Mitochondrial DNA hypervariable region-1 sequence variation and

American Journal of Primatology 69:1285–1306 (2007)

RESEARCH ARTICLE Mitochondrial DNA Hypervariable Region-1 Sequence Variation and Phylogeny of the Concolor Gibbons, Nomascus KERI MONDA1,2, RACHEL E. SIMMONS1,3, PHILIPP KRESSIRER1,4, BING SU1,5, 1 AND DAVID S. WOODRUFF 1 Ecology, Behavior & Evolution Section, Division of Biological Sciences, University of California, San Diego, La Jolla, California 2 Department of Epidemiology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 3 Genomic Variation Laboratory, Department of Animal Science, University of California, Davis, Davis, California 4 Medical Center of the Ludwig-Maximilians-University, Munich, Germany 5 Key Laboratory of Cellular and Molecular Evolution, Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, Yunnan, China

The still little known concolor gibbons are represented by 14 taxa (five species, nine subspecies) distributed parapatrically in China, Myanmar, Vietnam, Laos and Cambodia. To set the stage for a phylogeographic study of the genus we examined DNA sequences from the highly variable mitochondrial hypervariable region-1 (HVR-1 or control region) in 51 animals, mostly of unknown geographic provenance. We developed gibbon-specific primers to amplify mtDNA noninvasively and obtained 4477 bp sequences from 38 gibbons in North American and European zoos and 4159 bp sequences from ten Chinese museum skins. In hindsight, we believe these animals represent eight of the nine nominal subspecies and four of the five nominal species. Bayesian, maximum likelihood and maximum parsimony haplotype network analyses gave concordant results and show Nomascus to be monophyletic. Significant intraspecific variation within N. leucogenys (17 haplotypes) is comparable with that reported earlier in Hylobates lar and less than half the known interspecific pairwise distances in gibbons. Sequence data support the recognition of five species (concolor, leucogenys, nasutus, gabriellae and probably hainanus) and suggest that nasutus is the oldest and leucogenys, the youngest taxon. In contrast, the subspecies N. c. furvogaster, N. c. jingdongensis, and N. leucogenys siki, are not recognizable at this otherwise informative genetic locus. These results show that HVR-1 sequence is variable enough to define evolutionarily significant units in Nomascus and, if coupled with multilocus microsatellite or SNP genotyping, more than adequate to characterize their phylogeographic history. There is an urgent need to obtain DNA from Correspondence to: Dr. David Woodruff, Ecology, Behavior & Evolution Section, Division of Biological Sciences, University of California, San Diego, UCSD, La Jolla, CA 92093-0116. E-mail: [email protected]

Received 23 August 2006; revised 10 January 2007; revision accepted 12 March 2007 DOI 10.1002/ajp.20439 Published online 23 April 2007 in Wiley InterScience (www.interscience.wiley.com).

r 2007 Wiley-Liss, Inc.

1286 / Monda et al.

gibbons of known geographic provenance before they are extirpated to facilitate the conservation genetic management of the surviving animals. Am. J. Primatol. 69:1285–1306, 2007. c 2007 Wiley-Liss, Inc. Key words: Hylobatidae; control region; noninvasive genotyping; species; molecular evolution; taxonomy

INTRODUCTION Until recently, war and habitat inaccessibility frustrated scientific study of the Indochinese concolor gibbons of the genus Nomascus. They are variously referred to as the black or crested gibbons but these names are problematic; although adult males are black, adult females are primarily buff-colored, and the crest varies from prominent to indistinct, even within a species. The various species and subspecies are poorly defined and, not surprisingly, their taxonomic and phylogenetic relationships are still controversial [Bartlett, 2005]. To put our ignorance of these neglected apes in perspective consider the following report: ‘On the morning of 24 October 1999, the survey team recorded the songs of three gibbon pairs in Nang Lu foresty the first scientifically confirmed direct evidence of living N. c. concolor in Vietnam for about 30 years’ [Geissmann et al., 2000:63]. Since then another two species, N. nasustus and N. hainanus, with the distinction of being among the most critically endangered primate species in the world, have also been recognized in northeastern Vietnam and Hainan Island, China [Geissmann, 2005a,b, 2006b; Geissmann et al., 2003; Groves, 2004; Roos, in prep.]. Here, we present genetic evidence clarifying the relationships of four of the five species and nine nominal subspecies (Fig. 1, Appendix 1). For most of its taxonomic history the genus Nomascus has been treated as a subgenus of Hylobates and thought to contain only one species, Hylobates (Nomascus) concolor [Groves, 1972; Marshall & Sugardjito, 1986; and references therein] with six subspecies: concolor, lu, hainanus, leucogenys, siki and gabriellae. Ma and Wang [1986] proposed two new Chinese subspecies jingdongensis and furvogaster, and, following Dao [1983] divided the former subgenus into two species: the southern Hylobates leucogenys, characterized by white or yellow cheek-whiskers in black (male) individuals, and the northern H. concolor, characterized by the lack of contrasting cheek-whiskers in males. Groves and Wang [1990] supported this species distinction and elevated the southern yellow-cheeked gibbon H. gabriellae as a third species. Geissmann [1993] also recognized three species, H. concolor, H. leucogenys, and H. gabriellae, and subsequently recognized a fourth species, Hylobates (Nomascus) sp. cf. nasutus [Geissmann, 1995; Geissmann et al., 2000]. Although Zehr [1999] still considered the concolor group as a single species, most researchers have favored dividing the subgenus into four or five species and eight or nine subspecies. Furthermore, with the growing realization of the molecular genetic distances and antiquity of the four gibbon subgenera relative to the age of other hominoid genera most recent workers have elevated Nomascus from subgeneric to generic status [Bartlett, 2005; Brandon-Jones et al., 2004; Geissmann, 1995; Groves, 2001, 2004; Hall et al., 1998; Mootnick & Groves, 2005; Nadler & Streicher, 2004; Roos & Geissmann, 2001]. We support this as our own genetic research confirms the monophyly of the concolor group and its relatively deep divergence from the other three clades [Simmons, 2005, Simmons et al., in prep].

Am. J. Primatol. DOI 10.1002/ajp

Nomascus gibbon phylogeny / 1287

Fig. 1. Outline map showing the historical distribution of Nomascus species and subspecies [from Konrad & Geissmann, 2006, modified]. Ranges are shown as they are thought to have occurred 50 years ago; current populations are reduced in extent and highly fragmented. Taxa are N. concolor (con) and its subspecies N. c. lu (lu), N. nasutus (nas), N. hainanus (hai), N. leucogenys (leu) and its subspecies N. l. siki (siki), and N. gabriellae (gab). The three ?-marks indicate uncertainties surrounding the identities and current northern ranges of Chinese gibbons surviving east of the Red River; the identity of gibbons now extirpated from a large area of southwestern Yunnan; and the identity of gibbons found between the ranges of N. l. siki and N. gabriellae. The map is deceptive as it does not show either the former 1,000 km northern extension of concolor gibbons into China or the fact that Hainan Island was broadly connected to both Vietnam and China during Pleistocene low sea stands. Following tradition, most range limits follow rivers or political boundaries.


1288 / Monda et al.

Following the work of Prouty et al. [1983a,b] gibbons of the genus Nomascus are defined as having 26 pairs of chromosomes (members of the other genera have 2n 5 38, 44 or 50) and their species-level taxonomy has been based on geographic distribution, pelage and other morphological characters. Unfortunately, historical range limits of taxa are not well documented and the occurrence of natural hybridization, which could effect species-level differentiation, remains contentious [Fooden, 1996; Geissmann, 2002, 2006a]. Furthermore, the morphological criteria are unreliable guides to taxonomy as anatomical and morphological variation in Nomascus are limited. Although identification of black adult males is fairly straightforward, it is difficult to determine specific or subspecific status of females (buff-colored as adults) and immature males [Mootnick, 2006]. Other morphological traits are even more problematic. For example, species-level differentiation between N. leucogenys and N. gabriellae was based on differences in the penis bone [Groves, 1993; Groves & Wang, 1990]. Unfortunately, only one N. gabriellae baculum had been studied at that time and Geissmann and Lim [1994] later demonstrated that there was a significant variability in baculum size and the particular bone examined was incomplete and attributed to the wrong species [Geissmann, 1995]. So although Corbet and Hill [1992] have tabulated alleged differences between the various taxa traditionally associated with the name Hylobates concolor it should be emphasized that, apart from geographic range, these are minor and of limited utility, and/or ignore population variability. Groves [1993] was justified in expressing frustration with the mosaic geographic patterns of character variation in the genus. Two other approaches to species recognition promise to contribute to the resolution of the controversies surrounding concolor taxonomy. First, behavior may provide a more reliable guide to the species-level identification of adults than morphology. All species of gibbons produce elaborate patterns of vocalization often referred to as ‘‘songs’’ and many workers have shown how they have species-specific and sex-specific inherited characteristics [see: Brockelman & Schilling, 1984; Geissmann, 1993, 2002; Geissmann & Nijman, 2006; Konrad & Geissmann, 2006; and references therein]. Unfortunately, song variability in the wild is only now being documented and the songs of captive animals can be misleading as there is a learned component in their songs. Nevertheless, we can anticipate a phylogeny that is supported by these species-specific behavioral attributes [Geissmann, 2002]. Second, several researchers have begun to describe genetic markers that should serve as unequivocal identifiers of an individual gibbon’s taxonomic identity and relationships. In this context gibbons have featured significantly in the development of molecular primatology [Di Fiore & Gagneux, 2006]. In the first demonstration of noninvasive genotyping involving a non-human primate, Woodruff [1990, 1993] provided evidence that H. lar was variable at a microsatellite locus amplified from hair. In a preliminary study of the genetic differences between gibbon taxa, Garza and Woodruff [1992] sequenced a 252 bp segment of the cytochrome b (cyt b) gene of the mitochondrial genome from N. leucogenys, N. gabriellae and an animal identified as siki. The study revealed genetic evidence of a closer relationship between N. leucogenys and siki than either taxon to N. gabriellae, and provided evidence against the then prevailing view of Nomascus as a monospecific clade. Garza and Woodruff [1994] subsequently showed that taxon-specific cyt b sequence variation could be used to distinguished individuals of the three concolor taxa then in North American zoos (N. leucogenys, siki, and gabriellae). However, this data set included only white-cheeked or buff-cheeked individuals and they were consequently unable to



