Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

The Taxonomic and Phylogenetic Affinities of Bunopithecus sericus, a Fossil Hylobatid from the Pleistocene of China

  • Alejandra Ortiz ,

    Affiliations Center for the Study of Human Origins, Department of Anthropology, New York University, New York, New York, United States of America, New York Consortium in Evolutionary Primatology (NYCEP), New York, New York, United States of America

  • Varsha Pilbrow,

    Affiliation Department of Anatomy and Neuroscience, University of Melbourne, Melbourne, Victoria, Australia

  • Catalina I. Villamil,

    Affiliations Center for the Study of Human Origins, Department of Anthropology, New York University, New York, New York, United States of America, New York Consortium in Evolutionary Primatology (NYCEP), New York, New York, United States of America

  • Jessica G. Korsgaard,

    Affiliation Center for the Study of Human Origins, Department of Anthropology, New York University, New York, New York, United States of America

  • Shara E. Bailey,

    Affiliations Center for the Study of Human Origins, Department of Anthropology, New York University, New York, New York, United States of America, New York Consortium in Evolutionary Primatology (NYCEP), New York, New York, United States of America

  • Terry Harrison

    Affiliations Center for the Study of Human Origins, Department of Anthropology, New York University, New York, New York, United States of America, New York Consortium in Evolutionary Primatology (NYCEP), New York, New York, United States of America

The Taxonomic and Phylogenetic Affinities of Bunopithecus sericus, a Fossil Hylobatid from the Pleistocene of China

  • Alejandra Ortiz, 
  • Varsha Pilbrow, 
  • Catalina I. Villamil, 
  • Jessica G. Korsgaard, 
  • Shara E. Bailey, 
  • Terry Harrison


Fossil hylobatids are rare, but are known from late Miocene and Pleistocene sites throughout East Asia. The best-known fossil hylobatid from the Pleistocene of China is a left mandibular fragment with M2-3 (AMNH 18534), recovered from a pit deposit near the village of Yanjinggou in Wanzhou District, Chongqing Province. Matthew and Granger described this specimen in 1923 as a new genus and species, Bunopithecus sericus. Establishing the age of Bunopithecus has proved difficult because the Yanjinggou collection represents a mixed fauna of different ages, but it likely comes from early or middle Pleistocene deposits. Although the Bunopithecus specimen has featured prominently in discussions of hylobatid evolution and nomenclature, its systematic status has never been satisfactorily resolved. The present study reexamines the taxonomic and phylogenetic relationships of Bunopithecus by carrying out a detailed comparative morphometric study of its lower molars in relation to a large sample of modern hylobatids. Our results show that differences in M2 and M3 discriminate extant hylobatids fairly well, at least at the generic level, and that AMNH 18534 is not attributable to Hylobates, Nomascus or Symphalangus. Support for a close relationship between Bunopithecus and Hoolock is more equivocal. In most multivariate analyses, Bunopithecus presents a unique morphological pattern that falls outside the range of variation of any hylobatid taxon, although its distance from the cluster represented by extant hoolocks is relatively small. Our results support the generic distinction of Bunopithecus, which most likely represents an extinct crown hylobatid, and one that may possibly represent the sister taxon to Hoolock.


The fossil record documenting the evolutionary history of hylobatids (i.e., gibbons and siamangs) is extremely meager; the only definitive representatives of the family are known from localities in Asia dating to the late Miocene and Pleistocene. Yuanmoupithecus from the late Miocene (~7–8 Ma) of Yunnan, China [1] is considered a stem hylobatid based on the presence of a suite of dental specializations that are shared uniquely with extant members of the clade [23]. Pleistocene hylobatids are known from China, Vietnam, Thailand, Laos, Borneo, Sumatra and Java [326]. These are mostly isolated teeth, making their taxonomic and phylogenetic affinities difficult to ascertain, but they probably represent crown hylobatids and are likely attributable to extant genera [3, 22]. The earliest records of fossil hylobatids from the Quaternary of Asia come from cave sites in Guangxi in southern China with an estimated age of 2.2 Ma [2627]. From the early Pleistocene onwards, and even into historic times, gibbons were widely distributed across southern China [67, 22, 2830], whereas today they are restricted to Yunnan, Guangxi and Tibet in southwestern China and to the island of Hainan [3134]. The best-known fossil hylobatid from the Pleistocene of China is a partial mandible from Yanjinggou in Chongqing Province (formerly part of Sichuan Province). The specimen was discovered by Walter Granger in 1920–1921 and was later described by Matthew and Granger [35] as a new genus and species, Bunopithecus sericus. Even though its phylogenetic affinities have never been satisfactorily resolved, Bunopithecus has featured prominently in discussions of hylobatid evolution and nomenclature.

The Bunopithecus sericus type specimen (AMNH 18534) consists of a left mandibular fragment with M2-3 (Fig 1). The mandibular corpus is complete below M2 and M3 and the inferior margin extends anteriorly below the roots of P4 and M1. Posterior to M3, the anterior portion of the ramus is also preserved. The two molars are well-preserved and only lightly worn. The specimen was recovered from the pits and fissures near the village of Yanjinggou in Wanzhou District, Chongqing (Yen-ching-kao, Wan-hsien, Szechuan of [35]). Although Yanjinggou is one of the most famous and productive Quaternary fossil vertebrate sites in China, establishing the age of the fauna has proved difficult. Matthew and Granger [35] provisionally attributed the Yanjinggou fauna to the late Pliocene, but subsequent studies indicated an early to late Pleistocene age [3639]. The major impediment to ascertaining the age of Granger’s collection is that it probably represents a mixed fauna of different ages [40]. Recent renewed field work at Yanjinggou has helped to clarify the ages of the faunas, and it appears most likely that the Bunopithecus specimen came from early or middle Pleistocene deposits [40].

Fig 1. Photograph of the Bunopithecus sericus specimen (AMNH 18534) represented by a left mandibular fragment with M2-3.

A) Lateral (buccal) view. B) Medial (lingual) view. C) Occlusal view (lingual to the right).

