil collection and management in Linxia Basin. 127 Type locality—Longdan, Linxia Basin, Gansu Province, China. 128 Distribution and chronology—So far known from the early Early Pleistocene in Linxia 129 Basin (ca. 2.4-2.2Ma). 130 Description—The cranium is complete, but slightly dorsal-ventrally compressed, so the 131 ventro-dorsal height of this cranium is not original, but the width of the cranium is weakly 132 affected. Judging from the dental wear (apex of the P3 slightly worn) and development of 133 sagittal crest, the individual represents a full adult. 134 Dorsal view (Fig. 3C) The rostrum is relatively narrower than most other species of Lynx. 135 There is not distinct postcanine constriction, but the rostrum gets slightly narrower backward. 136 The forehead is slightly wider than the rostrum. The postorbital process of the frontal is 137 distinct but not strong. The temporal ridges unit into the sagittal crest at the point posterior to 138 the frontal-parietal suture. The lambdoid crest is well developed. Lateral view (Fig. 3A) The 139 anterior contour of the naris opening is concave. The highest point of the cranium is located 140 slightly behind the level of the postorbital process of the frontal in lateral view, in the anterior 141 part of the cranium. The sagittal crest is low. The lateral contour of the C-P3 diastema is 142 distinctly dorsally concave. The infraorbital foramen is located at the level of the mesial 143 border of the P3. The anterior border of the orbit is located at the level of the P3 main cusp. 144 The zygomatic arch has a uniform height and is slightly dorsally arched. The postorbital 145 process of the jugal is distinct. The central ridge in the jugal is located around at the middle 146 line of the bone. The mastoid process is distinct. Ventral view (Fig. 3B) The incisor row is 147 nearly straight. The palate is relatively narrow for a lynx. The toothrow gets wider backward. 148 The anterior border of the palatine is located at the level of the P3. The bony choana extends 149 distinctly posterior to the M1. The preglenoid process is relatively weak. The foramen ovale 150 is located medially to the glenoid fossa. The auditory bullae are large, whereas the 151 basisphenoid-basioccipital between the bullae is narrow. The posterior lacerated foramen is 152 located in the same fossa as the hypoglossal foramen. 153 Dentition The incisors are relatively small. The I1 and I2 are in a similar size, whereas 154 the I3 is distinctly larger. There are two strong vertical grooves in the buccal side of the 155 canine. A slightly weaker vertical groove is present in the anterior side of the tooth. There is 156 no P2 in the left, and a tiny root, which seems to be DP2, is present in the right side. The P3 157 main cusp is high crowned in buccal view. There is no mesial accessory cusp, whereas the 158 distal accessory cusp and posterior cingulum are well developed. There is weak distolingual 159 convexity in the P3. The P4 has a straight buccal border. There is no vertical groove in the 160 buccal side of the tooth. The ectoparastyle is absent. The parastyle is relatively short. The 161 protocone is moderate in size, and anteriorly located. The M1 is small and transversely 162 elongated. 163 Comparison and discussion 164 The characteristic craniodental traits of genus Lynx 165 The most widely known trait for Lynx genus is the stable absence of the P2 (Werdelin, 166 1981). We observed this trait in other felids based on observation of large collection in 167 AMNH and USNM. The P2 is nearly always present in Panthera but is sometimes absent in 168 Neofelis and Prionailurus bengalensis. The P2 seems to be frequently absent in some species 169 of Leopardus and in Pardofelis marmorata, and Caracal caracal. It is absent in all species of 170 Lynx we have observed, and all L. issiodorensis from Longdan, Gansu Province, and 171 Caozhuang, Shouyang, Shanxi Province, but in two crania from the Mazegou Formation of 172 the Yushe Basin (Late Pliocene), it is present, though very small (probably DP2). Therefore, 173 the absence of the P2, if taken alone, is not diagnostic for the genus Lynx, as several other 174 felids regularly lack this tooth. The presence of the P2, however, suggests the felid is not very 175 likely to be a Lynx. 176 Furthermore, the upper canines of Lynx genus display another discriminant character. In 177 its buccal surface, two very deep vertical grooves are present in all living species of Lynx 178 (Fig. 4). We studied a large sample of living felid species and found that such deep grooves 179 are not shown by other feline cats, e.g., Caracal or Leptailurus, except some species of 180 Leopardus, e.g., Le. pardalis. More interestingly, there is a distinct groove at the anterior side 181 of the canine, just buccal to the anterior ridge of the canine. We check through the living felid 182 species, and again, this groove is absent in all cats except most species of Leopardus. The 183 grooves in the canine, therefore, provide more useful evidence to judge whether a cat could 184 belong to Lynx genus. 185 The extant lynxes, except for L. canadensis, seldom develops mesial accessory cusp in 186 the P3. It is also absent for all L. issiodorensis from Longdan we observed and rarely present 187 in European L. issiodorensis and Early Pleistocene L. pardinus (Fig. 5). This trait is variable 188 in many felids, e.g., Leptailurus serval, Caracal caracal, Leopardus spp., Prionailurus spp., 189 and Felis spp. The widest part of the tooth lies at the level around the main cusp/distal 190 accessory cusp boundary (which means the distal side of the tooth is wider) in all species of 191 living Lynx except L. rufus. In the latter species, the widest part is often more anteriorly near 192 the apex of the main cusp. In Leopardus spp. and Prionailurus viverrinus, it is more common 193 to see the widest part lies anteriorly, and this point lies posteriorly in Prionailurus 194 bengalensis, Catopuma temminckii, Pardofelis marmorata, but quite variable in Leptailurus 195 serval, and Caracal caracal et al. Therefore, these two P3 traits only provide limited 196 evidence for genus judgment but could be used in species inference. 197 The P4 protocone of all living species of Lynx is often somewhat reduced in size (Fig. 198 5). This trait is commonly seen in L. lynx and L. pardinus, and less often seen in L. 199 canadensis and L. rufus. In L. issiodorensis, the protocone is also often slightly reduced in 200 size. A small preparastyle is quite common in all three living species, whereas in L. rufus, this 201 small cusp is present in around the half specimens we observed. In L. issiodorensis, this cusp 202 is also often present. Of particular interest, we noticed that in L. issiodorensis (both the Late 203 Pliocene and Early Pleistocene populations from Europe and Asia) there are distinct vertical 204 buccal grooves in most specimens (Fig. 6). This trait, however, is absent the living species 205 (occasionally as shallow imprint in L. lynx) or fossil L. pardinus. We regard this trait as an 206 important feature for L. issiodorensis. 