assess differences that might exist between these taxa and the other black gibbons. Moreover, the number of informative sites in the cyt b gene was relatively low (5% within the sample) and thus limited the resolving power of the locus. To answer questions left unresolved by the cyt b study we therefore examined the more rapidly evolving mitochondrial hypervariable region (HVR-1, sometimes called the control region and less accurately termed the d-loop). This paper combines and updates our results from three unpublished theses and an untraceable report in Chinese [Kressirer, 1993; Monda, 1995; Simmons, 2005; Su et al., 1996]. To interpret our data we also review other reports on variation in gibbon mtDNA [Chatterjee, 2006; Roos, 2004; Whittaker, 2005; Zhang, 1997] and nuclear loci [Zehr, 1999; Zhang, 1997; Zhang et al., 2004]. The phylogenetic principles underpinning these analyses are explained by Wagele [2005]. MATERIALS AND METHODS Our samples representing ten species and subspecies were derived primarily from gibbons in North American zoos supplemented with specimens of otherwise unavailable taxa from zoos and museums in Europe and China (Appendix 2). Our main analysis was based on 476–513 bp sequences of the mtDNA HVR-1 in 33 individuals representing seven taxa. We were also able to incorporate the results of a 159 bp mtDNA HVR-1 sequence survey for 11 Chinese concolor gibbons representing five taxa, including the only data for N. c. furvogaster [Su et al., 1996]. An additional four published hominoid sequences were used as outgroups in various analyses. All genotyping was performed non-invasively and non-destructively based on mtDNA extracted and amplified from 1 to 3 plucked hairs/individual of live animals or museum skins. We developed a variety of primers to amplify up to approx. 500 bp of the HVR-1. Most amplifications were performed using combinations of four oligonucleotides: L15997 50 -CACCATTAGCACCCAAAGCT-30 [Monda, 1995] L16205 50 -AACACAACATGCTTACAAGC-30 [Kressirer, 1993] H16431 50 -GTTGGTGATTTCACGGAGGA-30 [Kressirer, 1993] H16498 50 -CCTGAAGTAGGAACCAGATG-30 [Monda, 1995] The primers are identified according to the human nucleotide reference sequence [Anderson et al., 1981] of the 30 end, and L or H to indicate light or heavy strand, respectively. Our laboratory protocols are described elsewhere [Monda, 1995] and are available from DSW. We took several precautions to reduce the risk of amplifying nuclear sequences of mitochondrial origin (numts) including re-extracting and verifying key determinations in a second laboratory without potential contamination from other non-human primate DNA [Simmons, 2005; Simmons et al., in prep.] and comparing our sequences with those reported for two whole mitochondrial sequences of H. lar and a partial HVR-1 sequence of H. lar. Verifications typically involved amplification of a 325 bp fragment using the primers H16431 and L16007 [50 -CCCAAAGCTAAAATTCTAA-30 ; Roos & Geissmann, 2001]. We aligned the sequences using Clustal X Version 1.81 [Thompson et al., 1997] and after verification by eye we constructed maximum likelihood, parsimony, and distance trees using PAUP Version 4.0b [Swofford, 1998] and Bayesian inference trees with MrBayes Version 3.0b [Huelsenbeck & Ronquist, 2001]. Parsimony analyses used default settings of PAUP and treated gap states as ‘‘missing data.’’ Distance analyses were performed with default settings of PAUP to optimize minimum evolution. All trees were then rooted by outgroup.


1290 / Monda et al.

Bootstraps of parsimony and distance methods were each composed of 1,000 replicates. The maximum likelihood phylogeny had four steps: a maximum parsimony tree starting point and then three consecutive heuristic searches using maximum likelihood criteria. The first search’s likelihood settings employed a general time reversible model with gamma-distributed rate variation across sites (GTR1G) and estimated base frequencies, rate matrix, and shape. Nearest neighbor interchange (NNI) perturbation drove the state changes. Two subsequent heuristic searches used a tree bisection-reconnection (TBR) swapping and the starting point, rate matrix, base frequencies, and shape of the previous search. The maximum likelihood tree was then determined from the saved trees of these three searches. Bayesian trees used the GTR1G model on four MCMC chains. Generation number and sample frequency were determined by running a brief preliminary analysis, checking for probability convergence, and estimating the values necessary for a large set of equally likely trees. After the second analysis, each Bayesian tree’s probabilities file was examined to verify that the likelihoods converged around a small value range before determining if an appropriate burn-in value had been chosen, and the analysis was repeated twice to ensure that the MCMC chains were long enough and had converged on the same tree. Ultimately, we used MrBayes to searched for 5,000,000 generations and with a burn-in value of 100,000. Network (4.111; www.fluxus-engineering.com) was used to construct haplotype networks using median joining [Bandelt et al., 1999] and maximum parsimony [Polzin & Daneschmand, 2003] algorithms. The program TCS [Clement et al., 2000] was used to construct a statistical parsimony network joining all haplotypes that connect with a 95% or greater frequency. RESULTS The HVR-1 sequences are more informative than the previously published cyt b sequences. Comparing a subset of ten representative gibbons over 477 bp we found 123 variable and phylogenetically informative sites, involving 104 transitions and 38 transversions (TS/TV ratio of 2.7:1). Overall, we found a total of 123 (26%) variable positions and 26 different haplotypes. Phylogenetic relationships among the eight Nomascus taxa represented by more than 476 bp of mtDNA HVR-1 sequence are presented in Figures 2 and 3. All phylogenetic trees show similar patterns of tree topology, with long branch lengths between outgroup species representing other hominoids (Homo, Pan, Pongo) (Fig. 2). The concolor gibbons form a monophyletic group (Fig. 2) within which we see consistent evidence for four clusters of sequences that correspond to the species-level taxa N. leucogenys, N. concolor, N. gabriellae, and N. nasutus (Figs. 2 and 3). Bootstrap values (62–91%) show low to moderate support for these species-level clades (Fig. 2A). All phylogenetic methods indicate N. nasutus is the most basal group, distinct from all other clades with highly significant scores for distance bootstrap (87%) and Bayesian posterior probability (1.00). Depending on which algorithm is used either N. gabriellae or N. concolor branch off next. N. leucogenys appears as the youngest clade regardless of method used. Our two N. nasutus sequences are very similar differing by only 2.6%. Such differences are relatively small compared with the divergence seen among the leucogenys and comparable to those found within gabriellae and concolor. Although we initially treated one of our specimens as N. hainanus, on the advice of Christian Roos we have concluded that the animal was misidentified (Appendix 2).



Fig. 2. Phylogenetic relations of representative Nomascus gibbons based on variation in the mtDNA hypervariable region-1. A: distance bootstrap tree; B: Bayesian tree; C: maximum likelihood tree. Gibbon taxa are described in Appendix 1.


1292 / Monda et al.

Fig. 3. Maximum parsimony haplotype network showing phylogenetic relations of representative Nomascus gibbons. Gibbons are the same as those described in Fig. 2.