Matthew and Granger [35] noted similarities between the Yanjinggou specimen and extant gibbons, but opted to name a new genus and species for the fossil due to several distinctive features of the molars (i.e., greater crown width and larger hypoconulid size). Colbert and Hooijer [38], using a larger comparative sample, argued that the purported distinctive dental features of B. sericus are found among extant gibbons. They concluded that the generic distinction was unwarranted, and preferred to recognize the fossil as Hylobates (Bunopithecus) sericus. Subsequently, Frisch [41] and Groves [29] inferred that B. sericus was most closely related to hoolock gibbons. With the recognition that hoolock gibbons should be considered a distinct subgenus/genus [42], the apparent close relationship between Bunopithecus sericus and extant hoolock gibbons became central to deciding on an available genus-group name [43]. If the fossil species is included in the same genus, then Bunopithecus Matthew and Granger, 1923 becomes the oldest available name for hoolock gibbons (and one of the rare examples of a fossil type species for an extant genus). For the next two decades the subgenus/genus name Bunopithecus was widely employed as the valid name for the hoolock gibbons. However, further investigations cast doubt on the dental similarities between extant hoolocks and Bunopithecus, and led to a rethinking of their taxonomic association. Gu [7] suggested, without offering much in the way of supporting morphological evidence, that B. sericus most closely resembles Nomascus concolor. In their review of gibbon evolution, Jablonski and Chaplin [22] followed Gu [7] in referring Bunopithecus sericus to Nomascus (even though Bunopithecus Walter and Granger, 1923 has priority over Nomascus Miller, 1933). Most recently, Mootnick and Groves [44] pointed out that the dental characters of B. sericus fall outside the range of variation of extant gibbons and hypothesized that Bunopithecus probably represents an extinct genus, without a close evolutionary relationship to hoolock gibbons. In light of this, Mootnick and Groves [44] proposed a new genus name, Hoolock, for the extant hoolock gibbons, and excluded Bunopithecus from synonymy.

Currently, the affinities of Bunopithecus sericus remain unsettled. The main problems hampering previous interpretations of its taxonomic status are that the fossil is represented by a single fragmentary type specimen, that only small samples of modern gibbons have been used in comparisons, that the cheek teeth of extant hylobatids are notoriously difficult to discriminate (except by differences in size), and that most assessments have been based on just a few linear measurements and qualitative traits [35, 38, 41, 43, 4546]. The present study aims to clarify the taxonomic and phylogenetic relationships of Bunopithecus by carrying out a detailed comparative morphometric study of its lower molars in relation to a large sample of modern hylobatids.

Extant hylobatids are currently included in four genera–Hylobates, Hoolock, Symphalangus and Nomascus. The phylogenetic relationships between the extant genera have not been adequately resolved [4761], and the lack of consensus probably stems from the rapid radiation of the crown taxa and the confounding influences of species introgression [58, 61]. Nevertheless, the majority of studies provide support for the following set of relationships: (Nomascus (Symphalangus (Hoolock, Hylobates))) [4756, 60]. Molecular clock estimates indicate a date of ~19 Ma (with a range of ~16.3–21.8 Ma) for the divergence of the hylobatids from the other hominoids and ~8.4 Ma (with a range of 10.5–5.2 Ma) for the divergence of the extant genera [3, 4756, 60, 6265]. Given this inferred chronology, Bunopithecus postdates the differentiation of the extant generic lineages, which increases the likelihood that the fossil belongs to a member of the crown clade, but it does not entirely rule out the possibility that it represents a late-surviving stem hylobatid. The enduring question then is whether Bunopithecus is distinct enough to be retained as a valid genus (either as a stem hylobatid or as a distinctive crown clade) or whether it should be subsumed into one of the four currently recognized genera of extant hylobatids (and if so, which one?). If Bunopithecus is deemed to belong to one of the extant hylobatid genera, then priority of the name Bunopithecus over both Hoolock and Nomascus becomes a critical nomenclatural issue that will need to be considered. In addition to clarifying the taxonomic status of Bunopithecus, the results of our study have implications for understanding hylobatid biogeography and evolutionary relationships.

Materials and Methods


The present study examines the M2 and M3 of the B. sericus fossil specimen and makes comparisons with those of extant hylobatids. Access to the original B. sericus type specimen (AMNH 18534) was given to AO by the Division of Vertebrate Paleontology at the American Museum of Natural History (AMNH), New York, United States of America. Permission to study the fossil was notified by Ms. Judith Galkin from the Division of Vertebrate Paleontology, AMNH via email on 05/15/13. A total of 289 molar teeth represented by 172 extant individuals from Hylobates, Hoolock, Nomascus and Symphalangus were included in the comparative sample (Table 1). It should be noted that the sample size per molar type varies due to missing teeth and differential preservation and wear. Metrical data on extant hylobatids were collected from the following museums: AMNH; National Museum of Natural History (NMNH), Washington D.C.; Field Museum of Natural History (FMNH), Chicago; Museum of Comparative Zoology (MCZ), Cambridge; Natural History Museum (BMNH), London; Zoologisches Museum (ZMB), Berlin; Anthropologische und Zoologische Staassammlung (ZSM), Munich; Anthropologisches Institüt und Museum der Universität Zurich-Irchel (AS/Z), Zurich; and Muséum National d’Histoire Naturelle (MNHN), Paris. Only data from individuals of known provenance were collected. Provenance information was obtained from museum records and the nomenclature was updated to reflect the currently accepted taxonomy [32, 34]. Because molar size is not sexually dimorphic in extant hylobatids [46], sex was not included as a variable in this study. We attempted, however, to maintain an equal representation of male and female individuals. Information on dental wear was collected using Pilbrow’s [66] three-stage system and specimens with heavily worn teeth were excluded from analyses.

Table 1. Fossil and recent comparative sample of hylobatid lower molars used in this study.

Data Collection Procedures

High-resolution images of the occlusal surface of the M2 and M3 of B. sericus and those of the comparative samples were taken with either a Canon Digital Rebel XT camera with 18–55mm lens (AO) or a Minolta X-700 camera outfitted with a Sigma 50mm F2.8 macro lens (VP). As described elsewhere [6667], teeth were positioned so that the cervical border was perpendicular to the optical axis of the camera. A millimeter scale was included in each image, placed at the same horizontal plane as the cusp apices, and both the scale and the camera were leveled using standard bubble devices. The smallest aperture possible was used to maximize depth of field. Because only the left mandibular corpus is preserved in B. sericus, comparative data were collected on left molars when available. In the case of missing or damaged teeth, the right antimere was used and then digitally mirror-imaged to correspond to the left side. Data were collected separately for each molar type.