207 Regarding the cranial features, Lynx is most characteristic by its broad cranium (wide 208 zygomatic arch, wide rostrum, and wide forehead, see Fig. 7) for a medium-sized cat (smaller 209 cats, e.g., Felis, could have an even broader cranium). This trait, however, can only be 210 applied to L. lynx, L. pardinus and L. canadensis, whereas L. rufus only has a moderately 211 wide cranium, comparable to that of the similar sized other cats, e.g., Catopuma temminckii, 212 Leopardus pardalis, and Leptailurus serval. In lateral view, the ventral border of the palate in 213 the C-P3 diastema is distinctly dorsally concave in all living species of Lynx, as well as in L. 214 issiodorensis. This trait is generally absent or weak in Leopardus spp., Felis spp., 215 Prionailurus spp., but present in Caracal caracal, Catopuma temminckii, Leptailurus serval, 216 Prionailurus viverrinus, though in most of these latter species, the degree of convexity is not 217 so strong as seen in Lynx. The zygomatic arch is rather deep and robust in all three species 218 Lynx except L. rufus. None of the living small to medium-sized cats has such a deep 219 zygomatic arch. The supraorbital process is strong and sharp in all living species of Lynx and 220 being variable L. issiodorensis. In most feline cats, this process is sharp or variable. The 221 mastoid process is moderately developed in all living species of Lynx and L. issiodorensis. 222 The mastoid process is weaker in Felis spp., and most species of Prionailurus spp. but 223 similarly strong in medium-sized felids, e.g., Caracal caracal, Catopuma temminckii, 224 Leptailurus serval, Prionailurus viverrinus, so this trait is more likely a size-related character 225 in the feline. The separation of the posterior lactated foramen and hypoglossal foramen is 226 viewed as one of the most important traits to distinguish L. lynx from L. pardinus, in which 227 the two foramina are located in the same fossa (Boscaini et al., 2016; Mecozzi et al., 2021). 228 Recent observation suggests that the situation in L. issiodorensis is the same as in L. pardinus 229 (Mecozzi et al., 2021). We checked this trait across the living species as well as in fossil 230 Asian species. The confluence the posterior lactated and hypoglossal foramina is present in 231 most species, in L. rufus, L. hei as well as L. pardinus as previously recognized (Fig. 7). 232 However, it is clearly separated in L. canadensis, as in L. lynx. L. issiodorensis mostly has 233 two foramina in the same fossa. In most other felids, these two foramina are in the same 234 fossa, and this state is likely the primitive state for felids, including Lynx. 235 Lynx affinity of the small cranium from Longdan 236 The feline cranium HMV2010 from Longdan has characteristic Lynx craniodental 237 features. The cranium is relatively broad, and the upper canine has two deep buccal grooves, 238 and one distinct anterior groove. The P2 is absent. This trait combination is not seen in any 239 other genera of Felidae. Of noteworthy importance is the feature of the upper canine. As 240 mentioned above, the presence of anterior groove is only present in Lynx among the Old 241 World feline, and at least since the Late Pliocene. In the New World, most species of 242 Leopardus also have deep buccal grooves and distinct anterior groove, but this genus 243 typically retains P2 (except Leopardus colocolo, which frequently lack this tooth), and 244 distinctly lower zygomatic arch, so the Longdan felid is unlikely a member of Leopardus. 245 The Longdan Lynx is also different from all living and known fossil species of Lynx. The 246 body size of Longdan Lynx is distinctly smaller than that of L. issiodorensis from the same 247 locality and other eastern Asian records (e.g., Yushe Basin, Nihewan Basin) and is also 248 smaller than all living species except L. rufus (Fig. 8). The general morphology is similar to 249 that of L. rufus, e.g., relatively narrow rostrum, P4 parastyle not elongated. However, The 250 Longdan Lynx has a wider zygomatic width, and deeper zygomatic arch than those of L. 251 rufus. The other three living species of Lynx have broader cranium and more elongated P4 252 parastyle. In Europe, the early and middle Villafranchian lynx generally has a large body size, 253 but since the late Villafranchian or slightly earlier, some smaller-sized population appeared 254 (Mecozzi et al., 2021; Cuccu et al., 2022). The Longdan felid is still distinctly smaller than 255 these lynx population (P4 of the small individual of late Villafranchian L. issiodorensis is still 256 larger than 17 mm) (Werdelin, 1981). L. issiodorensis from Longdan shows clear affinity to 257 L. lynx in size, dental robustness (Fig. 8A-B), but also differs significantly in relatively P3 258 size (Fig. 8-E), rostrum length and width (Fig. 8G-I). A more detailed analysis of different 259 populations of L. issiodorensis and their comparison with the living species will be the topic 260 for future more specific works. Both the type of L. issiodorensis and the Longdan population 261 of this species are distinctly larger than L. hei in all lineal measurements, and has larger 262 postorbital process of the frontal, but the craniodental proportions of the two species are 263 similar. The type of L. issiodorensis also has a deeper zygomatic arch and proportionally 264 smaller auditory bulla. L. canadensis is only slightly larger than L. rufus, but it differs 265 significantly from the latter in several aspects, e.g., narrower cheek teeth, distinctly longer P4 266 parastyle and wider rostrum. The Longdan felid is therefore represents a new species of Lynx, 267 i.e., L. hei sp. nov. 268 The diversification of Lynx during the Plio-Pleistocene 269 To better study the phylogenetic position of L. hei, and the diversification of Lynx during 270 the Plio-Pleistocene, we performed phylogenetic analyses for all living species of Lynx, L. 271 issiodorensis, L. hei, and most living genera of Felidae, in fact the first phylogenetic analysis 272 ever performed in fossil lynx. Note that our matrix coding of L. issiodorensis is based on 273 materials from Shouyang, Yushe, and Longdan of eastern Asia. 274 In addition to Maximum Parsimony analysis, we perform Bayes Inference analysis 275 under Mkv model (Lewis, 2001), as it outperforms Maximum Parsimony for estimation of 276 phylogeny from discrete morphological data (Wright and Hillis, 2014), but see some debate 277 in Goloboff et al. (2018). Our morphological matrix is based on matrix of Salles (1992) and 278 Werdelin and Flink (2018), and expanded considerably based on our observation of all living 279 and selected fossil species of Felidae. Those only characteristic of machairodonts (subfamily 280 Machairodontinae) were not included in the analysis. Apomorphy for each species is retained 281 when available, as it is informative in the tip-dating analysis (Matzke and Irmis, 2018; Zhang 282 and Wang, 2019). The gray wolf Canis lupus and the fossa Cryptoprocta ferox were chosen as 283 the outgroup for Felidae following the choice of Werdelin and Flink (2018). Two Miocene 284 felids Proailurus lemanensis and Hyperailurictis intrepidus were chosen to represent the 285 primitive members of the subfamily. The majority of the living genera (here including 11 286 genera in addition to Lynx) were included in the analyses. In Bayes Inference analyses, we 287 performed both ordinary and tip-dating methods (Ronquist et al., 2012a). In the latter method, 288 we did not include molecular data, so the tip-dating mainly acts as a constraint on branch 289 length. As our fossil sampling of fossil feline is insufficient, we did not analyze the divergent 290 time inferred from the tip-dating method, which can be more reliable when more fossil 291 members were included in the analyses. 