Roos (in litt. to DSW, May 2006; in prep.) has now sequenced hainanus of known provenance and found them to be much more different from nasutus (cyt b 6.8%, HVR1 14%). This strongly suggests that our Berlin Museum specimen was misidentified. Bayesian (Fig. 2B) and maximum likelihood (Fig. 2C) trees show N. concolor jingdongensis within the concolor cluster; a distance bootstrap (Fig. 2A) shows it as a very closely related outgroup (nine mutations) to the three N. c. concolor gibbons (which vary among themselves by up to 12 mutations). The 19 gibbons identified as N. leucogenys form a very closely related cluster that is well separated from the cluster identified as N. gabriellae. N. gabriellae differs from N. leucogenys by three indels, five transversions and 11 transitions; sites 243–249 are especially variable and may be taxon-specific. The distinction between N. leucogenys and the monophyletic cluster of three N. gabriellae is supported at the 81% level in Bayesian and 62% levels in parsimony and distance. The gibbon identified as siki is seen as an outgroup to the N. leucogenys rather than N. gabriellae clade; parsimony, maximum likelihood and distance trees show siki as a closely related outgroup of the N. leucogenys. In all tree methods used, 15 N. leucogenys individuals form a monophyletic cluster with two prominent subclusters of closely related individuals we denote as clades A and B (as discovered by Monda, 1995 , and shown most clearly in Fig. 2A). We find approximately 2% divergence within each clade and 4% divergence between clades. These clades differ by four transitions at sites 288, 294, 307, 348. Two additional subclusters are moderately differentiated from the others (Fig. 2A shows two individuals of Clade C and one, leuc-16, in Clade D). The division between these subclusters is supported by high posterior probability scores (0.81) in the Bayesian (Fig. 2B) and distance bootstrap methods (85%) (Fig. 2A). There are 26 (5.1%) variable sites within clade A, and 25 (4.9%) within clade B, of which they share only six. Clade A is weakly supported in Bayesian analysis (posterior probability, 0.75) and weakly supported in parsimony (not shown) and distance bootstraps (Fig. 2A) (62 and 58%, respectively). Clade B is supported strongly in Bayesian analysis (0.98) but weakly in bootstraps (parsimony, 56%, distance, 75%). The maximum parsimony haplotype network (Fig. 3) shows the above patterns more clearly and also reveals N. leucogenys clades A and B, and a small



Fig. 4. Relationships among Chinese Nomascus gibbons. Most parsimonious tree based on 159 bp sequences of the HVR-1. Redrawn from Su et al. [1996: Fig. 2b] showing branch lengths proportional to the number of base substitutions.

group clustering with the N. l. siki individual as a closely related outlier. The median joining network gave a similar result. The TCS generated statistical parsimony network (not shown) recognized four clusters concordant with seven leucogenys A (excluding animal leuc 11) and five leucogenys B (excluding leuc 4), two leucogenys D, and the four concolor1jingdongensis. The remaining ten individuals (plus outgroups Hoolock hoolock and Symphalangus syndactylus) were not related at our stringent 95% level. Su et al. [1996] provide some additional information on sequence variation in five Chinese Nomascus taxa (Appendix 2). Unfortunately, and despite the use of the same primers as Kressirer [1993], technical problems experienced in amplifying mtDNA from the museum skin samples resulted in only 159 bp of comparable sequence for the six taxa compared. Su et al. found no indels and 39 (25%) of the sites (not including the outgroup Symphalangus) were variable with 8–23 transitions and 1–6 transversions. He used a 6:1 TS/TV ratio. The most parsimonious tree shown (Fig. 4) is identical to the maximum likelihood tree; tree length 5 78, CI 5 0.73. In agreement with our more comprehensive 477 bp analyses N. nasutus [misidentified in Su et al. as N. n. hainanus] is seen as the outgroup. The previously unknown N. c. furvogaster and N. c. jingdongensis are seen as clustering with N. c. concolor (0–2 TV; 0–17 TS). The two N. leucogenys from the town of Hekou on the Vietnamese border in southern Yunnan differ from one another at one transversion and eight transitions. DISCUSSION The results reported here were based primarily on DNA extracted from freshly plucked hair. Although shed hair is no longer regarded as a reliable source of high quality DNA [Morin & Goldberg, 2003; Thalmann et al., 2004; Woodruff, 2003] noninvasive genotyping has made it possible to study these apes which are otherwise very difficult to sample. We are confident the sequences presented here


1294 / Monda et al.

are verified mtDNA sequences and not confounded by mitochondrial pseudogenes (numts) or other hominoid contaminants [Bensasson et al., 2001; Simmons et al., in preparation; Thalmann et al., 2004]. Our results and their interpretation are critically dependent on the identification of the individuals studied. The specimen data presented in Appendix 2 will facilitate their replication and revision when necessary. The mtDNA relationships seen are supportive of, but less ambiguous than, the morphology based taxonomy. As this is the first report in which multiple individuals of each species have been considered, questions presupposing the availability of information on typical intraspecific genetic variation can now be addressed. In particular, the large sample of N. leucogenys (19 haplotypes: the average pairwise distance is 0.03570.014 with a 29.8% percent sequence difference) can now be compared to the recently published data on comparable variation in 46 Hylobates lar (0.02470.014 with a 17.9% sequence difference) [Woodruff et al., 2005]. These percent sequence differences are comparable to our estimates for taxa represented by smaller sample sizes: N. concolor (N 5 3; 0.049670.0145), N. gabriellae (N 5 3; 0.024770.0214), and N. nasutus (N 5 2; 0.0299). Including the single N. l. siki with the leucogenys changes the average pairwise distance only slightly: 0.03870.015. Such within taxon variability data will be invaluable in interpreting the observed between-taxa differences. At the mtDNA locus studied the concolor gibbons are a monophyletic clade. Their karyotypic distinction (2N 5 52) and mtDNA genetic distance from the other three groups of living gibbons supports their recognition as a separate genus. Divergence between Hylobates and Nomascus is described elsewhere but is comparable to or greater than that between Nomascus and either Symphalangus or Hoolock, with highly significant scores for distance bootstrap (100%) and Bayesian posterior probability (1.00) [Simmons et al., in prep.]. Within the genus are at least four species or species groups (N. concolor, N. leucogenys, N. nasutus, and N. gabriellae) that differ at more than the 5% divergence level; a level comparable to that seen in comparably documented species in Hylobates [Simmons et al., in prep]. Below that 5% divergence level it is clear that more genotyping of individuals of known geographic provenance is required to resolve the issues and that the following discussion of siki, jingdongensis, and furvogaster must be viewed as preliminary. Our genetic result showing N. nasutus as the basal group is in agreement with the findings of Roos [2004] which were based on his survey of cyt b variation. This conclusion is also supported by a preliminary phylogenetic tree based on vocal data for all gibbon species, including the four species of crested gibbons [Konrad & Geissmann, 2006]. Although the relevant bootstrap values in the behavioral analysis are very low, this tree appears to provide additional support for Groves [1993] hypothesis that, within Nomascus, northern species are more basal than the southern species. Nomascus nasutus is very different from Nomascus concolor, the northern gibbon with which they have been long confused. The genetic distance between our two samples is less than the variation between the A and B clades of N. leucogenys, or between the three N. gabriellae, or between the four N. c. concolor. This supports our argument that there is only one species represented in our sample and not two as reported earlier (see Appendix 1). Christian Roos (in litt. to DSW, April 2006) has recently discovered that Patzi (see Appendix 2) is genetically closely related to gibbons from the two remaining wild populations in Vietnam. His results, when published, should resolve the problems of the relationships between the surviving northeastern populations of gibbons.