Digital images were imported into SigmaScan Pro (Systat Software Inc.) imaging software in order to collect mesiodistal and buccolingual dimensions, the angle between cusps (i.e., cusp position), and cusp and crown base areas. All measurements used in our analysis have proven useful for differentiating between fossil and living hominins [6871] and great apes [66, 7274]. Importantly, previous studies have shown that intra and inter-observer errors in tooth crown orientation and measurement are relatively low (0.5–3.0%) and not statistically significant [66, 75].

As defined by Pilbrow [66], the mesiodistal length (MDLENGTH) was measured along the longitudinal developmental groove (Fig 2). Buccolingual breadths were taken at the tips of the mesial (BLMES) and distal (BLDIS) cusps. The positions of the protoconid (ANBCUSP) and metaconid (ANLCUSP) were collected as the angles formed between the lines connecting the tips of these cusps to the buccal and lingual sides of the longitudinal groove, respectively. Similarly, the position of the hypoconulid (ANHYCLD) was calculated as the angle formed by the apices of the entoconid, hypoconid and hypoconulid. Individual areas for the protoconid (ABSAPROTO), metaconid (ABSAMETA), entoconid (ABSAENTO), hypoconid (ABSAHYPCD) and hypoconulid (ABSAHYPCLD) were measured by tracing the outline of the crown and the main fissures dividing the cusps. Wood and Engleman’s [76] protocol was followed when minor corrections for interproximal wear were necessary, as well as for estimating the individual areas of the five main cusps in cases where additional cusps (e.g., C6, C7) or marginal tubercles were present. The total crown area (OCCLAREA) was calculated by summing the individual cusp areas. Similarly, the sum of the areas of the protoconid and metaconid and the areas of the hypoconid, hypoconulid and entoconid gave the total area of the trigonid (ABSATRIGD) and talonid (ABSATALD), respectively. Relative cusp areas (RELAPROTO, RELAMETA, RELAENTO, RELAHYPCD and RELAHYPCLD) were determined by dividing the individual cusp areas by the total crown area. The same protocol was used for calculating the relative trigonid (RELATRIGD) and talonid (RELATALD) areas. A total of 21 variables were analyzed for each molar.

Fig 2. Hylobatid left lower molar showing dental variables taken.

A) Linear dimensions: 1) mesiodistal length, 2) buccolingual width at mesial cusps, and 3) buccolingual width at distal cusps. B) Angles: 1) position of mesiobuccal cusp, 2) position of mesiolingual cusp, and 3) position of hypoconulid. C) Absolute cusp areas: 1) protoconid, 2) metaconid, 3) entoconid, 4) hypoconid, and 5) hypoconulid.

Statistical Analyses

Univariate and multivariate statistical analyses were performed on PAST [77]. Unless otherwise noted, differences were considered significant at α = 0.05. Because sample sizes differ between taxa and the data are non-parametric, Kruskal-Wallis tests were performed for each variable to determine whether sample medians were significantly different from each other. Mann-Whitney U tests were performed on each variable to determine where such differences lay, with p-values subjected to a Bonferroni correction. Z-tests were also performed on the logged data to determine whether B. sericus, for which a single data point is available for each variable, can be excluded from each extant group. To determine group differences and the identity of B. sericus using all available data, discriminant function analyses (DFAs) were performed using both absolute values that contain information about size and relative values that contain information primarily on shape. Linear measurements BLMES and BLDIS were standardized relative to MDlength, whereas measurements of cusp areas were standardized as a proportion of the total occlusal area. Either absolute or relative cuspal occlusal area values were used, with the total occlusal area itself excluded from all DFAs to avoid excessive weighting of tooth area in the results. Angle measurements were converted to radians for all DFAs to normalize their distribution and reduce their range of variation. DFAs were performed for M2 and M3 separately and combined. Accuracy of the DFAs was quantified by determining the percent of individuals correctly classified, with and without jackknifing. Individuals with missing data were not included in analyses.


Descriptive statistics for the 21 variables of M2 and M3 for both B. sericus and extant hylobatids are presented in S1 and S2 Tables, respectively. Results of the Kruskal-Wallis statistical analysis are presented in Table 2 and details of the pair-wise Mann-Whitney U tests are provided in S3 Table. Our results show that size has a greater discriminatory power than shape in both M2 and M3, with all linear dimensions and absolute areas showing significant differences among sample medians (Table 2). With few exceptions (e.g., the position and relative size of the hypoconulid), shape variables such as cusp angles and relative areas do not appear useful for differentiating among hylobatid taxa. Previous studies have argued that due to the existence of only minor differences in the dental morphology between genera and the high degree of intra-specific variability, dental characters are not particularly informative for hylobatid systematics [22, 41, 4546]. However, the results of the Mann-Whitney U tests performed on the samples included in this study reveal the potential of several dental variables to differentiate among groups, at least at the generic level. For M2, the following variables are significantly different between sample medians in all four extant genera: MDlength, ABSAHYPCD, ABSAHYPCLD (except for Hoolock vs. Nomascus), ABSATRIGD, ABSATALD and OCCLAREA (S3 Table). Similarly, significant differences in all pair-wise comparisons were found for each of the following M3 variables: MDlength (except for Hylobates vs. Nomascus), ABSAHYPCD (except for Hoolock vs. Nomascus), ABSATRIGD, ABSATALD (except for Hylobates vs. Nomascus) and OCCLAREA (S3 Table). In general, among extant hylobatids, Symphalangus shows the largest mean values for each of the absolute cusp areas, and concomitantly the largest mean values for the trigonid, talonid and total occlusal base. Molar size is second largest in Hoolock, followed by Nomascus and finally Hylobates, which shows the smallest absolute size with regards to the variables examined (S1 and S2 Tables).

S1 and S2 Tables also provide the values for each variable for the B. sericus specimen. Comparisons with the samples of extant genera show that most absolute values for B. sericus approximate the mean values of Hoolock more closely than those of other taxa, although there is considerable overlap. Confidence intervals at the 95% confidence level (± 2SD) suggest that the M2 of B. sericus and Symphalangus are significantly different in MDlength and ABSAHYPCD. Similarly, there is a statistically significant difference between the M2 of B. sericus and Nomascus in BLMES, BLDIS and ANLCUSP, and between B. sericus and Hylobates in ABSATRIGD and OCCLAREA. Data for the M3 at the 95% confidence level of the absolute measurements and cusp position variables indicate that B. sericus significantly differs from Nomascus in BLMES, ANBCUSP and ANLCUSP and from Hylobates in MDlength and ABSAHYPCD. It differs significantly from both Nomascus and Hylobates in ABSAMETA, ABSATRIGD, ABSATALD and OCCLAREA. Finally, the ABSAHYPCLD of B. sericus M3 is significantly different from that of Hoolock and Hylobates.