292 In two of the three analyses (MP, BI, tip-dating BI), the monophyly of Lynx is not 293 supported, and Otocolobus manul mostly groups within Lynx. Even though molecular 294 phylogeny clearly suggested that O. manul has a closer relationship to Prionailurus (Johnson 295 et al., 2006; Li et al., 2016), our result is expected from a morphological aspect, as there is 296 considerable convergence. O. manul has a broad cranium, wide forehead, absence of the P2, 297 and reduction of the P4 protocone, elongated m1, that are also present in three living species 298 of Lynx (not in L. rufus). In BI analysis, Acinonyx jubatus groups into Lynx, as there is also 299 convergence between them, e.g., p4 anterior accessory cuspid enlarged, deep mandibular 300 corpus. In MP analysis, the two species of Panthera are grouped with Acinonyx, together in 301 Lynx, in a similar result of convergence. In the final analyses, we use constraint forcing 302 Otocolobus grouped with Prionailurus, and Acinonyx grouped with Puma, as we regard that 303 molecular lineage defined by Johnson et al. (2006) as correct, considering its data size. 304 Nevertheless, all Bayes analyses (Fig. 9 and appendix), L. hei and L. issiodorensis are 305 grouped within the crown group of Lynx. In MP analysis, L. issiodorensis is grouped with L. 306 lynx, L. pardinus and L. canadensis, whereas the position of L. hei is not resolved in the strict 307 consensus tree. We checked the original equal parsimony tree and found that this species is 308 either sister to L. rufus, or sister to the group of L. issiodorensis + three living species of 309 Lynx. In the BI analyses after constraining Otocolobus and Acinonyx out of Lynx, both tip- 310 dating and ordinary analyses support the sister group L. hei and L. rufus, and L. issiodorensis 311 to the other three species of Lynx, but in ordinary BI analysis, the posterior probability for the 312 internal node within Lynx is rather low. In the tip-dating BI analysis without constraint, L. 313 issiodorensis is grouped with L. hei + L. rufus, but with rather low posterior probability for 314 this node. In most analyses (e.g., ordinary BI and MP analyses without constraint), the Asiatic 315 golden cat Catopuma is supported as a sister group to Lynx. This result is congruent with the 316 recent genomic phylogeny (Li et al., 2016). Another relationship of the recent genomic 317 phylogeny, which Felis and Prionailurus are sister groups, is also supported in our analyses. 318 Some other relationship, e.g., the sister group of Puma and Acinonyx, is not supported in any 319 non-constraint analyses. This result is likely due to the lacking in intermedium form between 320 the much more specialized Acinonyx and more conservative Puma, leaving a considerable 321 morphological gap between them. Adding more fossil members will be helpful to break this 322 long branch. 323 Our analyses provide interesting new information on the diversification of the crown 324 members of Lynx. The fossil records of Lynx are quite common in Eurasia, but mostly 325 assigned to L. issiodorensis, which was supposed to be the common ancestor for the three 326 other living species of Lynx, suggesting a simple widely distributed common ancestor- 327 vicariance-speciation model. Our new finding suggested that Lynx crown group already 328 experienced diversification during the Plio-Pleistocene. Molecular phylogenetic dating (Li et 329 al., 2016) suggests a middle Pliocene divergence bobcat L. rufus to the rest of living species. 330 Our new fossil and phylogeny support such a scenario. Lynx began to diverge into two 331 ecomorphologically different species during the middle Pliocene, differs mainly by their body 332 size. One population of the small-sized lineage migrated to North America and evolved there 333 since then, finally evolved to the living L. rufus, whereas L. hei represents the population of 334 this small-sized lineage retained in eastern Asia. At this evolutionary level, both large and 335 small-sized lineage still have primitive craniodental traits, e.g., skull relatively narrow, and 336 cheek teeth not enlarged. These traits are retained in L. rufus, whereas L. issiodorensis 337 experienced to the second wave of diversification since the Pleistocene, and acquired more 338 derived traits, e.g., broad skull, shorter rostrum. 339 It is interesting to note that after the larger sized lineage acquirement of a broad skull, 340 shorter rostrum, it then began to occupy the niche of smaller body size: living L. canadensis 341 has a comparative body size similar to L. rufus. This pattern suggests the first divergence of 342 ecomorphology is achieved by evolving into different body size, and during the evolutionary 343 history, the two lineages gradually their unique morphology, large enough to allow them to 344 explore different niches without body size differences, as is seen in the living L. canadensis 345 and L. rufus (Fig. 10). 346 Conclusion 347 The new Early Pleistocene fossil cranium from Longdan described here, represents a 348 new species of lynx, L. hei, phylogenetically closely related to the modern North American L. 349 rufus. Our phylogeny and comparative analyses suggest that the crown group of Lynx 350 experienced its initial diversification during the mid-Pliocene in Eurasia into two lineages 351 i.e., L. hei-rufus, and L. issiodorensis-other living species. The two lineages mainly differ in 352 their body size during the Plio-Pleistocene. Whereas latter larger sized lineage acquired new 353 craniodental morphology, and experienced a second wave of diversification in radiation in 354 body size/morphology during the Pleistocene. 