In reconstructing the evolutionary relations between N. nasutus and N. hainanus it is important to realize that today’s isolated island population (Fig. 1) has been repeatedly connected with those on the mainland whenever sea levels fell 425 m (Appendix 1); Neogene hypothermal phase paleogeography could also have brought hainanus and siki into geographic proximity. Our data permit the first assessment of the genetic merits of the two subspecies of Chinese N. concolor named by Ma and Wang [1986]. Our 170 bp analysis involved four representatives of the central subspecies jingdongensis and they cluster closely with the three N. c. concolor. N. jingdongensis would not stand out as a distinct taxon if it were a H. lar or N. leucogenys. It is a little different from the southern N. c. concolor but such differences would be expected over the geographic distances involved. In the case of the western subspecies, furvogaster, we have only one museum specimen. Nevertheless, Su et al. [1996] found this individual’s sequence (Fig. 4) sits even closer to topotypic N. c. concolor than the four jingdongensis. This suggests that furvogaster has little merit if mtDNA genetic distances are the metric of subspecific status. There are two possible explanations for the multiple clades within N. leucogenys. One is that N. leucogenys contains several distinct haplotypes that exist in sympatry. This hypothesis can be tested by establishing the geographic distribution of the various clades, especially clades A and B. It is also possible that a cryptic taxon has been absorbed into today’s N. leucogenys. The range of N. leucogenys is large (Fig. 1), but not sufficiently large or geographically heterogeneous to lead one to expect to find multiple allopatric haplotypes. The average distance between A and B group individuals (0.036470.0018) is less than between subspecies in comparable groups, but by a very small margin [macaques Z0.041; Rosenblum et al., 1997; orangutans Z0.04117.0013; Warren et al., 2001]. Genotyping N. leucogenys individuals of known geographic provenance will be crucial to interpret the genetic variation. In either situation (sympatric variants or allopatric subspecies), it is desirable to maintain the full extent of this species’ variability in the captive population. Our mtDNA data do not support the elevation of siki to species status. Three of the four individuals that cluster with our H. l. siki are identified as N. leucogenys. Couturier and Lernould [1991] found no reliable morphological distinction between female siki and N. leucogenys; adult males are of course indistinguishably black with white cheeks. The pairwise differences between our sample identified as siki and the leucogenys are within the range of intraspecific variation seen in N. leucogenys and H. lar. Although Couturier and Lernould [1991] found their specimens of siki and leucogenys differed by a reciprocal translocation between chromosomes 1 and 22, no other workers appear to have karyotyped their samples and the phylogenetic significance of this discovery remains unexplored. Some additional data are provided by Zhang [1997] in a little known report on cytochrome b variation. Zhang compared five new 252 bp sequences for one H. hoolock of unknown origin, two northern concolor from the Vietnam–China border, and two N. leucogenys from northern Vietnam to 11 of the 26 sequences reported by Garza and Woodruff [1992]. His maximum parsimony analysis showed that Nomascus was a well-defined sister group (bootstrap value 5 98) to the other three genera (Hoolock and Syndactylus and Hylobates) and, within its cluster of nine individuals, he found siki as the outgroup to leucogenys, and well differentiated from both concolor and gabriellae. Although he argued that this result favored the recognition of siki as a separate species in our opinion the bootstrap value is too low (86) to warrant such revision.


1296 / Monda et al.

Two additional cytochrome b analyses have been reported. Hall et al. [1998] examined the entire 1,040 bp sequence in six species and all four genera but in only eight individuals. They confirmed the species-level divergence of leucogenys and gabriellae but found insufficient variation to resolve the relationships among the genera. Roos [2003, 2004] provides a more recent analysis of the entire sequences of 24 Nomascus. In close agreement with our results, he found four species-level taxa within his samples: nasutus, concolor, leucogenys, gabriellae. N. nasutus [represented by Patzi] had the deepest root (8% interspecific divergence). Six N. leucogenys and three N. gabriellae show moderate intraspecific divergence (4% each) and more than was found among nine N. concolor. Four siki were examined: two individuals were about 2% different from gabriellae and two were about 1% different from the leucogenys. In the absence of specimen and locality data it is not possible to interpret this finding of paraphyly in siki as it could be due to specimen misidentification and/or hybridization. Both natural and artificial (captive) hybrids have been identified [Mootnick, 2006, personal communication, May 2006] and, more generally, hybridization has to be recognized as a normal process in gibbon evolution [Arnold & Meyer, 2006]. If the historical taxonomic problems of the gibbons have arisen in part from the practice of defining taxa based on characters of single individuals we should be cautious not to further contribute to such malpractice. Although we have tried to go beyond dependence on a typological approach there are several issues that still frustrate the elucidation of concolor gibbon phylogeny. First, taxa should not be defined solely on the basis of maternally transmitted mtDNA patterns. Comparable genetic studies of nuclear variation (autosomal sequences, Y-chromosome sequences, and microsatellite allele frequencies) and karyotypes are essential components of contemporary species characterization. Although family-wide surveys have not yet been undertaken, an indication of their resolving power is illustrated by, for example, Chambers et al. [2004], Tanaka et al. [2004] and Zhang et al. [2004]. Second, vocalization, which may serve as a partial reproductive isolating mechanism must also be considered [Konrad & Geissmann, 2006]. Third, it is important to recognize that today’s geographic ranges for some species are recently contracted and fragmented. Subfossil teeth and old paintings show that until 1,000–2,000 years ago gibbons ranged 1,600 km further north in China to the level of the Huang (Yellow) River [Geissmann, 1995; Groves, 1972; Marshall & Sugardjito, 1986; van Gulik, 1967]; Hoolock and northern Nomascus have been extirpated over most of their historical ranges. Finally, there is a need to pursue additional karyological studies of gibbons as their chromosomes are very unusual among the mammals in exhibiting a ten-fold higher incidence of chromosomal rearrangements (especially translocations), most of which appear to be species-specific [Arnold et al., 1996; Couturier et al., 1992; Bigoni & Stanyon, 2006; Carbone et al., 2006; Hirai et al., 2005; Jauch et al., 1992; Koehler et al., 1995; Mueller et al., 2003; Nie et al., 2001]. As chromosomal rearrangements can function as postmating reproductive isolating mechanisms and contribute to stasipatric speciation [White, 1978; King, 1993] they warrant close examination in species complexes exhibiting parapatric distribution patterns like the gibbons. Effort should be made to characterize Nomascus karyologically before key populations are extirpated. Consideration of the processes of karyotypic evolution suggested to Mueller et al. [2003] that Nomascus was the last (youngest) hylobatid genus to diverge [between 10 and 5 Mya according to Groves, 2004]; a conclusion supported by our own work [Simmons et al., in prep.] and Chatterjee [2006], and at odds with Roos and



Geissmann [2001] who concluded that Nomascus was basal within the family. This difference of opinion is based on different data to those described here and with the increased concordance between data-sets it is clear that within a few years the phylogeny of the surviving gibbons will be fully resolved and stable. The concolor gibbons are variously listed by regional governments and NGOs as globally threatened or, in the cases of N. nasutus and N. hainanus, critically endangered [IUCN, 2006]. Gibbons are hunted for food, the alleged medicinal value of their parts, and as trophies and pets [Sterling et al., 2006]. In addition, habitat degradation results in the survivors living in small isolated patches of forest and their disappearance across much of their historic range [Chivers, 2005; Konstant et al., 2003; D.S. Woodruff, personal observation in Xishuangbanna, July 2006]. This is vividly illustrated by the widely separated confirmedoccurrence dots on recent maps of Vietnam [Geissmann et al., 2000; Nadler & Streicher, 2004]. Time is fast disappearing to document the genetic variability of the remaining animals and provide a foundation for the sound conservation management of both captive and free-ranging gibbons. The real promise of multilocus genetic data is that it will permit a partial reconstruction of the original phylogeographic patterns. Multilocus genotyping of all available museum specimens and noninvasive genotyping of the surviving wild individuals will permit resolution of the taxonomic ambiguities and the estimation of historical rates of gene flow and natural hybridization. ACKNOWLEDGMENTS We thank Kurt Benirschke, Warren Brockelman, Helen Chatterjee, Thomas Geissmann, Colin Groves, Alan Mootnick and Christian Roos for sharing their knowledge of gibbons. Ronald Tilson and Cathy Castle facilitated Monda’s original study of the AZA gibbons. Francine Frasier, Dawn Field, Hopi Hoekstra, John Huelsenbeck, Pascal Gagneux, Nick Mundy, Sukamol Srikwan and Romel Hokanson provided advice or technical assistance in the laboratory or with the phylogenetic analyses. Our studies were supported, in part, by the AZA Gibbon Taxon Advisory Group, the US National Science Foundation, the Chinese Academy of Sciences, the University of Munich, and the University of California. REFERENCES Anderson S, Bankier AT, Barrell BG, de Bruijin MHL, Coulson AR, Drouin J, Eperon IC, Nierlich DP, Roe BA, Sanger F, Schreier PH, Smith AJH, Staden R, Young IG. 1981. Sequence and organization of the human mitochondrial genome. Nature 290:457–465. Arnold ML, Meyer A. 2006. Natural hybridization in primates: one evolutionary mechanism. Zoology 109:261–276. Arnold N, Stanyon R, Jauch A, O’Brien P, Weinberg J. 1996. Identification of complex chromosome rearrangements in the gibbon by fluorescent in situ hybridization (FISH) of a human chromosome 2q specific microlibrary, yeast artificial chromosomes, and reciprocal chromosome painting. Cytogenet Cell Genet 74:80–85. Baker LR, Geissmann T, Nadler T, Long B, Walston J. 2002. Cambodia: Primate field

guide. Phnom Penh, Cambodia: Fauna & Flora Cambodia. 8 p. Bandelt H-J, Forster P, Rohl A. 1999. Medianjoining networks for inferring intraspecific phylogenies. Mol Biol Evol 16:37–48. Bartlett TQ. 2005. The hylobatidae. In: Campbell CJ, Fuentes A, MacKinnon KC, Panger M, Bearder SK, editors. Primates in perspective. New York: Oxford University Press. p 274–289. Bensasson D, Zhang D-X, Hartl DL, Hewitt GM. 2001. Mitochondrial pseudogenes: evolution’s misplaced witnesses. TREE 16: 314–321. Bigoni F, Stanyon R. 2006. Hylobates concolor. In: O’Brien SJ, Menninger JC, Nash WG, editors. Atlas of mammalian chromosomes. Hoboken, NJ: Wiley. p 160. Bleisch WV, Jiang X. 2000. Action plan for conservation of the gibbons of the Wuliang