The results of the Z-tests are summarized in Table 3. As noted above, these tests were conducted on each variable to determine whether B. sericus can be excluded from any of the extant hylobatid genera. Contrary to early claims by Matthew and Granger [35] using a small comparative sample, B. sericus has significantly narrower tooth crowns across the mesial cusps (BLMES) in both M2 and M3 than the four extant hylobatid genera (see also S1 and S2 Tables). In general, with the exception of ANBCUSP, all linear measurements, absolute areas and cusp angles in M2 indicate that B. sericus does not belong to Symphalangus. The exclusion of B. sericus as a member of Symphalangus based on information contained in M3 is, however, slightly less robust. Comparisons with the other three extant groups are also more ambiguous. In addition to BLMES, B. sericus significantly differs from extant gibbons in having a smaller relative hypoconid size (RELAHYPCD) in M2 and, except for the M2 of extant hoolocks, a smaller ANLCUSP in both M2 and M3. It is also excluded from Nomascus by BLDIS in both molars and by ANHYCLD in M2. The fossil significantly differs from Hoolock also in ANHYCLD for M2, and absolute (ABSAENTO) and relative (RELAENTO) entoconid size for M3, and from Hylobates in RELAMETA and RELATALD for M2, and RELAPROTO for M3. When information based on confidence intervals and Z-tests is contrasted for more robust assessments, significant values in agreement between univariate analyses demonstrate that B. sericus can be excluded from Nomascus by BLMES and ANLCUSP in both molars, and by BLDIS in M3. We also found that B. sericus does not align with Symphalangus in MDlength and ABSAHYPCD for M2.

Figs 3 and 4 illustrate the plots of the first two discriminant functions for individual molars showing the relative placement of B. sericus using linear measurements, cusp angles and absolute areas. S4 Table provides the average Mahalanobis distances between hylobatid taxa for each analysis. For M2, the first function is responsible for 84.4% of the variance, while the following two functions explain 10.4% and 4.1%, respectively. Similar values were obtained for M3, as the first three functions account for the 87.1%, 6.8% and 4.1% of the variance, respectively. The DFAs performed on M2 and M3 independently reveal that the greatest degree of overlap among hylobatids occurs between Nomascus and Hylobates. As expected, given their larger dentitions, siamangs are the most distinctive of the hylobatids. As illustrated in Fig 3, the M2 of B. sericus falls within the range of overlap among Nomascus, Hoolock and Hylobates, but when all axes are considered, the average pair-wise Mahalanobis comparisons indicate a closest distance to hoolock gibbons (S4 Table). In contrast, data for M3 for the first two discriminant functions show that B. sericus does not cluster with any extant group, although distances between B. sericus and the cluster represented by hoolock gibbons are relatively small (Fig 4). Although Bunopithecus' average Mahalanobis distance is smaller to Hylobates (1.3104) than to Hoolock (1.3159), the average Mahalanobis distance within Hylobates is much smaller (1.2086) than within Hoolock (1.3828), which places B. sericus well within the possible spread of Hoolock but not within the spread of Hylobates (S4 Table). The likelihood of individuals being accurately classified ranges between 80.15% (not jackknifed) and 71.76% (jackknifed) for M2 and between 79.05% (not jackknifed) and 62.86% (jackknifed) for M3.

Fig 3. Plot of the first two discriminant functions (DF1 and DF2) of the M2 analysis using linear measurements, cusp angles and absolute areas.

Eigenvalues: 4.19 (DF1) and 0.52 (DF2); variance: 84.44% (DF1) and 10.41% (DF2).

Fig 4. Plot of the first two discriminant functions (DF1 and DF2) of the M3 analysis using linear measurements, cusp angles and absolute areas.

Eigenvalues: 40.22 (DF1) and 0.31 (DF2); variance: 87.07% (DF1) and 6.80% (DF2).

When data for M2 and M3 are combined, the results of the DFA using the same set of variables (i.e., linear measurements, cusp angles and absolute areas) are slightly more robust with respect to grouping patterns among extant hylobatids. Separation among clusters and the relative position of B. sericus on the scatter plots of the first two discriminate functions are shown in Fig 5. The first function accounts for 74.32% of the variance, with ABSATRIGD and ABSATALD in both molars contributing the most to this axis. The second and third axes, on the other hand, encompass 11.41% and 9.46% of the variance, respectively. As illustrated in Fig 5, with the exception of the overlap between Nomascus and Hylobates, each genus is distinct. Again, B. sericus falls outside the range of variation observed among extant gibbons. However, the distance between B. sericus and the cluster represented by extant hoolock gibbons is quite small (see also S4 Table). The likelihood of individuals being correctly classified ranges between 90.63% (not jackknifed) and 62.50% (jackknifed), which is lower than for either M2 or M3 alone.

Fig 5. Plot of the first two discriminant functions (DF1 and DF2) in the analysis of size variables, M2 and M3 combined.

Eigenvalues: 4.79 (DF1) and 0.73 (DF2); variance: 74.32% (DF1) and 11.41% (DF2).

Finally, a DFA of M2 and M3 data combined was performed using exclusively shape variables, with absolute size components removed (Fig 6). In this analysis, the first function explains 42.88% of the variance, while the following two account for 27.96% and 19.36%, respectively. In agreement with results based on Kruskal-Wallis and Mann-Whitney U tests, this analysis demonstrates that shape variables alone have less discriminatory power than when size is considered, as shape variables alone result in no discernible differences between groups. Based on shape variables alone, B. sericus falls outside the range of variation of extant hylobatids, but closest to Hoolock (see also S4 Table). The likelihood of the classification of individuals into their correct group is relatively lower (75% not jackknifed and 40.63% jackknifed).

Fig 6. Plot of the first two discriminant functions (DF1 and DF2) in the analysis of shape variables, M2 and M3 combined.