355 356 Acknowledgements: We thank J. Meng, R. O’Leary, and J. Galkin for their help in accessing 357 the AMNH fossil mammal collections; M. Surovy, E. Hoeger, and S. Ketelsen for their help 358 in accessing the AMNH modern mammal collections; A. Millhouse, D. Lunde and J. J. 359 Ososky for their help in accessing the USNM fossil and modern mammal collections; and Z. 360 Qiu and J. Chen, W. He, S. Chen for help in accessing fossil collections of the IVPP and 361 HMV. The current work was supported by the Strategic Priority Research Program of 362 Chinese Academy of Sciences (Grant No. XDB26000000 and XDA20070203), Key Frontier 363 Science Research Program of the Chinese Academy of Sciences (Grant Nos. QYZDY-SSW- 364 DQC-22 and GJHZ1885), the National Natural Science Foundation of China (Grant Nos. 365 42102001, 41430102, 41872001, 41872005 and 41772018), Second Comprehensive 366 Scientific Expedition on the Tibetan Plateau [2019QZKK0705], China Scholarship Council, 367 and Frick Fund, Division of Vertebrate Paleontology, AMNH. J.M.-M. was funded by the 368 Agencia Estatal de Investigación-European Regional Development Fund of the European 369 Union (CGL2017-82654-P, AEI/FEDER-UE) and the Generalitat de Catalunya (CERCA 370 Program). J.M.-M. is member of the consolidated research group 2017 SGR 116 (AGAUR, 371 Generalitat de Catalunya). 372 373 References 374 Batsch, A.J.G.C., 1788. Versuch einer Anleitung, zur Kenntniβ und Geschichte der Thiere und Mineralien, für 375 akademische Vorlesungen entworfen und mit den nöthigsten Abbildungen versehen Erster Theil. 376 Academischen Buchhandlung, Jena. 377 Bellani, G.G., 2019. Felines of the World: Discoveries in Taxonomic Classification and History. Academic Press, 378 London. 379 Boscaini, A., Alba, D.M., Beltrán, J.F., Moyà-Solà, S., Madurell-Malapeira, J., 2016. Latest Early Pleistocene 380 remains of Lynx pardinus (Carnivora, Felidae) from the Iberian Peninsula: taxonomy and evolutionary 381 implications. Quaternary Science Reviews 143, 96-106. 382 Bowdich, T.E., 1821. An Analysis of the Natural Classification of Mammalia, for Use of Students and Travelers. 383 J. Smith, Paris. 384 Chamberlain, M.J., Leopold, B.D., Conner, L.M., 2003. Space use, movements and habitat selection of adult 385 bobcats (Lynx rufus) in central Mississippi. The American Midland Naturalist 149, 395-405. 386 Croizet, J.B., Jobert, A.C.G., 1828. Recherches sur les ossemens fossiles du département du Puy-de-Dôme. Paris: 387 Chez les principaux libraires, 1-224. 388 Cuccu, A., Valenciano, A., Azanza, B., DeMiguel, D., 2022. A new lynx mandible from the Early Pleistocene of 389 Spain (La Puebla de Valverde, Teruel) and a taxonomical multivariate approach of medium-sized felids. 390 Historical Biology, 1-12. 391 Geraads, D., 1980. Un nouveau felide (Fissipeda, mammalia) du pleistocene moyen du Maroc: Lynx thomasi N. 392 sp. Geobios 13, 441-444. 393 Geraads, D., 2016. Pleistocene Carnivora (Mammalia) from Tighennif (Ternifine), Algeria. Geobios 49, 445–458. 394 Goloboff, P.A., Catalano, S.A., 2016. TNT version 1.5, including a full implementation of phylogenetic 395 morphometrics. Cladistics 32, 221-238. 396 Goloboff, P.A., Farris, J.S., Nixon, K.C., 2008. TNT, a free program for phylogenetic analysis. Cladistics 24, 774- 397 786. 398 Goloboff, P.A., Torres, A., Arias, J.S., 2018. Weighted parsimony outperforms other methods of phylogenetic 399 inference under models appropriate for morphology. Cladistics 34, 407-437. 400 Hendey, Q.B., 1974. The Late Cenozoic Carnivora of the Southwestern Cape Province South Africa. Annals of 401 the South African Museum 63, 1-369. 402 Heptner, V.G., 1992. Mammals of the Soviet Union, Volume 2 Part 2 Carnivora (Hyenas and Cats). Brill. 403 Huelsenbeck, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17, 404 754-755. 405 Jayne, H., 1898. Mammalian Anatomy, Part I. The Skeleton of the Cat. Lippincott Company, London. 406 Johnson, W.E., Eizirik, E., Pecon-Slattery, J., Murphy, W.J., Antunes, A., Teeling, E., O'Brien, S.J., 2006. The late 407 Miocene radiation of modern Felidae: a genetic assessment. Science 311, 73-77. 408 Kerr, R., 1792. The animal kingdom or zoological system, of the celebrated Sir Charles Linnaeus…, London. 409 Kurtén, B., Werdelin, L., 1984. The relationships of Lynx shansius Teilhard. Annales Zoologici Fennici, 129-133. 410 Lewis, P.O., 2001. A likelihood approach to estimating phylogeny from discrete morphological character data. 411 Syst Biol 50, 913-925. 412 Li, G., Davis, B.W., Eizirik, E., Murphy, W.J., 2016. Phylogenomic evidence for ancient hybridization in the 413 genomes of living cats (Felidae). Genome Res 26, 1-11. 414 Lovallo, M.J., Anderson, E.M., 1996. Bobcat (Lynx rufus) home range size and habitat use in northwest Wisconsin. 415 American Midland Naturalist, 241-252. 416 Maddison, D.R., Maddison, W.P., 2021. Mesquite: a modular system for evolutionary analysis. Version 3.40 417 https://www.mesquiteproject.org/home.html. 418 Matzke, N.J., Irmis, R.B., 2018. Including autapomorphies is important for paleontological tip-dating with 419 clocklike data, but not with non-clock data. PeerJ 6, e4553. 420 Mecozzi, B., Sardella, R., Boscaini, A., Cherin, M., Costeur, L., Madurell-Malapeira, J., Pavia, M., Profico, A., 421 Iurino, D.A., 2021. The tale of a short-tailed cat: New outstanding Late Pleistocene fossils of Lynx 422 pardinus from southern Italy. Quaternary Science Reviews 262, 106840. 423 Poole, K.G., 2003. A review of the Canada lynx, Lynx canadensis, in Canada. The Canadian Field-Naturalist 117, 424 360-376. 425 Qiu, Z.X., Deng, T., Wang, B.Y., 2004. Early pleistocene mammalian fauna from Longdan, Dongxiang, Gansu, 426 China. Science Press, Beijing. 427 Qiu, Z.X., Deng, T., Wang, B.Y., 2014. A Late Miocene Ursavus skull from Guanghe, Gansu, China. Vertebrata 428 PalAsiatica 52, 265-302. 429 Ronquist, F., Klopfstein, S., Vilhelmsen, L., Schulmeister, S., Murray, D.L., Rasnitsyn, A.P., 2012a. A total- 430 evidence approach to dating with fossils, applied to the early radiation of the hymenoptera. Syst Biol 61, 431 973-999. 432 Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Hohna, S., Larget, B., Liu, L., Suchard, 433 M.A., Huelsenbeck, J.P., 2012b. MrBayes 3.2: efficient Bayesian phylogenetic inference and model 434 choice across a large model space. Syst Biol 61, 539-542. 435 Salles, L.O., 1992. Felid phylogenetics: extant taxa and skull morphology (Felidae, Aeluroidea). American 436 Museum Novitates 3047, 1-67. 437 Schultz, C.B., Martin, L.D., 1972. Two lynx-like cats from the Pliocene and Pleistocene. Buletin of the University 438 of Nebraska State Museum 9, 197-203. 439 Sunquist, M., Sunquist, F., 2002. Wild cats of the world. University of Chicago press, Chicago. 440 Teilhard de Chardin, P., Leroy, P., 1945. Les Felides de Chine. Institut de géobiologie, Pekin. 441 Tumlison, R., 1987. Felis lynx. Mammalian species 269, 1-8. 442 Werdelin, L., 1981. The evolution of lynxes, Annales Zoologici Fennici. JSTOR, pp. 37-71. 443 Werdelin, L., Flink, T., 2018. The phylogenetic context of Smilodon, in: Werdelin, L., Mcdonald, H.G., Shaw, 444 C.A. (Eds.), Smilodon: The Iconic Sabertooth. Johns Hopkins University Press, Baltimore, pp. 14-29. 