1298 / Monda et al. Mountains. Sino-Dutch Forest Conservation and Community Development Project, Kunming, Yunnan, China. Bleisch W, Zhang Y. 2004. The view across the border: China and the future of Vietnamese primates. In: Nadler T, Streicher U, Ha Thang L, editors. Conservation of primates in Vietnam. Hanoi: Frankfurt Zool Soc, Vietnam Primate Conserv Progr, Cuc Phuong National Park, Endangered Primate Rescue Center. p 107–114. Brandon-Jones D, Eudey AA, Geissmann T, Groves CP, Melnick DJ, Morales JC, Shekelle M, Stewart CB. 2004. Asian primate classification. Int J Primatol 25:97–164. Brockelman WY, Schilling D. 1984. Inheritance of stereotyped gibbon songs. Nature 312:634–636. Carbone L, Vessere GM, Hallers BFH, Zhu B, Osoegawa K, Mootnick M, Kofler A, Wienberg J, Rogers J, Humphray S, Scott C, Harris RA, Milosavljevic A, de Jong PA. 2006. A high-resolution map of synteny disruptions in gibbon and human genomes. PLoS Genet 2:2162–2175. Chambers KE, Reichard UH, Moller A, Nowak K, Vigilant L. 2004. Cross-species amplification of human microsatellite markers using noninvasive samples from white-handed gibbons (Hylobates lar). Am J Primatol 64: 19–27. Chan BPL, Fellowes JR, Geissmann T, Zhang J, editors. 2005. Status survey and conservation action plan for the Hainan gibbon— version I (last updated November 2005). Hong Kong: Kadoorie Farm & Botanic Garden Technical Report 3. p 1–33. Chatterjee HJ. 2006. Phylogeny and biogeography of gibbons: a dispersal-vicariance analysis. Int J Primatol 27:699–712. Chivers DJ. 2005. Gibbons: the small apes. In: Caldecott J, Miles L, editors. World atlas of great apes and their conservation. Berkeley: University of California Press. p 205–214. Clement M, Posada D, Crandall KA. 2000. TCS: a computer program to estimate gene genealogies. Mol Ecol 9:1657–1659. Corbet GB, Hill JE. 1992. The mammals of the Indomalayan Region: a systematic review. Oxford: British Museum (Natural History) & Oxford University Press. Couturier J, Lernould JM. 1991. Karyotypic study of four gibbon forms provisionally considered as subspecies of Hylobates (Nomascus) concolor (Primates, Hylobatidae). Folia Primatol 56:95–104. Couturier J, Dutrillaux B, Turleau C, DeGrouchy J. 1992. Comparative karyotyping of four gibbon species or subspecies. Ann Genet 25:5–10. Dao VT. 1983. On the north Indochinese gibbons (Hylobates concolor) (Primates:


Hylobatidae) in North Vietnam. J Hum Evol 12:367–372. DiFiore A, Gagneux P. 2006. Molecular primatology. In: Campbell CJ, Fuentes A, MacKinnon KC, Panger M, Bearder SK, editors. Primates in perspective. New York: Oxford University Press. p 369–393. Fooden J. 1996. Zoogeography of Vietnamese primates. Int J Primatol 67:845–899. Fooden J, Quan G, Luo Y. 1987. Gibbon distribution in China. Acta Theriol Sin 7: 161–167. Garza JC, Woodruff DS. 1992. A phylogenetic study of the gibbons (Hylobates) using DNA obtained noninvasively from hair. Mol Phylogen Evol 1:202–210. Garza JC, Woodruff DS. 1994. Crested gibbon (Hylobates [Nomascus]) identification using noninvasively obtained DNA. Zoo Biol 13: 383–387. Geissmann T. 1989. A female black gibbon, Hylobates concolor subspecies, from northeastern Vietnam. With appendix: a note on the Laotian black gibbon, H. concolor lu. Int J Primatol 10:455–476. Geissmann T. 1993. Evolution of communication in gibbons (Hylobatidae). Ph.D. dissertation. Zurich University, Switzerland. Geissmann T. 1995. Gibbon systematics and species identification. Int Zoo News 42: 65–77. Geissmann T. 1997. New sounds from the crested gibbons (Hylobates concolor group): first results of a systematic revision. In: Zissler D, editor. Verhandlungen der Deutschen Zoologischen Gesellschaft: Kurzpublikationen, 90. Jahresversammlung, Gustav Fischer, Stuttgart. p 170. Geissmann T. 2002. Taxonomy and evolution of gibbons. Evol Anthropol 11(Suppl 1): 28–31. Geissmann T. 2005a. Der Hainan-Schopfgibbon: Der bedrohteste Menschenaffe der Welt. Gibbon J 1:10–12. Geissmann T. 2005b. Auf der Suche nach den letzten Gibbons von Hainan. Gibbon J 1: 18–22. Geissmann T. 2006a. Gibbon research lab. Available from: http://www. gibbons.de Geissmann T. 2006b. Schutz des HainanSchopfgibbons, des seltensten Menschenaffen der Welt: ein Projektbericht. Gibbon J 2: 11–13. Geissmann T, Lim KKP. 1994. Extraction of bacula from tanned gibbon skins. Raffles Bull Zool 42:29–41. Geissmann T, Nijman V. 2006. Calling in wild silvery gibbons (Hylobates moloch) in Java (Indonesia): behavior, phylogeny, and conservation. Am J Primatol 68:1–19. Geissmann T, Nguyen Xuan D, Lorme´e N, Momberg F. 2000. Vietnam primate conservation status review 2000. Part 1:

Nomascus gibbon phylogeny / 1299 Gibbons. Hanoi: Fauna & Flora International, Indochina Programme. 139p. Geissmann T, La Quang T, Trinh Dinh H, Vu Dinh T, Dang Ngoc C, Pham Duc T. 2003. Rarest ape species rediscovered in Vietnam. Asian Primates 8(3–4):8–9. Geissmann T, Traber S, von Allmen A. 2006. Das Nangunhe-Naturreservat, Provinz Yunnan, China: Ein Projektbericht [Nangunhe Nature Reserve, Yunnan Province, China: a project report]. Gibbon J 2:14–17. Groves CP. 1972. Systematics and phylogeny of gibbons. In: Rumbaugh, DM, editor. Gibbon and siamang, vol. 1. Basel: Karger. p 1–89. Groves CP. 1993. Speciation in living hominoid primates. In: Kimbel WH, Martin LB, editors. Species, species concepts, and primate evolution. New York: Plenum. p 109–121. Groves CP. 2001. Primate taxonomy. Washington, DC: Smithsonian Institution Press. Groves CP. 2004. Taxonomy and biogeography of primates in Vietnam and neighbouring regions. In: Nadler T, Streicher U, Ha Thang Long, editors. Conservation of primates in Vietnam. Hanoi: Frankfurt Zool Soc, Vietnam Primate Conservation Programme, Cuc Phuong National Park, Endangered Primate Rescue Center. p 15–22. Groves CP, Wang Y. 1990. The gibbons of the subgenus Nomascus (Primates, Mammalia). Zool Res 11:148–154. Hall LM, Jones D, Wood B. 1998. Evolution of the gibbon subgenera inferred from cytochrome b DNA sequence data. Mol Phylogen Evol 10:281–286. Hirai H, Wijayanto H, Tanaka H, Mootnick AR, Hayano A, Perwitasari-Farajallah D, Iskandriati D, Sajuthi D. 2005. A whole-arm translocation (WAT8/9) separating agile gibbons and its evolutionary features. Chromosome Res 13:123–133. Huelsenbeck JP, Ronquist F. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17:754–755. IUCN. 2006. IUCN Red list of threatened species. Available from: http://www.iucnredlist. org. Jauch A, Weinberg J, Stanyon R, Arnold N, Tofanelli S, Ishida T, Cremer T. 1992. Reconstruction of genomic rearrangements in great apes and gibbons by chromosome painting. Proc Natl Acad Sci USA 89: 8611–8615. Jiang XL, Wang YX. 1999. Population and distribution of black-crested gibbons (Nomascus concolor jingdongensis) in Wuliang Nature Reserve, Jingdong, Yunnan. Zool Res 20:421–425. King M. 1993. Species evolution: the role of chromosome change. Cambridge: Cambridge University Press.