Eigenvalues: 0.95 (DF1) and 0.62 (DF2); variance: 42.88% (DF1) and 27.96% (DF2).


Our results show that differences in M2 and M3 discriminate modern hylobatids fairly well, at least at the generic level, and that these differences have the potential to contribute to the taxonomic identification of fossil members of the clade. This is noteworthy given that most previously recognized distinctions between hylobatid genera relate to soft tissue features (e.g., pelage and hair coloration patterns, ear size, nose shape, and hand and feet color) and karyology, although some skeletal differences in cranial size and shape, anatomy of the baculum, and number of thoracic vertebrae have also been reported [29, 7880]. Informative dental features include MDlength, ABSATRIGD, ABSATALD and OCCLAREA, and to a lesser extent, the absolute area of each lower molar cusp. Since M3 exhibits a greater degree of intra-specific variability [81], M2 has a greater discriminatory power than that of the M3. Moreover, variables that contain size and shape data, rather than those containing information about shape alone, are considerably more useful for identifying taxonomic differences in extant hylobatids. In general, Symphalangus has the largest teeth, followed in decreasing order by Hoolock, then Nomascus and finally Hylobates. As shown by our study, the linear dimensions of B. sericus teeth overlap with those observed among all four extant hylobatid genera, although they fall closest to the mean values for Hoolock. However, it is necessary to be cautious about using overall dental size or size-based dental features in taxonomic assessments since there has been a general trend among many Asian mammals, including primates, to undergo dental size reduction during the course of the Pleistocene [9, 8283]. As a consequence, shared morphological specializations are a much better guide to taxonomic affinities than dental size.

The results of our study show that B. sericus is unlikely to belong to Hylobates, Nomascus or Symphalangus. In fact, most of our univariate and multivariate analyses reveal that B. sericus exhibits the greatest differences with extant Nomascus and Symphalangus, especially when size variables are considered. In addition, as noted by Frisch [41], Nomascus is the only genus with a relatively high frequency (61.7%) of buccal cingula on the lower molars, a feature that is lacking in B. sericus. The presence of a lingual cingulum is also common on Nomascus upper molars and traces are occasionally observed in Hylobates. In contrast, hoolocks appear to have lost the cingulum on both upper and lower molars. Frisch [41] also noted that Symphalangus exhibits the least reduced lower third molars, with Hoolock and Nomascus occupying an intermediate position. Although highly variable, the metaconid tends to be distal to the protoconid and the hypoconulid is placed in the midline (or slightly buccal to the midline) in Nomascus and Hoolock molars, while in Hylobates the metaconid and protoconid are more transversely aligned and the hypoconulid is more lingually positioned [41, 46].

Support for a close relationship between B. sericus and modern hoolocks is more equivocal. Taxon means and confidence intervals for each variable, as well as DFA and Z-test results, suggest that although B. sericus is most similar to Hoolock relative to other hylobatid genera, it remains dentally distinct. In most multivariate analyses, B. sericus presents a unique morphological pattern that falls outside the range of variation of any hylobatid taxon, although its distance from the cluster represented by extant hoolocks is relatively small. Our findings support the conclusions reached by Mootnick and Groves [44] that Bunopithecus should be considered a distinct genus. Given the morphological and metrical evidence, it seems likely that Bunopithecus represents an extinct crown hylobatid, one that may possibly represent the sister taxon to Hoolock. However, since the Hoolock sample used in this study is relatively small (n = 20 individuals), and may not adequately encompass the full extent of the intra-generic variation, we cannot entirely rule out the possibility that B. sericus merely represents an early hoolock gibbon that should be included in the genus Hoolock. At present, given the results of our study, we prefer to recognize Bunopithecus as a separate genus.

This opens up the intriguing possibility that an extinct gibbon taxon, in the form of Bunopithecus, may have occupied parts of China to the north and east of the current geographic distribution of extant gibbons during the Pleistocene-Holocene, and may have even survived into historic times (Fig 7). Evidence from historical records shows that gibbons in China were much more widely distributed in the recent past than they are today, extending as far north as the Yellow River and eastwards as far as Zhejiang Province [7, 28, 30, 80]. Records of hylobatids south of the Xijiang River are almost certainly attributable to Nomascus, and these serve to fill the present-day geographic divide between the disjunct distribution of Nomascus in western China, Vietnam and Laos and isolated populations on Hainan (Fig 7). In addition, teeth of fossil gibbons from cave sites in Guangxi Zhuang Autonomous Region are consistent in size with those of Nomascus and, like their modern counterparts, they retain a high incidence of cingula on the upper and lower molars [7]. These lines of evidence indicate that Nomascus occupied much of southern China from at least the Early Pleistocene onwards. The taxonomic identity of recently extirpated gibbons to the north of the Xijiang River is much harder to establish. Geissmann [80] has suggested that paintings of gibbons that were living in Hunan and Hubei Provinces in central China during the eleventh century are strikingly similar to Hoolock. This could potentially extend the geographic range of the genus eastwards more than 1,200 km beyond its present-day distribution. However, an alternative interpretation is conceivable. The historic records of gibbons from central China south of the Yangtze River may refer to an extinct genus of hylobatid (Fig 7). Since these occurrences are in the same general region as Yanjinggou, it is plausible that the extinct genus was represented during the Pleistocene by Bunopithecus sericus. In this case, the apparent similarities to Hoolock in the historical depictions of gibbons from central China would not be unexpected given that the results of our study indicate that Bunopithecus is likely to be the sister taxon of Hoolock. It may well be that the mandibular fragment from the Pleistocene of Yanjinggou represents an extinct hylobatid genus, Bunopithecus, which was once widely distributed across central and eastern China before becoming extinct in historic times.

Fig 7. Map of East and Southeast Asia showing the historical and present distribution of gibbons (Hoolock, Hylobates and Nomascus).

The black star indicates the location of the village of Yanjinggou (Wanzhou District, Chongqing Province, China), where Bunopithecus sericus was found. Adapted from Gu [7], Gao et al. [30] and Geissmann [80].

Supporting Information

S1 Table. Descriptive statistics for M2 size and shape variables in B. sericus and extant hylobatid genera.


S2 Table. Descriptive statistics for M3 size and shape variables in B. sericus and extant hylobatid genera.


S3 Table. Extant hylobatid variance analysis.

Left: Mann-Whitney U test values; Right: Bonferroni adjusted p-values with significant values bolded.