445 Wright, A.M., Hillis, D.M., 2014. Bayesian Analysis Using a Simple Likelihood Model Outperforms Parsimony 446 for Estimation of Phylogeny from Discrete Morphological Data. PloS one 9, e109210. 447 Zhang, C., Wang, M., 2019. Bayesian tip dating reveals heterogeneous morphological clocks in Mesozoic birds. 448 R Soc Open Sci 6, 182062. 449 450 Figure captions 451 452 Fig. 1. Location of Longdan. 453 Fig. 2. Craniodental measurement of Lynx used in this study. For measurements see 454 Abbreviation section. 455 Fig. 3. Holotype of Lynx hei sp. nov, HMV2010, from Longdan. 456 Fig. 4. Anterior view of rostrum of various Lynx species, marking the presence of the 457 anterior groove in the upper canine. A. L. hei sp. nov., HMV2010; B. L. issiodorensis, 458 HMV2015; C. L. canadensis, USNM212620; D. L. rufus, AMNH 24666. Not to scale. 459 Fig. 5. Comparison of occlusal view of upper cheek tooth rows of Lynx spp.. A. L. 460 issiodorensis, A1. HMV2012, Longdan; A2. AMNH 96-B1042 (reversed/mirrored), 461 Xiazhuang, and A3. AMNH 62-B756, both Shouyang; B. L. hei sp. nov., HMV2010, 462 Longdan; C. L. rufus, C1. AMNH M14822; C2. AMNH M243108; D. L. pardinus, D1. 463 AMNH M169492; D2. USNM 152619; E. L. canadensis, E1. USNM 077095; E2. USNM 464 212617; F. L. lynx, F1. AMNH 41337; F2. IOZ T0139. 465 Fig. 6. Comparison of buccal view of Lynx spp. A. Lynx canadensis, USNM 025726; B. L. 466 rufus, USNM 243108; C. L. hei sp. nov., HMV 2010; D. L. issiodorensis, ; E. L. 467 issiodorensis, Longdan, HMV2016; F. L. issiodorensis, Xiazhuang, Shouyang, AMNH 96- 468 B,1024. Not to scale. 469 Fig. 7. Crania of Lynx hei, and selected living felid species, lateral, ventral and dorsal 470 views. A. L. hei, HMV2010; B. L. rufus, AMNH M24665; C. L. candaensis, USNM33126; 471 D. L. pardinus, AMNH M169492; E. L. lynx, IOZ09419, F. Caracal caracal, AMNH 472 M116512. 473 Fig. 8. Craniodental measurements and ratio of Lynx hei sp. nov. and related taxa. A. P3 474 length vs width/length ratio; B. P4 length vs width/length ratio; C. P4 length vs blade 475 width/length ratio; D. P4 length vs parastyle length/length ratio; E. P4 length vs P3/P4 length 476 ratio; F. condylobasal length vs toothrow length; G. toothrow length vs diastema/toothrow 477 length ratio; H. condylobasal length vs rostrum width cross P4 and m1/condylobasal length 478 ratio; I. condylobasal length vs rostrum width cross C/condylobasal length ratio. 479 Fig. 9. Bayes Inference tip-dating analysis, with constraint on Otocolobus with 480 Prionailurus. See the results of other phylogenies in appendix. 481 Fig. 10. Model of diversification pattern of Lynx proposed in this paper. The specimen of 482 L. issiodorensis HMV2015 is from Longdan, China. 483 484 1993 old No. No. note Character 0 1 2 3 1 canine Upper canine morphology height reduced (<0.13) crown not elongated (~13–20% of comoderately elongate (~20–25%) 25-35 (>35% of CBL) 2 canine Upper canine buccal groove 0 1 2 3 3 canine Upper canine lingual groove 0 1 2 4 canine Lower canine buccal groove 0 1 2 4 5 canine lower canine morphology caniniform and large small very small and incisiform 6 M1 M1 transverelg shortening not shortened (>2) shortened (<2) 7 M1 M1 cusp presence of large, multicuspid M1 si presence of reduced, multi-cuspid Mno distinct cusp 7 8 P4 P4 protocone morphology presence of large, usually cusped proprotocone small but still distinct protocone very small or absent 8 9 P4 P4 protocone angulation P4 protocone angulation strongly anteriorly directed lingually directed 9 10 P4 P4 protocone position anterior to the paracone-parastyle junat the paracone-parastyle junction posterior to the paracone-parastyle junction 10 11 P4 P4 parastyle anteroposterior length absent or cingulum-like small (<0.2) moderate (0.2-0.25) large (>0.25) 12 P4 P4 buccal concavity absent one two 11 13 P4 P4 preparastyle size absent small large 14 P4 P4 preparastyle direction nearly perpendicular to parastyle with distinct angle with parastyle nearly in line with parastyle 15 P4 P4 buccal vertical groove absent present 12 16 P4 Relative size of P4 P4 short relative to PL (<25% of PL P4 long relative to PL (>25% of PL) 13 17 P3 Relative length P3/P4 P3 long relative to P4 (>60% of LP4P3 moderate (ca. 50–60% of LP4) P3 short (ca. 35–50% of LP4) P3 very short (<35% of LP4) 14 18 P3 P3 posterior cingulum absent or weak distinct cusp-like 15 19 P3 P3 anterior accessory cusp absent present large 20 P3 P3 anterior cingulid cusp absent often present 21 P3 anterior end of P3 anterior ridge close to the anterior border distinctly lingually 9 22 P3 P3 lingual ridge absent present 16 23 P3 Greatest width of P3 at main cusp posterior part distal part 17 24 P2 P2 morphology large with multiple cusps present but very small, usually only wabsent 25 P2 P2 root double single 18 26 P1 P1 morphology present present but very small absent 19 27 m3 m3 presence Present absent 20 28 m2 m2 presence Present absent 21 29 m1 m1 metaconid-talonid morphology large talonid basin with distinct hypolarge without basin small absent 30 m1 m1 metaconid and talonid differential differential undistinguishable 31 m1 m1 widest part posterior to the paraconid-protoconidat the paraconid-protoconid notch slightly anterior to the paraconid-pro near the middle part of the paraconid 32 m1 m1 robustness (W/L) slender (mostly <45%) robust (~50%) 33 m1 anterior border vertical or slightly posteriorly inclinedistinctly posteriorly inclined 24 34 m1 Relative size of m1 short relative to mandible length (<1 long relative to mandible length (>15%) 22 35 p4 p4 relative length long (>80% of m1 length) short (<80% of m1 length) 25 36 p4 p4 paraconid direction posterior inclined vertical 37 p4 p4 paraconid size distinctly smaller than the posterior asimilar to the posterior accessory cuslarger than the posterior accessory cuspid 38 p4 p4 posterior cingulid absent or weak clearly distinguished cusp-like 39 p4 middle constriction generally absent generally present 47 40 p3 p3 size (versus m1 ) small (<0.3%) or absent relatively small (0.3-0.5) large 41 p3 p3 height same or slightly lower than the p4 (>distinctly lower than the p4 (<0.7) 42 p3 p3 anterior accessory cusp absent indistinct distinct 43 p3 p3 posterior cingulid absent or weak clearly distinguished cusp-like 26 44 p2 p2 presence Present absent 27 45 p1 p1 presence Present absent 46 cranium dorsal concavity between C-P4 absent present 47 cranium zygomatic arch deepth narrow deep 31 48 cranium Anterior border of orbit at anterior border of P3 P3 P3/P4 to P4 paracone P4 metacone 34 49 cranium Infraorbital fenestra position at anterior border of P3 P3 at around P3/P4 posterior to P3/P4 50 cranium Infraorbital fenestra size very small (Acinonyx) small (Felis) large (Panthera) very large (Smilodon) 32 51 cranium Snout area elevation compared with braincase snout area low snout area distinctly elevated 52 cranium cranial dorsal contour nearly straight with distinct angle rounded 53 cranium forehead width