Koehler U, Bigoni F, Weinberg J, Stanyon R. 1995. Genomic reorganization in the concolor gibbon (Hylobates concolor) revealed by chromosome painting. Genomics 30: 287–292. Konrad R, Geissmann T. 2006. The vocal diversity and taxonomy of crested gibbons (genus Nomascus) in Cambodia. Int J Primatol 27:713–745. Konstant WR, Mittermeier RA, Rylands AB, Butynski TM, Eudey AA, Ganzhorn J, Kormos R. 2003. The world’s top 25 most endangered primates. Asian Primates 8: 29–34. Kressirer P. 1993. Eine Molekulare Phylogenie der Gibbons (Hylobatidae). Munchen: Diplomarbeit der Fakultat fur Biologie der Ludwig-Maximilians-Universitat. Ma S, Wang Y. 1986. The taxonomy and distribution of gibbons in southern China and its adjacent region, with description of three new subspecies. Zool Res 7: 393–410. Ma S, Wang Y, Poirier FE. 1988. Taxonomy, distribution, and status of gibbons (Hylobates) in southern China and adjacent areas. Primates 29:277–286. Marshall J, Sugardjito J. 1986. Gibbon systematics. In: Swindler DR, Erwin J, editors. Comparative primate biology, vol. 1: systematics, evolution and anatomy. New York: Liss. p 137–185. Monda K. 1995. A phylogenetic study of the gibbons (Hylobates) using the control region of the mitochondrial genome. San Diego: Master’s thesis, University of California. Mootnick AR. 2006. Gibbon (Hylobatidae) species identification recommended for rescue or breeding centers. Primate Conserv 21:103–138. Mootnick AR, Groves CP. 2005. A new generic name for the hoolock gibbon (Hylobatidae). Int J Primatol 26:971–976. Morin PA, Goldberg TL. 2003. Determination of genealogical relationships from genetic data: a review of methods and applications. In: Chapais B, Berman CM, editors. Kinship and behavior in primates. Oxford: Oxford University Press. p 15–45. Mueller S, Hollatz M, Weinberg J. 2003. Chromosomal phylogeny and evolution of gibbons (Hylobatidae). Hum Genet 113: 493–501. Nadler T, Streicher U. 2004. The primates of Vietnam—an overview. In: Nadler T, Streicher U, Ha Thang L, editors. Conservation of primates in Vietnam. Hanoi: Frankfurt Zool Soc, Vietnam Primate Conservation Programme, Cuc Phuong National Park, Endangered Primate Rescue Center. p 5–11. Nie W, Rens W, Wang J, Yang F. 2001. Conserved chromosome segments in


1300 / Monda et al. Hylobates hoolook revealed by human and H. leucogenys paint probes. Cytogenet Cell Genet 92:248–253. Polzin T, Daneschmand SV. 2003. On Steiner trees and minimum spanning trees in hypergraphs. Oper Res Lett 31:12–20. Prouty LA, Buchanan PD, Pollitzer WS, Mootnick AR. 1983a. Bunopithecus: a genus-level taxon for the hoolock gibbon (Hylobates hoolock). Am J Primatol 5:83–87. Prouty LA, Buchanan PD, Pollitzer WS, Mootnick AR. 1983b. A presumptive new hylobatid subgenus with 38 chromosomes. Cytogen Cell Genet 35:141–142. Roos C. 2003. Molekulare Phylogenie der Halbaffen, Schlankaffen und Gibbons. PhD thesis. Germany: Technical Univ Munich. Roos C. 2004. Molecular evolution and systematics of Vietnamese primates. In: Nadler T, Streicher U, Ha Thang L, editors. Conservation of primates in Vietnam. Hanoi: Frankfurt Zool Soc, Vietnam Primate Conservation Programme, Endangered Primate Rescue Center, Cuc Phuong National Park. p 23–28. Roos C, Geissmann T. 2001. Molecular phylogeny of the major hylobatid divisions. Mol Phylogen Evol 19:486–494. Rosenblum LL, Supriatna J, Melnick DJ. 1997. Phylogeographic analysis of pigtail macaque populations (Macaca nemestrina) inferred from mitochondrial DNA. Amer J Phys Anthro 104:35–45. Sheeran, LK, Zhang Y, Poirier FE, Yang D. 1998. Preliminary report on the behavior of the Jingdong black gibbon (Hylobates concolor furvogaster). Trop Biodiv 5:113–125. Simmons RE. 2005. A phylogenetic study of the gibbons (Hylobatidae). Master’s thesis. San Diego: University of California. Simmons RE, Monda K, Woodruff DS. Phylogenetics of gibbons (Hylobatidae): mitochondrial and nuclear DNA sequence data resolve long-standing questions. in prep. Sterling EJ, Hurley MM, Minh LD. 2006. Vietnam: a natural history. New Haven: Yale University Press. Su B, Kressirer P, Wang W, Jiang X, Wang YX, Woodruff DS, Monda K, Zhang YP. 1996. Molecular phylogeny of Chinese N. c. concolor gibbons. Sci China (C) 26:414–419. Swofford DL. 1998. Phylogenetic analysis using parsimony and other methods (PAUP Version 4.0b). Sunderland, MA: Sinauer. Takacs Z, Morales JC, Geissmann T, Melnick DJ. 2005. A complete species-level phylogeny of the Hylobatidae based on mitochondrial ND3-ND4 gene sequences. Mol Phylogen Evol 36:456–467. Tanaka H, Wijayanto H, Mootnick AR, Iskandriati D, Perwitasari-Farajallah D, Sajuthi D, Hirai H. 2004. Molecular phylogenetic


analyses of subspecific relationships in agile gibbons (Hylobates agilis) using mitochondrial and TSPY gene sequences. Folia Primatol 75(Suppl 1):418. Thalmann O, Hebler J, Poinar HN, Paabo S, Vigilant L. 2004. Unreliable mtDNA data due to nuclear insertions: a cautionary tale from analysis of humans and other great apes. Mol Ecol 13:321–335. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. 1997. The CLUSTAL-X windows interface: Flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25:4876–4882. Trung LQ, Hoang TD. 2004. Status review of the Cao Vit black-crested gibbon (Nomascus nasutus nasutus) in Vietnam. In: Nadler T, Streicher U, Ha Thang L, editors. Conservation of primates in Vietnam. Hanoi: Frankfurt Zool Soc, Vietnam Primate Conservation Progr. Cuc Phuong National Park, Endangered Primate Rescue Center. p 90–94. van Gulik RH. 1967. The gibbon in China. An essay in Chinese animal lore. Leiden: Brill. Wagele J-W. 2005. Foundations of phylogenetic systematics. Munich: Pfeil. Wang Y. 2003. A complete checklist of mammal species and subspecies in China—a taxonomic and geographic reference. Beijing: China Forestry Publishing House. 394p. Warren KS, Verschoor EJ, Langenhuijzen S, Swan RA, Vigilant L, Heeney JL. 2001. Speciation and intrasubspecific variation in Bornean orangutans, Pongo pygmaeus pygmaeus. Mol Biol Evol 18:472–480. White MJD. 1978. Modes of speciation. San Francisco: Freeman. Whittaker DJ. 2005. Evolutionary genetics of Kloss’s gibbons (Hylobates klossii): systematics, phylogeography, and conservation. Ph.D. thesis. The City University of New York. 201p. Woodruff DS. 1990. Genetics and demography in the conservation of biodiversity. J Sci Soc Thailand 16:117–132. Woodruff DS. 1993. Non-invasive genotyping of primates. Primates 34:333–346. Woodruff DS. 2003. Noninvasive genotyping and field studies of free-ranging nonhuman primates. In: Chapais B, Berman CM, editors. Kinship and behavior in primates. Oxford: Oxford University Press. p 46–68. Woodruff DS, Monda K, Simmons RE. 2005. Mitochondrial DNA sequence variation and subspecific taxonomy in the white-handed gibbon, Hylobates lar. Nat Hist J Chulalongkorn Univ 1:71–78. Zehr SM. 1999. A nuclear and mitochondrial phylogeny of the lesser apes (Primates, genus Hylobates). Ph.D. thesis. Harvard University. 221p.

Nomascus gibbon phylogeny / 1301 Zhang Y. 1992. Hainan gibbon (Hylobates concolor hainanus) threatened. Asian Primates 2:6. Zhang Y, Sheeran L. 1994. Current status of the Hainan black gibbon (Hylobates concolor hainanus). Asian Primates 3:3. Zhang YP. 1997. Mitochondrial DNA sequence evolution and phylogenetic relationships of gibbons. Acta Genet Sin 24:231–237.

Zhang YW, Luo HR, Ryder OA, Zhang YP. 2004. Evolution of the tandem repeats in thymidylate synthase enhancer region (TSER) in primates. Gene 338:47–54. Zhou J, Wei F, Li M, Zhang J, Wang D, Pan R. 2005. Hainan black-crested gibbon is headed for extinction. Int J Primatol 26: 453–465.