S4 Table. Average pair-wise Mahalanobis distances for each DFA.



We would like to thank Judith Galkin from the Division of Vertebrate Paleontology at the American Museum of Natural History for providing access to the Bunopithecus sericus type specimen. For providing access to skeletal collections, the following institutions are also gratefully acknowledged: American Museum of Natural History (New York, USA), National Museum of Natural History (Washington D.C., USA), Field Museum of Natural History (Chicago, USA), Museum of Comparative Zoology (Cambridge, USA), Museum of Natural History (London, UK), Zoologisches Museum (Berlin, Germany), Anthropologische und Zoologische Staassammlung (Munich, Germany), Anthropologisches Institüt und Museum der Universität Zurich-Irchel (Zurich, Switzerland), and Musée National d’Histoire Naturelle (Paris, France). Fossil collections were studied at the Institute of Vertebrate Paleontology and Paleoanthropology (Beijing, China). Finally, we would like to thank Laurent Viriot, Colin Groves and an anonymous reviewer for their helpful comments on the original version of this manuscript.

Author Contributions

Conceived and designed the experiments: AO TH. Performed the experiments: AO VP JGK. Analyzed the data: AO CIV. Contributed reagents/materials/analysis tools: VP SEB. Wrote the paper: AO CIV TH.


  1. 1. Pan Y. Primates Linnaeus, 1758. In: Qi G, Dong W, editors. Lufengpithecus hudienensis Site. Beijing: Science Press; 2006. pp. 131–148, 320–322.
  2. 2. Harrison T, Ji X, Zheng L. Renewed investigations at the late Miocene hominoid locality of Leilao, Yunnan, China. Am J Phys Anthropol. 2008;135 (S46): 113.
  3. 3. Harrison T. The fossil record and evolutionary history of hylobatids. In: Reichard UH, Hirohisa H, Barelli C, editors. Evolution of Gibbons and Siamang: Phylogeny, Morphology and Cognition. New York: Springer; In press.
  4. 4. Badoux DM. Fossil mammals from two fissure deposits at Punung (Java). Utrecht: Kemink en Zoon; 1959.
  5. 5. Hooijer DA. Quaternary gibbons from the Malay Archipelago. Zool Verhand Leiden. 1960;46: 1–42.
  6. 6. Gu Y. Preliminary research on the fossil gibbon of Pleistocene China. Acta Anthropol Sinica. 1986;5: 208–219.
  7. 7. Gu Y. Preliminary research on the fossil gibbons of the Chinese Pleistocene and recent. Hum Evol. 1989;4: 509–514.
  8. 8. Ciochon RL, Olsen JW. Paleoanthropological and archaeological research in the Socialist Republic of Vietnam. J Hum Evol. 1986;15: 623–633.
  9. 9. Harrison T. The paleoecological context at Niah Cave Sarawak: Evidence from the primate fauna. Bull Indo—Pacific Prehist Assoc. 1996;14: 90–100.
  10. 10. Harrison T. Vertebrate faunal remains from Madai Cave (MAD 1/28), Sabah, East Malaysia. Bull Indo—Pacific Prehist Assoc. 1998;17: 85–92.
  11. 11. Harrison T. Archaeological and ecological implications of the primate fauna from prehistoric sites in Borneo. Bull Indo—Pacific Prehist Assoc. 2000;20: 133–146.
  12. 12. Nisbett RA, Ciochon RL. Primates in northern Viet Nam: A review of the ecology and conservation status of extant species, with notes on Pleistocene localities. Int J Primatol. 1993;14: 765–795.
  13. 13. Schwartz JH, Long VT, Cuong NL, Kha LT, Tattersall I. A diverse hominoid fauna from the late Middle Pleistocene breccia cave of Tham Khuyen, Socialist Republic of Vietnam. Anthropol Papers Am Mus Nat Hist. 1994;73: 1–11.
  14. 14. Wu X, Poirier FE. Important fossil hominoids in China. In: Wu X, Poirier FE, editors. Human evolution in China: A metric description of the fossils and a review of the sites. New York: Oxford University Press; 1995. pp. 241–283. pmid:9140545
  15. 15. Gu Y, Huang W, Chen D, Guo X, Jablonski NG. Pleistocene fossil primates from Luoding, Guangdong. Vert PalAs. 1996;34: 235–250.
  16. 16. Ciochon R, Long VT, Larick R, Gonzales L, Grün R, de Vos J, et al. Dated co-occurrence of Homo erectus and Gigantopithecus from Tham Khuyen Cave, Vietnam. Proc Nat Acad Sci. 1996;93: 3016–3020. pmid:8610161
  17. 17. Jablonski NG, Whitfort MJ, Roberts-Smith N, Xu Q. The influence of life history and diet on the distribution of catarrhine primates during the Pleistocene in eastern Asia. J Hum Evol. 2000;39: 131–157. pmid:10968926
  18. 18. Tougard C. Biogeography and migration routes of large mammal faunas in South-East Asia during the late Middle Pleistocene: Focus on the fossil and extant faunas from Thailand. Palaeogeog Palaeoclimatol Palaeoecol. 2001;168: 337–358.
  19. 19. Bacon A-M, Demeter F, Tougard C, de Vos J, Sayavongkhamdy T, Antoine PO, et al. Redécouverte d’une faune pléistocène dans les remplissages karstiques de Tan Hang au Laos: Premiers résultats. C R Palevol. 2008;7: 277–288.
  20. 20. Storm P, de Vos J. Rediscovery of the Late Pleistocene Punung hominin sites and the discovery of a new site Gunung Dawung in East Java. Senckenbergiana Lethaea. 2006;86: 271–281.
  21. 21. Harrison T, Krigbaum J, Manser J. Primate biogeography and ecology on the Sunda Shelf Islands: A paleontological and zooarchaeological perspective. In: Lehman SM, Fleagle J, editors. Primate Biogeography. New York: Springer; 2006. pp. 331–372.