distinctly wider than the rostrum similar to the rostrum 35 54 cranium extension of posterior border of the nasals relative to maxi not extended similar more extended 45 55 cranium posterior arch nasals weakly arched arched 56 cranium posterior border shape of the nasals smooth pointed 57 cranium anterior process of frontal weak strong 58 cranium nasal width narrow wide 37 59 cranium nasal middle constriction (Rostral constriction) absent present 38 60 cranium Nasal curvature absent present 61 cranium depression at the posterior end of nasal absent present 31 62 cranium Frontal bone, outer surface depression absent present 43 63 cranium anterior border of jugal strongly anteriorly inclined moderately anteriorly inclined 44 64 cranium Lower rim of the orbit vertical flat 36 65 cranium Size of the jugal postorbital process small, rounded, often almost absent rounded triangle long, pointed 37 66 cranium Angle of the occipital condyles low (~45–60) high (~70–100) 38 67 cranium rostrum width vs palatine width narrow, <0.7 wide 0.7-0.8 very wide >0.8 39 68 cranium palatine width versus CBL widest part of the palate comparativewidest part of the palate comparativewidest part of the palate comparatively narrow wide (>45%) 40 69 cranium Alisphenoid canal Present absent 70 cranium occipital condyle position low, much lower than the middle po high, close to the middle point of inion and mastoid process 46 71 cranium Orientation of P3 and P4 tiny angle or nearly in line small angle large angle 72 cranium extension of bony choana at or slightly posterior to the M1 distinctly posterior to the M1 73 cranium palate concavity medial to the M1 absent or weak distinct 74 cranium position of rounded foramen very close to the orbital fissure at the middle of orbital fissure and oval foramen 75 cranium pterygoid fossa absent small distinct 76 cranium middle ear opening moderate enlarged 77 cranium sagittal crest always present often or always absent 78 cranium postorbital bar absent often present 79 cranium size of ectotympanic distinctly smaller than the entotympaenlarged to similar size of ectotympanic 80 cranium anterior projection of jugal absent present 81 cranium postorbital constriction absent or indistinct distinct 82 cranium posterior expansion of frontal absent or very weak distinct 83 mandible number of mental foramina mostly double mostly single 84 mandible mandible height at canine very low < 0.9 moderate (0.9-1.1) high (1.1-1.3) very high (>1.3) 85 mandible lower border profile nearly straight distinctly convex 43 86 mandible Mandibular fossa termination posterior to the carnassial well anterior of the posterior edge of the carnassial, frequently terminating around carnassial saddle or even at the m1/p4 junction 44 87 mandible Posterior part of horizontal dentary ramus height relative t slender (~15–19%) robust (~20–23%) very robust >0.23 catalogue Locality Age taxa CL CW CH CW/L CH/L P3L P3W P3H P3W/L P3H/L P4L P4W P4BW P4PaL P4MeL 前尖高PH P4W/L P4BW/L P4PL/L P4ML/L M1L M1W M1W/L C.P4 P3.P4 L1 L2 L3 LT W1 W2 W3 LDP HMV 2010 X1813 Longdan Early PleistoLynx hei 7,7 5,57 17,68 0,723 2,296 9,44 4,9 0,519 14,52 7,19 5,3 2,63 5,61 0,495 0,365 0,181 0,386 2,85 5,42 1,902 0,53 0,65 0,373 120,64 114,49 107,91 37,63 88,27 46,46 32,67 HMV 2010 X1813 Longdan Early PleistoLynx hei 7,71 6,15 17,31 0,798 2,245 14,66 7,21 5,36 2,05 5,34 0,492 0,366 0,14 0,364 2,35 5,55 2,362 0,526 0,379 41,01 88,27 8,55 V13539 Longdan Early PleistoLynx shansius 13,23 10,11 0,764 20,12 9,12 6,5 4,37 7,35 0,453 0,323 0,217 0,365 V13539 Longdan Early PleistoLynx shansius 13,1 6,31 11,04 0,482 0,843 62,89 41,91 7,43 V13540 Longdan Early PleistoLynx shansius 12,07 6,29 9,57 0,521 0,793 47,31 97,92 61,24 37,68 7,78 V13540 Longdan Early PleistoLynx shansius 9,28 7,43 20,78 0,801 2,239 18,78 8,92 6,44 3,97 6,71 0,475 0,343 0,211 0,357 47,16 HMV2013 X1799 Longdan Early PleistoLynx shansius 9,97 8,42 0,845 11,71 6,23 9,29 0,532 0,793 17,7 8,49 6,55 3,25 0,48 0,37 0,184 0,563 0,662 48,44 63,1 40,65 8,93 HMV2013 X1799 Longdan Early PleistoLynx shansius 9,23 8,32 0,901 11,09 5,87 9,65 0,529 0,87 HMV2016 X2126 Longdan Early PleistoLynx shansius 10,38 12,7 19,32 3,67 7,5 0,19 0,388 0,537 0,657 180,48 53,4 9,15 HMV2016 X2126 Longdan Early PleistoLynx shansius 10,31 13,3 19,45 3,93 7,36 0,202 0,378 0,53 0,684 53,35 59,63 42,6 8,32 HMV2015 X1849 Longdan Early PleistoLynx shansius 9,62 23,47 2,44 19,36 4,22 7,05 0,218 0,364 0,497 172,42 152,18 50,51 107,62 58,14 45,14 7,74 HMV2015 X1849 Longdan Early PleistoLynx shansius 9,64 22,19 2,302 12,92 19 4,45 6,99 0,234 0,368 0,507 0,68 50,89 8,65 HMV2017 01429 Longdan Early PleistoLynx shansius 10,52 HMV2017 01429 Longdan Early PleistoLynx shansius 10,48 12,29 19,72 0,531 0,623 171,02 151,32 49,95 106,27 60,27 43,96 8,07 HMV2014 X1821 Shitougou Early PleistoLynx shansius 9,96 12,14 20,76 0,48 0,585 166,52 153,82 52,11 113,14 63,89 44,87 8,42 HMV2014 X1821 Shitougou Early PleistoLynx shansius 9,92 20,46 0,485 49,61 HMV 2011 X1775 Longdan Early PleistoLynx shansius 9,05 6,86 0,758 11,21 5,96 0,532 18,56 8,41 6,63 2,89 6,89 0,453 0,357 0,156 0,371 4,46 6,76 0,488 0,604 0,364 137,73 HMV 2011 X1775 Longdan Early PleistoLynx shansius 8,79 7,44 21,79 HMV 2012 X1781 Longdan Early PleistoLynx shansius 8,38 6,3 10,88 5,4 17,17 5,6 2,99 6,31 9,3 0,326 0,174 0,368 3,12 5,71 0,488 0,634 0,333 157,72 137,52 45,82 55,06 39,47 9,58 HMV 2012 X1781 Longdan Early PleistoLynx shansius 7,93 6,05 10,9 5 16,4 7,49 5,47 3,17 5,76 0,457 0,334 0,193 0,351 2,78 5,55 0,484 0,665 0,338 45,78 9,34 X1812 Longdan Early PleistoLynx shansius 9,22 7,14 12,16 5,63 17,84 0,517 0,682 45,64 7,01 X1812 Longdan Early PleistoLynx shansius 11,1 5,53 17,87 0,621 43,65 5,57 X1868 Longdan Early PleistoLynx shansius 11,35 9,9 12,47 6,51 9,62 18,81 9,37 7,68 0,498 0,408 0,603 0,663 145,97 50,68 8,94 X1868 Longdan Early PleistoLynx shansius 13,02 51,87 55,37 39,88 9,02 X1870 Longdan Early PleistoLynx shansius 8,96 7,5 20,01 12,51 6,24 10,8 18,33 9,27 6,96 0,506 0,38 0,489 0,682 47,63 65,22 44,64 8,09 X1870 Longdan Early PleistoLynx shansius 8,89 7,1 HMV1230 Longdan Early PleistoLynx shansius 10,4 12,07 5,65 18,24 0,57 0,662 158,82 146,28 50,1 112,69 59,65 42,76 10,46 HMV1230 Longdan Early PleistoLynx shansius 9,82 50,54 9,37 HMV1226 Longdan Early PleistoLynx shansius 10,44 7,8 12,94 6,29 10,55 19,49 9,53 6,94 3,69 7,16 10,98 0,489 0,356 0,189 0,367 3,31 6,32 0,536 0,664 0,324 161,02 150,91 52,49 110,53 62,66 42,79 9,31 HMV1226 Longdan Early PleistoLynx shansius 10,8 8,37 24,54 13,12 6,1 10,74 19,85 9,49 6,97 4,37 7,29 10,8 0,478 0,351 0,22 0,367 5,83 0,544 0,661 0,294 52,23 10,49 AMNH M4137 Romania modern Lynx lynx 9,2 6,92 20,43 0,752 2,221 11,5 6,44 9,76 0,56 0,849 17,6 8,16 7 3,4 6,26 10,69 0,464 0,398 0,193 0,356 3,3 5,96 1,806 0,523 0,653 0,339 135,46 121,35 42,93 93,14 57,68 35,66 4,1 AMNH M4137 Romania modern Lynx lynx 9,26 6,95 19,66 0,751 2,123 11,58 6,02 10,26 0,52 0,886 17,6 8,29 6,93 3,54 5,99 11,01 0,471 0,394 0,201 