APPENDIX 1 Annotated list of currently recognized Nomascus species and subspecies Nomascus concolor Harlan, 1826. Western black gibbon or black crested gibbon. The tortuous taxonomic history of N. concolor is discussed by Geissmann [1989]. Our genetic data are in agreement with the cyt b study by C. Roos (unpublished) and support the recognition of this taxon as a full species, and do not support the recognition of three parapatric subspecies—concolor, jingdongensis, furvogaster. Endangered [IUCN, 2006] with less than 2,000 individuals in China, Vietnam and Laos. Recent reports include Bartlett [2005], Bleisch and Zhang [2004], Sheeran et al. [1998], Nadler and Streicher [2004]. Nomascus concolor concolor Harlan, 1826. Black gibbon or Tonkin black crested gibbon. South China (eastern Yunnan) and northwest Vietnam, between the Black (Song Da) and Red (Song Hong) rivers. Range historically parapatric with N. c. jingdongensis, N. nasutus, and N. leucogenys [Geissmann et al., 2000; Nadler & Streicher, 2004], but interactions with those taxa are undocumented and probably no longer occur. N. concolor jingdongensis Ma and Wang, 1986. Black gibbon or central Yunnan black crested gibbon. South China (central Yunnan) between the Mekong and Black rivers, Wenbu, Jingdong, and Wuliang Mountains [Bleisch & Jiang, 2000; Jiang & Wang, 1999; Sheeran et al., 1998]. Extirpated from most of range [Geissmann et al., 2000, 2006]. Our genetic data support subjugation within N. concolor, a view supported by Brandon-Jones et al. [2004], Geissmann et al. [2000] and C. Roos (unpublished), as the diagnostic features involve only minor color differences in females. Nomascus concolor furvogaster Ma and Wang, 1986. Black gibbon or west Yunnan black crested gibbon. South China (western Yunnan), between the Mekong and Salween rivers. [Ma & Wang, 1986; Ma et al., 1988]. Genetic data reported here support subjugation within N. concolor, a view supported by Geissmann [1995], Geissmann et al. [2000] as the diagnostic features were based on a subadult female. Geissmann et al. [2006] provide limited recent observations. Nomascus concolor lu Delacour, 1951. Loatian black crested gibbon. Bokeo Province, Laos, westernmost Laos at Ban Nam Khueng and Khao Tham Phra on the Mekong river and Nam Kan valley at about 201N, where Geissmann recently confirmed their presence and identity [Geissmann et al., 2000; see also Mootnick, 2006]. This is an isolated allopatric population of a few hundred individuals separated today from the main population of N. concolor by 250 km occupied by N. leucogenys. Probably a synonym of N. c. concolor with minor differences possibly a result of limited hybridization with N. leucogenys [Geissmann, 1989]. Until genetic data are available to test this hypothesis the subspecies should be recognized; preliminary genetic data support subjugation within N. concolor [C. Roos, unpublished].


1302 / Monda et al.

Nomascus nasutus Kunckel d’Herculais, 1884. Eastern black crested gibbon. Northeast Vietnam, north and east of the Red River delta and the adjacent coastal China. Largely extirpated, 20–28 gibbons were re-discovered in Cao Bang and Hoa Binh Provinces in 2002 [Bleisch & Zhang, 2004; Geissmann et al., 2000; Trung & Hoang, 2004]. Critically Endangered [IUCN, 2004; Konstant et al., 2003]. This taxon was referred to awkwardly for many years as Hylobates (Nomascus) sp. cf. nasutus because of uncertainties over the source of the type specimen (see notes in Appendix 2) and its relation to N. concolor [see Geissmann, 1989, 2002]. Genetically, we have shown that it is clearly specifically distinct from N. concolor, a conclusion reached independently by Takacs et al. [2005]. Until recently only one captive animal (Patzi, see Appendix 2) had been examined and her relationship to the remaining wild gibbons had not been established. Patzi was unusual in looking like a concolor gibbon but having a distinctive hainanus-like vocalization [Geissmann, 1989, 1997] and she may have been a hybrid or an unrecognized taxon [Geissmann et al., 2002; Groves, 2004]. However, Roos (in litt. April 2006) has just genotyped apes from both surviving Vietnamese populations and found them to be very closely related to Patzi at the cyt b sequence studied. Furthermore, as reported in the Results above, he discovered the mainland gibbons were significantly different from Hainan Island animals at this locus and accordingly proposes to elevate nasutus to species rank [Roos, in prep.]; a decision anticipated by Nadler and Streicher [2004]. N. hainanus Thomas 1892. Hainan black crested gibbon. The population isolated today on Hainan Island, China, is now restricted to the Bawangling Nature Reserve [Zhang, 1992; Zhang & Sheeran, 1994]. Dao [1983] thought the island gibbons were subspecifically distinct from nasutus and several museum specimens in Vietnamese collections are referred to the subspecies N. n. hainanus. Roos’ (in prep, in litt. to DSW) cyt b genetic sequence divergence data indicate, however, that although related to nasutus, hainanus merits full species level status. It is unknown when gibbons first arrived on the continental island of Hainan but they could have dispersed across broad (50–70 km wide) dry land connections during one or more Pleistocene hypothermal phase(s). Today’s population has been physically isolated for less than 10,000 years, since the sea rose above –25 m and flooded the narrow Qiongzhou Strait. Their genetic divergence from nasutus suggests they have been separated for much longer and Chatterjee [2006] estimates the species antiquity to be 0.3–1.8 My. The identity of mainland gibbons referred to this species cannot be accepted without genetic confirmation. Groves [2001] and Wang [2003] treat hainanus as a full species, without comment. Now critically endangered, this population fell from an estimated 2,000 animals in the 1950s, to 23 in 1998, and 13 in 2004 [Bleisch & Zhang, 2004; Chan et al., 2005; Chivers, 2005; Geissmann et al., 2000; Zhou et al., 2005]. N. leucogenys Ogilby, 1840. Northern white-cheeked crested gibbon. Historical range: China (Mengla county, Xishuangbanna, southern Yunnan), northern Laos, and northwestern Vietnam [Fooden et al., 1987; Geissmann et al., 2000]. Corbet and Hill [1992] reported it is sympatric with N. concolor but this seems unlikely as the historical interactions with other taxa are undocumented. IUCN [2006]: data deficient. Endangered in Vietnam and o100 individuals in China [Bleisch & Zhang, 2004; Geissmann et al., 2000; Nadler & Streicher, 2004]. The southern populations are referred to the subspecies N. l. siki. N. leucogenys siki Delacour, 1951. Southern white-cheeked crested gibbon. Southern Laos, central Vietnam. During the last 15 years various authors have included siki as either a subspecies of N. gabriellae [Corbet & Hill, 1992; Groves & Wang, 1990], or N. leucogenys [Geissmann, 1995], or treated it as a separate species [Groves, 2001, 2004; Zhang, 1997]. The size and shape of the white cheek



patches of adult males and juveniles are diagnostic features for siki, but adult females of N. leucogenys and siki are morphologically indistinguishable although both differ from females of N. gabriellae [Geissmann, 1995; Geissmann et al., 2000; Mootnick, 2006]. The songs of the three taxa are different [Konrad & Geissmann, 2006] and siki resembles N. leucogenys more than that of any other form of crested gibbon including N. gabriellae [Konrad & Geissmann, 2006]. The boundary between N. leucogenys and siki lies near the lower Ca River south of the town of Vinh and east of the Annamite Mountains in Nghe An Province, Vietnam. Further south, N. l. siki is replaced by N. gabriellae, but the southern range limits of N. l. siki are undocumented. Konrad and Geissmann [2006] have described vocalizations in northeast Cambodia (Rattanakiri) and provisionally assign those apes to N. l. siki because of their resemblance to calls of topotypic siki from nearby Bach Ma, Vietnam. They postulate that a taxon boundary exists somewhere between Rattanakiri and southern Mondulkiri in eastern Cambodia, and discuss the roles of the Srepok river and dry dipterocarp forest as possible distribution barriers. Our mitochondrial DNA sequences suggest that siki is more closely related to N. leucogenys than to N. gabriellae [Garza & Woodruff, 1992, 1994; Zhang, 1997; herein] but we recognize that a couple of specimens reported by others have the opposite affinity; we regard the problem as unresolved until animals of known geographic provenance are genotyped. Konrad and Geissmann [2006] discuss the possibility that siki and gabriellae are separated by a broad zone of intergradation (hybridization) or, alternatively, that a currently unrecognized taxon occupies a large area of south central Vietnam between their ranges. N. gabriellae Thomas, 1909. Yellow-cheeked (or buff-cheeked) crested gibbon. Eastern Cambodia, southern Laos, and southern Vietnam, south of 151300 . Historically thought to be parapatric or hybridize with siki, but their interactions are poorly documented and the two species are allopatric in Cambodia today [Baker et al., 2002; Geissmann et al., 2000; Konrad & Geissmann, 2006]. Endangered.