  22. 22. Jablonski NG, Chaplin G. The fossil record of gibbons. In: Lappan S, Whittaker DJ, editors. The gibbons, Developments in Primatology: Progress and Prospects. New York: Springer; 2009. pp. 111–130.
  23. 23. Zeitoun V, Lenoble A, Laudet F, Thompson J, Rink WJ, Mallye JB, et al. The cave of the Monk (Ban Fa Suai, Chiang Dao Wildlife Sanctuary, northern Thailand). Quatern Int. 2010;220: 160–173.
  24. 24. Ingicco T, de Vos J, Huffman OF. The oldest gibbon fossil (Hylobatidae) from insular Southeast Asia: Evidence from Trinil, (East Java, Indonesia), Lower/Middle Pleistocene. PLOS ONE. 2014;9(6): e99531. pmid:24914951
  25. 25. Zhang Y, Jin C, Cai Y, Kono R, Wang W, Wang Y, et al. New 400–320 ka Gigantopithecus blacki remains from Hejiang Cave, Chongzuo City, Guangxi, South China. Quatern Int. 2014;354: 35–45.
  26. 26. Takai M, Zhang Y, Kono RT, Jin C. Changes in the composition of the Pleistocene primate fauna in southern China. Quatern Int. 2014;354: 75–85.
  27. 27. Harrison T, Jin C, Zhang Y, Wang Y. Fossil Pongo from the Early Pleistocene Gigantopithecus fauna of Chongzuo, Guangxi, southern China. Quatern Int. 2014;354: 59–67.
  28. 28. Van Gulik RH. The Gibbon in China: An Essay in Chinese Animal Lore. Leiden: EJ Brill; 1967.
  29. 29. Groves CP. Systematics and phylogeny of gibbons. In: Rumbaugh DM, editor. Gibbon and Siamang: Evolution, Ecology, Behavior, and Captive Maintenance. Basel: Karger; 1972. pp. 1–89.
  30. 30. Gao Y, Wen H, He Y. The change of historical distribution of Chinese gibbons (Hylobates). Zool Research. 1981;2: 1–8.
  31. 31. Ji W, Jiang X. Primatology in China. Int J Primatol. 2004;5: 1077–1092.
  32. 32. Geissmann T. Status reassessment of the gibbons: Results of the Asian Primate Red List Workshop 2006. Gibbon J. 2007;3: 5–15.
  33. 33. Fan PF, Huo S. The northern white-cheeked gibbon (Nomascus leugogenys) is on the edge of extinction in China. Gibbon J. 2009;5: 44–52.
  34. 34. Mittermeier RA, Rylands AB, Wilson DE. Handbook of Mammals of the World. Volume 3. Primates. Barcelona: Lynx Edicions; 2013.
  35. 35. Matthew WD, Granger W. New fossil mammals from the Pliocene of Szechuan, China. Bull Amer Mus Nat Hist. 1923;48: 568–598.
  36. 36. Teilhard de Chardin P, Young CC, Pei WC, Chang HC. On the Cenozoic formations of Kwangsi and Kwangtung. Bull Geol Soc Chin. 1935;14: 179–205.
  37. 37. Colbert EH. Pleistocene mammals from the Ma Kai Valley of northern Yunnan, China. Am Mus Nov. 1940;1099: 1–10.
  38. 38. Colbert EH, Hooijer DA. Pleistocene mammals from the limestone fissures of Szechuan, China. Bull Am Mus Nat Hist. 1953;102: 1–134.
  39. 39. Kahlke HD. On the complex of the Stegodon-Ailuropoda-fauna of southern China and the chronological position of Gigantopithecus blacki V. Koenigswald. Vert PalAsiat. 1961;2: 83–108.
  40. 40. Chen SK, Pang LB, He CD, Wei GB, Huang WB, Yue ZY, et al. New discoveries from the classic Quaternary mammalian fossil area of Yanjinggou, Chongqing, and their chronological explanations. Chin Sci Bull. 2013;58: 3780–3778.
  41. 41. Frisch JE. Trends in the evolution of the hominoid dentition. In: Hoffer H, Schultz AH, Starck D, editors. Bibliotheca Primatologica. Volume 3. Basel: Karger; 1965. pp. 1–130.
  42. 42. Prouty LA, Buchanan PD, Pollitzer WS, Mootnick AR. A presumptive new hylobatid subgenus with 38 chromosomes. Cytogenet Cell Genet. 1983a;35: 141–142.
  43. 43. Prouty LA, Buchanan PD, Pollitzer WS, Mootnick AR. Bunopithecus: A genus-level taxon for the hoolock gibbon (Hylobates hoolock). Am J Primatol. 1983b;5: 83–87.
  44. 44. Mootnick A, Groves C. A new generic name for the hoolock gibbon (Hylobatidae). Int J Primatol. 2005;26: 971–976.
  45. 45. Frisch JE. The hylobatid dentition. In: Rumbaugh DM, editor. Gibbon and Siamang. Volume 2: Anatomy, Dentition, Taxonomy, Molecular Evolution and Behavior. Basel: Karger; 1973. pp. 56–95.
  46. 46. Swindler DR. Primate Dentition. New York: Cambridge University Press; 2002.
  47. 47. Hayashi S, Hayasaka K, Takenaka O, Horai S. Molecular phylogeny of gibbons inferred from mitochondrial DNA sequences: Preliminary report. J Mol Evol. 1995;41: 359–365. pmid:7563122
  48. 48. Roos C, Geissmann T. Molecular phylogeny of the major hylobatids divisions. Mol Phylogenet Evol. 2001;19: 486–494. pmid:11399155
  49. 49. Chatterjee HJ. Phylogeny and biogeography of gibbons: a dispersal-vicariance analysis. Int J Primatol. 2006;27: 699–712.
  50. 50. Chatterjee HJ. Evolutionary relationships among the gibbons: A biogeographic perspective. In: Lappan S, Whittaker DJ, editors. The Gibbons, Developments in Primatology: Progress and Prospects. New York: Springer; 2009. pp. 13–36.
  51. 51. Fabre P-H, Rodrigues A, Douzery EJP. Patterns of macroevolution among primates inferred from a supermatrix of mitochondrial and nuclear DNA. Mol Phylogenet Evol. 2009; 53: 808–825. pmid:19682589
  52. 52. Thinh VN, Mootnick AR, Geissmann T, Li M, Ziegler T, Agil M, et al. Mitochondrial evidence for multiple radiations in the evolutionary history of small apes. BMC Evol Biol. 2010;10: 74. pmid:20226039
  53. 53. Chan Y-C, Roos C, Inoue-Murayama M, Inoue E, Shih C-C, Pei KJC, et al. Mitochondrial genome sequences effectively reveal the phylogeny of Hylobates gibbons. PLOS ONE. 2010;5(12): e124419.