0,34 3,22 5,96 1,851 0,526 0,658 0,339 43,98 5,06 IOZ08119 辉南 modern Lynx lynx 10,3 7,6 21,1 0,738 2,049 12,1 7 9,7 0,579 0,802 20,52 9,66 6,6 3 7,7 0,471 0,322 0,146 0,375 3,5 6 1,714 0,502 0,59 0,292 158,32 143,22 51,22 116,18 68,62 44,44 5,98 IOZ08119 辉南 modern Lynx lynx 10,1 7,9 0,782 12,3 7,1 10 0,577 0,813 20,72 9,72 6,9 3,3 7,5 11,4 0,469 0,333 0,159 0,362 3,6 6,7 1,861 0,487 0,594 0,323 50,74 6,78 IOZ09426 黑龙江密山modern Lynx lynx 10 7,5 0,75 11,8 6,8 0,576 19,3 9,6 6,7 3,1 6,7 10,9 0,497 0,347 0,161 0,347 3,6 6,3 1,75 0,518 0,611 0,326 155,42 142,12 48,22 110 68,42 42,52 6,9 IOZ09426 黑龙江密山 modern Lynx lynx 9,5 7,5 0,789 11,6 7 9,8 0,603 0,845 19,7 9,6 7,1 3,3 6,9 11,1 0,487 0,36 0,168 0,35 3,3 6 1,818 0,482 0,589 0,305 48,92 6,8 IOZ07809 吉林敦化 modern Lynx lynx 8,52 7,14 20,54 0,838 2,411 11,18 5,94 10,42 0,531 0,932 20 8,74 6,52 4,36 8 11,86 0,437 0,326 0,218 0,4 3,92 6,82 1,74 0,426 0,559 0,341 146,28 131,62 45,62 101,42 61,52 38,64 5,92 IOZ07809 吉林敦化 modern Lynx lynx 8,86 7,16 21 0,808 2,37 11,76 6,02 10,72 0,512 0,912 20,14 8,64 6,56 4,22 7,9 11,38 0,429 0,326 0,21 0,392 3,9 6,8 1,744 0,44 0,584 0,338 45,86 5,9 IOZ09422 大约是东北modern Lynx lynx 10,92 8,92 0,817 12,9 6,74 10,74 0,522 0,833 20,18 9,32 7,12 4,32 7,72 11,5 0,462 0,353 0,214 0,383 4,12 6,16 1,495 0,541 0,639 0,305 163,24 148,82 50,12 116,04 70,08 44,72 6,92 IOZ09422 大约是东北modern Lynx lynx 10,5 8,82 23,38 0,84 2,227 12,94 7,08 10,92 0,547 0,844 20,32 9,24 7,22 4,24 7,48 12,54 0,455 0,355 0,209 0,368 4,48 7,3 1,629 0,517 0,637 0,359 148,82 49,02 6,68 IOZ08120 吉林安？ modern Lynx lynx 9,52 7,32 21 0,769 2,206 12 6,32 10 0,527 0,833 19,02 9,32 6,8 4,36 6,82 10,72 0,49 0,358 0,229 0,359 4,18 7,18 1,718 0,501 0,631 0,377 148,74 134,52 45,62 108,4 61,52 38,48 6,02 IOZ08120 吉林安？ modern Lynx lynx 8,84 7,12 20,2 0,805 2,285 11,74 6,32 9,72 0,538 0,828 18,84 9,22 6,74 4,02 7,02 11,02 0,489 0,358 0,213 0,373 4,06 7,08 1,744 0,469 0,623 0,376 45,58 6,34 IOZ09419 大约是东北modern Lynx lynx 10,02 7,9 0,788 12,18 6,68 0,548 19,42 9,32 7 4,02 7,62 11,28 0,48 0,36 0,207 0,392 3,92 7,22 1,842 0,516 0,627 0,372 149,12 135,42 48 105,7 65,1 43,64 7,18 IOZ09419 大约是东北modern Lynx lynx 10 8,08 0,808 12,06 6,56 0,544 19,4 9,38 6,98 4,04 7,38 11,46 0,484 0,36 0,208 0,38 3,94 7,1 1,802 0,515 0,622 0,366 47,66 6,46 IOZ T0139 西藏 modern Lynx lynx 8,28 6,68 0,807 11,38 6,02 9,76 0,529 0,858 18,72 8,72 6,48 3,54 6,88 10,62 0,466 0,346 0,189 0,368 3,64 6,96 1,912 0,442 0,608 0,372 134,72 122,34 42,62 95,44 58,76 34,78 4,54 IOZ T0139 西藏 modern Lynx lynx 8,12 6,58 0,81 11,34 6,02 9,82 0,531 0,866 18,84 8,58 6,48 4 6,92 10,86 0,455 0,344 0,212 0,367 3,66 6,76 1,847 0,431 0,602 0,359 42,62 5,4 IOZ T0138 西藏 modern Lynx lynx 8,76 7,42 0,847 12,52 7,04 0,562 19,62 9,12 6,34 4,1 7,18 0,465 0,323 0,209 0,366 4,52 6,42 1,42 0,446 0,638 0,327 142,34 129,54 45,74 100,32 60,92 39,62 4,96 IOZ T0138 西藏 modern Lynx lynx 9 7,32 0,813 12,54 6,92 0,552 19,52 9,02 6,38 4,02 7,44 10,52 0,462 0,327 0,206 0,381 0,461 0,642 45,18 4,62 IOZT0323 西藏 modern Lynx lynx 7,65 6,48 0,847 10,33 5,25 6,93 0,508 0,671 16,69 8,24 5,51 3,74 5,84 8,59 0,494 0,33 0,224 0,35 2,17 4,97 2,29 0,458 0,619 0,298 132,44 121,22 41,42 86,85 52,6 32,99 6,07 IOZT0323 西藏 modern Lynx lynx 7,89 6,45 0,817 10,6 5,6 7,27 0,528 0,686 16,64 8,38 5,6 3,96 5,69 9,11 0,504 0,337 0,238 0,342 2,12 4,73 2,231 0,474 0,637 0,284 41,36 6,07 USNM152619 Spain, Huelvmodern Lynx pardinus 7,47 6,27 15,79 0,839 2,114 10,52 5,08 8,17 0,483 0,777 16,1 7,56 5,97 2,95 6,07 9,25 0,47 0,371 0,183 0,377 2,64 4,85 1,837 0,464 0,653 0,301 39,78 54,14 36,09 5,96 USNM152619 Spain, Huelvmodern Lynx pardinus 7,39 6,37 17,67 0,862 2,391 9,92 5,26 8,53 0,53 0,86 15,98 7,26 5,78 3,39 5,67 9,26 0,454 0,362 0,212 0,355 2,76 4,44 1,609 0,462 0,621 0,278 38,78 5,18 USNM152618 Spain, Huelvmodern Lynx pardinus 7,39 5,99 17 0,811 2,3 10,18 4,92 8,49 0,483 0,834 14,49 6,96 5,49 2,77 5 8,67 0,48 0,379 0,191 0,345 0,51 0,703 36,39 52,36 32,97 4,95 USNM152618 Spain, Huelvmodern Lynx pardinus 7,25 5,66 16,98 0,781 2,342 10,06 5,03 8,66 0,5 0,861 14,36 6,95 5,37 3,05 4,83 9,43 0,484 0,374 0,212 0,336 2,31 4,33 1,874 0,505 0,701 0,302 120,11 36,28 4,69 AMNH M169492 Spain, Toledmodern Lynx pardinus 7,26 5,9 0,813 10,11 5,01 0,496 15,47 7,34 5,6 2,9 5,78 8,27 0,474 0,362 0,187 0,374 2,65 4,26 1,608 0,469 0,654 0,275 131,62 118,86 38,36 93,01 52,66 35,13 5,83 AMNH M169492 Spain, Toledmodern Lynx pardinus 10,09 4,97 0,493 15,34 7,53 5,7 3,1 5,58 0,491 0,372 0,202 0,364 0,658 38,22 5,74 USNM81910 Washington modern Lynx canadensis 7,4 5,92 17,89 0,8 2,418 10,8 4,96 8,36 0,459 0,774 16,49 7,62 5,6 4,08 6,32 8,95 0,462 0,34 0,247 0,383 2,67 4,04 1,513 0,449 0,655 0,245 127,79 118,91 41,09 89,96 54,18 35,39 6,72 USNM81910 Washington modern Lynx canadensis 7,69 5,96 18,88 0,775 2,455 10,15 4,8 8,1 0,473 0,798 16,46 7,66 5,22 4,31 5,95 9,26 0,465 0,317 0,262 0,361 2,8 4,52 1,614 0,467 0,617 0,275 40,41 6,76 USNM77095 Washington modern Lynx canadensis 6,58 4,93 17,13 0,749 2,603 9,4 4,61 7,86 0,49 0,836 15,1 7,07 5,19 3,57 5,79 8,5 0,468 0,344 0,236 0,383 2,24 4,19 1,871 0,436 0,623 0,277 123,82 113,61 37,62 87,22 52,11 32,94 6,83 USNM77095 Washington modern Lynx canadensis 6,41 5,09 15,94 0,794 2,487 9,53 4,45 7,7 0,467 0,808 15,4 6,89 4,97 3,87 5,73 8,69 0,447 0,323 0,251 0,372 2,28 4,45 1,952 0,416 0,619 0,289 37,8 6,75 USNM212611 Yukon modern Lynx canadensis 8,39 6,22 20,77 0,741 2,476 10,99 5,38 9,26 0,49 0,843 17,17 7,89 5,77 3,64 6,47 9,71 0,46 0,336 0,212 0,377 2,84 5,21 1,835 0,489 0,64 0,303 134,28 121,83 41,97 95,06 57,77 36,73 6,03 USNM212611 Yukon modern Lynx canadensis 8,26 6,21 21,3 0,752 2,579 10,78 5,47 9,1 0,507 0,844 16,94 7,74 5,74 3,92 6 10,05 0,457 0,339 0,231 0,354 3,09 5,24 1,696 0,488 0,636 0,309 42,02 6,86 USNM212620 Yukon modern Lynx canadensis 7,06 5,67 0,803 10,71 4,95 8,62 0,462 0,805 16,8 8,13 5,29 3,88 6,61 9,18 0,484 0,315 0,231 0,393 2,86 5,12 1,79 0,42 0,638 0,305 130,5 119,52 39,86 90,15 53,16 35,56 6,55 USNM212620 Yukon modern Lynx canadensis 7,08 5,59 18,51 0,79 2,614 10,92 5,11 8,84 0,468 0,81 16,72 7,93 5,29 4,05 6,36 9,54 0,474 0,316 0,242 0,38 3,19 5,36 1,68 0,423 0,653 0,321 39,82 6,51 USNM212617 Yukon modern Lynx canadensis 7,45 5,61 18,06 0,753 2,424 10,7 5,1 8,42 0,477 0,787 16,8 7,83 5,57 3,99 6,49 8,92 0,466 0,332 0,238 0,386 3,07 4,82 1,57 0,443 0,637 0,287 127,68 121,02 40,01 90,65 54,05 33,54 5,61 USNM212617 Yukon modern Lynx canadensis 7,26 5,62 18,18 0,774 2,504 10,77 5,16 8,82 0,479 0,819 16,71 7,97 5,69 4,11 6,24 10,06 0,477 0,341 0,246 0,373 2,91 5,02 1,725 0,434 0,645 0,3 40,41 5,61 USNM33126 Minnesota modern Lynx canadensis 7,69 5,98 0,778 10,27 5,25 9,26 0,511 0,902 16,58 7,59 5,77 3,3 6,61 9,71 0,458 0,348 0,199 0,399 2,7 4,24 1,57 0,464 0,619 0,256 135,05 124 40,72 96,85 55,62 36,49 6,36 USNM33126 Minnesota modern Lynx canadensis 7,4 5,98 0,808 10,58 5,43 9,51 0,513 0,899 16,49 7,63 5,84 3,66 6,19 9,97 0,463 0,354 0,222 0,375 2,63 4,57 1,738 0,449 0,642 0,277 40,63 5,89 AMNH11802 Ark. 15 Mi smodern Lynx canadensis 8,42 5 7,02 0,594 0,834 13,97 6,15 4,45 3,22 5,14 7,28 0,44 0,319 0,23 0,368 0,603 USNM224357 Oregon modern Lynx rufus USNM224357 Oregon modern Lynx rufus USNM224359 Oregon modern Lynx rufus USNM224359 Oregon modern Lynx rufus USNM224342 Oregon modern Lynx rufus USNM224342 Oregon modern Lynx rufus USNM225357 Oregon modern Lynx rufus USNM225357 Oregon modern Lynx rufus AMNH M14822 Texas modern Lynx rufus 6,92 5,82 16,29 0,841 2,354 9,53 5,26 7,78 0,552 0,816 15,29 6,82 6,01 2,86 5,96 7,97 0,446 0,393 0,187 0,39 3,73 6,01 1,611 0,453 0,623 0,393 124,25 113,12 36,62 81,97 49,52 33,57 5,04 AMNH M14822 Texas modern Lynx rufus 6,6 5,89 16,21 0,892 2,456 9,65 5,07 8,07 0,525 0,836 15,12 7,15 6,03 2,73 5,63 8,27 0,473 0,399 0,181 0,372 3,14 5,63 1,793 0,437 0,638 0,372 36,87 5,29 AMNH M14823 Texas modern Lynx rufus 6,92 5,24 0,757 9,04 4,53 7,45 0,501 0,824 13,59 6,03 5,09 2,4 4,98 7,86 0,444 0,375 0,177 0,366 2,37 3,84 1,62 0,509 0,665 0,283 119,02 109,42 35,29 80,61 47,13 29,97 6,29 AMNH M14823 Texas modern Lynx rufus 6,63 5,46 15,52 0,824 2,341 9,14 4,57 7,51 0,5 0,822 13,97 6,2 5,28 2,68 4,68 8,3 0,444 0,378 0,192 0,335 2,24 4,26 1,902 0,475 0,654 0,305 35,21 5,47 AMNH M243108 Florida modern Lynx rufus 7,51 6,6 17,64 0,879 2,349 9,42 5,16 7,87 0,548 0,835 14,4 7,28 5,87 2,36 5,36 8,34 0,506 0,408 0,164 0,372 2,47 4,83 1,955 0,522 0,654 0,335 131,11 120,07 37,62 89,99 52,71 34,97 6,45 AMNH M243108 Florida modern Lynx rufus 7,27 6,54 18,66 0,9 2,567 9,36 5,08 7,71 0,543 0,824 14,51 7,08 5,86 2,57 5,17 8,51 0,488 0,404 0,177 0,356 2,52 4,86 1,929 0,501 0,645 0,335 37,62 6,43 AMNH M24666 Mexico， Simodern Lynx rufus 6,04 4,54 14,28 0,752 2,364 8,53 4,5 6,68 0,528 0,783 13,07 5,93 4,78 2,29 4,78 7,51 0,454 0,366 0,175 0,366 2,15 3,86 1,795 0,462 0,653 0,295 115,88 105,03 33,86 77,18 46,39 28,47 5,83 AMNH M24666 Mexico， Si modern Lynx rufus 5,98 4,45 14,52 0,744 2,428 8,55 4,63 6,97 0,542 0,815 12,68 5,72 4,86 2,52 4,36 7,59 0,451 0,383 0,199 0,344 2,24 4,13 1,844 0,472 0,674 0,326 33,68 5,93 AMNH M24665 Mexico， Simodern Lynx rufus 6,28 5,06 0,806 9,2 4,53 6,98 0,492 0,759 13,89 6,6 5,2 2,3 5,03 8,12 0,475 0,374 0,166 0,362 2,35 4,58 1,949 0,452 0,662 0,33 133,19 118,04 37,89 87,44 49,26 31,72 8,27 AMNH M24665 Mexico， Si modern Lynx rufus 6,08 5,06 15,19 0,832 2,498 9,22 4,67 7,43 0,507 0,806 13,8 6,38 5,15 2,47 4,79 0,462 0,373 0,179 0,347 2,31 4,83 2,091 0,441 0,668 0,35 37,81 8,46 IOZ23613 云南永德 modern Catopuma temminckii 9,14 6,52 18,5 0,713 2,024 10,22 6 0,587 16,82 7,58 5,88 3,28 6,02 9,12 0,451 0,35 0,195 0,358 2,72 5,18 1,904 0,543 0,608 0,308 144,88 129,66 43,72 94,04 53,72 34,28 9,02 IOZ23613 云南永德 modern Catopuma temminckii 9,02 6,2 0,687 10,42 6,12 7,72 0,587 0,741 16,7 7,42 5,82 3,18 6,18 9 0,444 0,349 0,19 0,37 2,68 5,2 1,94 0,54 0,624 0,311 43,74 8,32 IOZ25947 湖北房县 modern Catopuma temminckii 8,64 6,25 17,18 0,723 1,988 10,5 5,33 6,89 0,508 0,656 17,48 8,48 5,93 3,14 6,42 8,62 0,485 0,339 0,18 0,367 2,42 4,73 1,955 0,494 0,601 0,271 138,12 42,57 86,7 54,22 33,09 5,92 IOZ25947 湖北房县 modern Catopuma temminckii 8,62 6,27 16,94 0,727 1,965 10,86 5,46 7,62 0,503 0,702 17,72 8,44 5,88 3,69 6,27 8,72 0,476 0,332 0,208 0,354 2,36 4,49 1,903 0,486 0,613 0,253 42,24 5,57 IOZ29533 湖北房县 modern Catopuma temminckii 6,89 5,34 16,46 0,775 2,389 9,52 5,28 7,62 0,555 0,8 16,52 7,52 5,79 2,85 5,85 8,25 0,455 0,35 0,173 0,354 2,25 4,45 1,978 0,417 0,576 0,269 127,6 115,84 38,42 81,51 48,92 30,35 6,63 IOZ29533 湖北房县 modern Catopuma temminckii 7,01 5,34 16,46 0,762 2,348 9,12 5,28 6,97 0,579 0,764 16,34 7,46 5,63 3,02 5,88 8,47 0,457 0,345 0,185 0,36 2,42 4,89 2,021 0,429 0,558 0,299 39,13 7,2 IOZ 无号 modern Catopuma temminckii 18,14 8,72 0,481 IOZ 无号 modern Catopuma temminckii 18,56 8,52 0,459 IOZ25038 四川平武 modern Catopuma temminckii 7,18 5,62 14,78 0,783 2,058 10,24 5,12 0,5 16,72 7,52 5,62 3,28 6,12 0,45 0,336 0,196 0,366 0,429 0,612 130,86 117,04 41,18 84,48 51,74 30,78 7,68 IOZ25038 四川平武 modern Catopuma temminckii 7,32 5,52 15,24 0,754 2,082 10,48 5,18 0,494 16,46 7,72 5,52 3,24 6 0,469 0,335 0,197 0,365 0,445 0,637 40,54 7,56 IOZ23601 四川平武 modern Catopuma temminckii 7,29 5,73 16,49 0,786 2,262 10,29 5,01 0,487 15,75 7,73 5,38 2,71 5,72 0,491 0,342 0,172 0,363 0,463 0,653 131,4 118,69 40,59 85,35 52,07 30,29 7,45 IOZ23601 四川平武 modern Catopuma temminckii 7,26 5,91 17,17 0,814 2,365 10,43 5,19 0,498 15,79 7,72 5,64 2,69 5,55 0,489 0,357 0,17 0,351 2,27 4,93 2,172 0,46 0,661 0,312 40,03 7,32 USNM259012 四川汶川 modern Catopuma temminckii 8,88 6,58 0,741 11,17 6,24 7,64 0,559 0,684 17,76 8,41 5,82 4,3 5,99 0,474 0,328 0,242 0,337 2,33 5,16 2,215 0,5 0,629 0,291 154,31 140,76 45,21 102,09 56,28 36,83 8,63 USNM259012 四川汶川 modern Catopuma temminckii 8,68 6,61 0,762 10,92 6,17 7,57 0,565 0,693 18,1 8,55 5,91 4,1 6,33 0,472 0,327 0,227 0,35 2,24 4,54 2,027 0,48 0,603 0,251 44,61 8,07 USNM182310 Kenya, Ulu modern Caracal caracal 8,16 5,95 14,68 0,729 1,799 9,81 4,94 7,45 0,504 0,759 16,15 7,9 5,05 3,18 5,91 8,68 0,489 0,313 0,197 0,366 2,81 4,67 1,662 0,505 0,607 0,289 131,98 119,5 39,52 87,65 49,58 31,04 5,49 USNM182310 Kenya, Ulu modern Caracal caracal 7,65 5,75 13,95 0,752 1,824 10,16 4,94 7,43 0,486 0,731 16,05 7,8 5,09 3,14 5,61 9,1 0,486 0,317 0,196 0,35 2,85 4,74 1,663 0,477 0,633 0,295 39,3 6,01 USNM182218 Kenya, Ulu modern Caracal caracal 7 4,91 0,701 8,63 15,04 6,73 4,87 2,51 5,41 8,34 0,447 0,324 0,167 0,36 2,39 4,1 1,715 0,465 0,574 0,273 117,18 106,63 36,67 82,13 48,57 30,83 5,57 USNM182218 Kenya, Ulu modern Caracal caracal 7,17 5,15 13,63 0,718 1,901 8,71 15,36 6,72 4,94 2,64 5,55 8,26 0,438 0,322 0,172 0,361 0,467 0,567 36,59 5,76 USNM368503 Botswana modern Caracal caracal 8,86 7,07 0,798 11,69 17,61 5,11 0,503 0,664 0,29 150,88 136,46 45,04 109,73 58,96 41,68 6,44 USNM368503 Botswana modern Caracal caracal 11,88 17,76 5,44 0,669 0,306 45,21 6,53 USNM40164 Botswana modern Caracal caracal 7,66 5,13 14,83 0,67 1,936 9,02 4,73 7,76 0,524 0,86 16,23 6,77 4,82 2,94 6,32 8,86 0,417 0,297 0,181 0,389 2,57 4,18 1,626 0,472 0,556 0,258 120,97 110,36 37,37 80,53 48,17 28,44 4,97 USNM40164 Botswana modern Caracal caracal 7,48 5,14 14,43 0,687 1,929 9,07 4,88 7,96 0,538 0,878 15,88 6,95 4,71 2,89 5,99 9,04 0,438 0,297 0,182 0,377 2,41 4,7 1,95 0,471 0,571 0,296 37,36 4,94 USNM384162 Botswana modern Caracal caracal USNM384162 Botswana modern Caracal caracal Appendix for A small Lynx from Longdan, Gansu Province and the diversification of Lynx lineage during the Plio-Pleistocene Fig. S1. Maximum parsimony analysis, without constraint. Fig. S2. Maximum parsimony analysis, with constraint on Panthera with Neofelis, Acinonyx with Puma, and Otocolobus with Prionailurus. Fig. S3. Bayes Inference analysis, without constraint. Fig. S4. Bayes Inference analysis, with constraint on Acinonyx with Puma, and Otocolobus with Prionailurus. Fig. S5. Bayes Inference tip-dating analysis, without constraint. Fig. S6. Bayes Inference tip-dating analysis, with constraint on Otocolobus with Prionailurus.