  54. 54. Chan Y-C, Roos C, Inoue-Murayama M, Inoue E, Shih C-C, Vigilant L. A comparative analysis of Y chromosome and mtDNA phylogenies of the Hylobates gibbons. BMC Evol Biol. 2012;12: 150. pmid:22909292
  55. 55. Matsudaira K, Ishida T. Phylogenetic relationships and divergence dates of the whole mitochondrial genome sequences among three gibbon genera. Mol Phylogenet Evol. 2010; 55: 454–459. pmid:20138221
  56. 56. Perelman P, Johnson WE, Roos C, Seuánez HN, Horvath JE, Moreira MA, et al. A molecular phylogeny of living primates. PLOS Genet. 2011;7(3): e1001342. pmid:21436896
  57. 57. Israfil H, Zehr SM, Mootnick AR, Ruvolo M, Steiper ME. Unresolved molecular phylogenies of gibbons and siamangs (Family: Hylobatidae) based on mitochondrial, Y-linked, and X-linked loci indicate a rapid Miocene radiation or sudden vicariance event. Mol Phylogenet Evol. 2011;58: 447–455. pmid:21074627
  58. 58. Meyer TJ, McLain AT, Oldenburg M, Faulk C, Bourgeois MG, Conlin EM, et al. An Alu-based phylogeny of gibbons (Hylobatidae). Mol Biol Evol. 2012;29: 3441–3450. pmid:22683814
  59. 59. Springer MS, Meredith RW, Gatesy J, Emerling CA, Park J, Rabosky DL, et al. Macroevolutionary dynamics and historical biogeography of primate diversification inferred from a species supermatrix. PLOS ONE. 2012;7(11): e49521. pmid:23166696
  60. 60. Finstermeier K, Zinner D, Brameier M, Meyer M, Kreuz E, Hofreiter M, et al. A mitogenomic phylogeny of primates. PLOS ONE. 2013;8(7): e69504. pmid:23874967
  61. 61. Carbone L, Harris RA, Gnerre S, Veeramah KR, Laurent-Galdos B, Huddleston J, et al. Gibbon genome and the fast karyotype evolution of small apes. Nature. 2014;513, 195–201. pmid:25209798
  62. 62. Purvis A. A composite estimate of primate phylogeny. Phil Trans R Soc Lond B. 1995;348: 405–421. pmid:7480112
  63. 63. Porter CA, Page SL, Czelusniak J, Schneider H, Schneider MPC, Sampaio I, et al. Phylogeny and evolution of selected primates as determined by sequences of the Ɛ-globin locus and 5' flanking regions. Int J Primatol. 1997;18: 261–295.
  64. 64. Raaum RL, Sterner KN, Noviello CM, Stewart CB, Disotell TR. Catarrhine primate divergence dates estimated from complete mitochondrial genomes: concordance with fossil and nuclear DNA evidence. J Hum Evol. 2005;48: 237–257. pmid:15737392
  65. 65. Bininda-Emonds ORP, Cardillo M, Jones KE, MacPhee RDE, Beck RMD, Grenyer R, et al. The delayed rise of present-day mammals. Nature. 2007;446: 507–512. pmid:17392779
  66. 66. Pilbrow VC. Dental variation in African apes with implications for understanding patterns of variation in species of fossil apes. Ph.D. Dissertation, New York University. 2003.
  67. 67. Bailey SE. Neandertal dental morphology: implications for modern human origins. Ph.D. Dissertation, Arizona State University. 2002.
  68. 68. Wood BA, Abbott SA. Analysis of the dental morphology of Plio-Pleistocene hominids I. Mandibular molars: crown area measurements and morphological traits. J Anat. 1983;136: 197–219. pmid:6403498
  69. 69. Wood BA, Abbott SA, Graham SH. Analysis of the dental morphology of Plio-Pleistocene hominids II. Mandibular molars: study of cusp areas, fissure pattern and cross sectional shape of the crown. J Anat. 1983;137: 287–314. pmid:6415025
  70. 70. Bailey SE. A morphometric analysis of maxillary molar crowns of Middle- Late Pleistocene hominins. J Hum Evol. 2004;47: 183–198. pmid:15337415
  71. 71. Quam R, Bailey SE, Wood B. Evolution of M1 crown size and cusp proportions in the genus Homo. J Anat. 2009;214: 655–670. pmid:19438761
  72. 72. Uchida A. Craniodental variation among the great apes. Cambridge: Peabody Museum Bulletins 4; 1996.
  73. 73. Pilbrow V. Population systematics of chimpanzees using molar morphometrics. J Hum Evol. 2006;51: 646–662. pmid:16965803
  74. 74. Pilbrow V. Dental and phylogeographic patterns of variation in gorillas. J Hum Evol. 2010;59: 16–34. pmid:20494403
  75. 75. Bailey SE, Pilbrow VC, Wood BA. Interobserver error involved in independent attempts to measure cusp base areas of Pan M1s. J. Anat. 2004;205: 323–331. pmid:15447691
  76. 76. Wood BA, Engleman CA. Analysis of the dental morphology of Plio-Pleistocene hominids, V. Maxillary postcanine tooth morphology. J Anat. 1988;161: 1–35. pmid:3254883
  77. 77. Hammer Ø, Harper DAT, Ryan PD. PAST: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontologia Electronica. 2001;4: 1–9.
  78. 78. Groves CP. Primate Taxonomy. Washington, D.C.: Smithsonian Institution Press; 2001.
  79. 79. Marshall J, Sugardjito J. Gibbon systematics. In: Swindler DR, Erwin J, editors. Comparative Primate Biology, Volume 1: Systematics, Evolution, and Anatomy. New York: Alan R. Liss; 1986. pp. 137–185.
  80. 80. Geissmann T. Gibbon systematics and species identification. Int Zoo News. 1995;42: 467–501.
  81. 81. Dahlberg AA. The changing dentition of man. J Am Dent Assoc. 1945;32: 676–690.
  82. 82. Hooijer DA. The orang-utan in Niah Cave pre-history. Sarawak Mus. J. 1961;9: 408–421.
  83. 83. Hooijer DA. Prehistoric bone: The gibbons and monkeys of Niah Great Cave. Sarawak Mus J. 1962;10: 428–449.