APPENDIX 2 Specimens examined North American zoo gibbons used in this study All taxonomic identifications are those provided by the owners and are the same as those recorded in the AZA concolor studbook [1990] and ISIS. For each individual the following data are provided: [the ID number used in Figures in this paper in square brackets], Studbook number/ISIS number, Sex, House name, Holding Zoo, Origin, GenBank accession number, and no. of base pairs (bp) of mtDNA sequenced. Notes: Studbook number: two individuals had no studbook number (]nsb). Sex and House name: male or female and name (or not available, na). Holding Zoo: at the time hair sample was provided, typically 1984–5. Origin: captive born (cb) or wild born (wb); country or zoo of origin or unknown.

Nomascus leucogenys (24 individuals) [leuc 1] 11/32055C, F Betsy, National, wb, unknown, EF203867, 497 bp. [leuc 2] 18/92, F na, Minnesota, wb unknown, EF203868, 530 bp. [leuc 3] 21/1345, F Muneca, Gladys Porter, wb unknown, EF203869, 486 bp. [leuc 4] 24/36233, M Joe, National, wb unknown, EF203870, 510 bp. [leuc 5] 28/36336, F


1304 / Monda et al.

Beryl National, wb unknown, EF203871, 497 bp. [leuc 6] 30/92, F China, Gladys Porter, wb unknown, EF203872, 485 bp. [leuc 7] 31/93, F Goldie, Gladys Porter Zoo, wb unknown, EF203873, 480 bp. [leuc 8] 42/1318, F Phyllis, Washington Park, wb unknown, EF203874, 285 bp. [leuc 9] 43/101675, F Mae, National, wb unknown, EF203875, 521 bp. [leuc 10] 53/101519, M Gilly, Cheyenne, wb unknown, EF203876, 300 bp. [leuc 11] 55/91, M Archie, Minnesota, wb unknown, EF203877, 300 bp. [leuc 12] 196/107858, M Mekong, National, cb National, EF203878, 485 bp. [leuc 13] 213/5926, F Minnesota, cb Minnesota, EF203879, 513 bp. [leuc 14] 229/109732, F Melaka, National, cb National, EF203880, 176 bp. Comparable sequences were obtained from the following individuals but are not shown here as they are subsumed within the clusters revealed by the above animals [Monda, 1995; Woodruff et al., in prep.]. Some are the descendants of females listed above and were found to have identical mtDNA, as expected. 1 nsb/110265, M, Beryl’s baby, National, unknown, 474 bp. 2 nsb/91200, ?F, Hue, Washington Park, cb Washington Park 285 bp. 14/1325, M, Gunther, Washington Park, wb unknown. 79/100739, M, Bert, National, cb, National, 521 bp. 134/105020, M, Ralph, National, cb National, son of 28 and sib of 1 nsb. 143/107869, F, Siam, National, cb, National, 499 bp. 146/2044, F, Deborah, Gladys Porter, cb unknown, 536 bp. 207/108291, F, Burma, National, cb National, daughter of 28 and sib of 1 nsb. 209/88071, M, Tanh Linh, Washington Park, cb Washington Park, 291 bp. 231/109740, F, Sena, National, cb National, 474 bp.

Nomascus gabriellae (2) [gabr 1], 122/94241, F Robin, Los Angeles, cb Los Angeles, daughter of 65/ 94111, F Bahmetoo, Los Angeles, wb S. Vietnam and 64/94110, M Koo, Los Angeles, wb S. Vietnam, EF203886, 492 bp. [gabr 2], 180/95141, M Yang Menggangu, Los Angeles, cb Los Angeles, sib of 122, EF203887, 492 bp. Symphalangus syndactylus (1) [Symphalangus], 267/870020, F, Juice, Cheyene Mountain Zoo, cb Cheyene, EF203866, 385 bp. Samples studied by Kressirer [1993] Gibbons were adults unless noted and all identifications were provided by the owners. For each individual the following data are provided: [the ID number used in this paper in square brackets], Kressirer [1993] specimen no., Sex, House name, Holding Zoo or Museum, mtDNA source, GenBank accession number.

N. l. leucogenys [leuc 15] 4 M juvenile (parents: Jack & Jacqueline), Mulhouse Zoo, tissue sample 1986, EF203883, 476 bp. [leuc 16] 280 M Claude, Mulhouse Zoo, blood sample 1986, EF203881, 476 bp. Used by Kressirer [1993] to represent this taxon and identified as P3leuc in some analyses by him, and subsequently by Monda [1995] and Simmons [2005]. [leuc 17] 281 M Jack, Mulhouse Zoo, blood sample 1986, EF203882, 476 bp. [leuc 18] 342 M Dodo, Budapest Zoo, arrived from southeast Asia in 1968, hair sample 1991, EF203884, 476 bp. N. l. siki [siki 1] 20 M Charly, Hellabrunn Zoo, Munich, hair sample in 1991, EF203885, 495 bp. Used by Kressirer [1993] to represent this taxon and identified



as by him as P1siki in some of his analyses, and subsequently by Monda [1995] and Simmons [2005]. Identical but shorter sequences were obtained from the following individuals but are not used here: 21 F Charlotte, Hellabrunn Zoo, Munich, hair sample in 1991; 22 F Mimi, Hellabrunn Zoo, Munich, hair sample in 1991.

N. gabriellae [gabr 3] 340 M Tschico, Budapest Zoo, arrived from Laos in 1987, hair sample 1991, EF203888, 407 bp. Used by Kressirer [1993] to represent this taxon and identified as by him as P2gabriellae in some of his analyses, and subsequently by Monda [1995] and Simmons [2005]. An identical but shorter sequence was obtained from the following individual but is not used here: 341 F Juschka, Budapest Zoo, transferred from Moscow Zoo 1991, born ca. 1968 in Vietnam, hair sample 1991. N. nasutus [nasu 1] 409/410 F Patzi, Humboldt University Museum, ZMB 70036, from Tierpark Berlin (1962–1986), reported to have been shipped from Hon Gai (north-eastern Vietnam) but geographic origin unknown, skin sample, EF203889, 446 bp. This individual is discussed in detail by Geissmann [1989] who regards is as sufficiently different in coloration from the other known northeastern Vietnamese black gibbons to warrant subspecific status. Used by Kressirer [1993] to represent this taxon and identified as by him as P4nasutus in some of his analyses, and subsequently by Monda [1995], Su et al. [1996] and Simmons [2005]. [nasu 2] Kressirer ]0.2, M, Museum Naturkunde, Berlin, 85357, captive animal labeled as N. n. hainanus from ‘‘Houchow (?), China’’, EF203890, 519 bp. Our efforts to establish the source locality of this specimen have failed; there is no Houchou on Hainan Island and there is a reasonable possibility that this animal was actually from the mainland as an anonymous reviewer of this manuscript has kindly drawn our attention to a locality named Hou Chau (or Kou Chau) at 221380 N, 104150 E in Vietnam. Used by Kressirer [1993] to represent this taxon and identified as by him as P5hainanus in some of his analyses, and subsequently by Monda [1995], Su et al. [1996] and Simmons [2005]. We now believe this specimen was misidentified and report it herein as N. nasutus. Partial sequences obtained by Su et al. [1996] For each individual the following data are provided: [ID number in this paper] specimen identification ] in Su et al. [1996], sex, collection locality, mtDNA source, collection year, GenBank accession number. Skins are in the Museum of the Kunming Institute of Zoology unless noted. Comparable sequences were 143 bp, and all shared a 16 bp deletion relative to Symphalangus. [leuc 19] N. leucogenys-1, M, Hekou, Yunnan, fresh hair sampled, 1994, EF212884. [leuc 20] N. leucogenys-2, M, Hekou, Yunnan, fresh hair sampled, 1994, EF212885. [jing 1] N. concolor jingdongensis-1, F, Jingdong, Yunnan, collected 1964, old skin, EF203891. [jing 2] N. c. jingdongensis-2, M, Jingdong, Yunnan, collected 1964 old skin, EF212886. [jing 3] N. c. jingdongensis-3, M, Jingdong, Yunnan, collected 1964 old skin, EF212887. [jing 4] N. c. jingdongensis-4, F, Jingdong, Yunnan, old skin, details unknown, EF212888.


1306 / Monda et al.

[furv 1] N. c. furvogaster, M, Cangyuan, Yunnan, collected 1983, old skin, EF212889. [conc 1] N. c. concolor-1, sex unknown, Jianshui, Yunnan, collected 1987 old skin, EF203893. [conc 2] N. c. concolor-2, F, Luchum, Yunnan, collected 1972, old skin, EF203894. [conc 3] N. c. concolor-3, F Jianshui, Yunnan, collected 1989, old skin, EF203892. [nasu 2] N. nasutus hainanus is Kressirer [1993] P5hainanus: ]0.2, M, Museum Naturkunde, Berlin, 85357, captive animal from ‘‘Houchow (?), China’’, and now thought to be N. nasutus from Vietnam (see above). [Symphalangus] Symphalangus syndactylus is Monda’s ]267/870020, F, Juice, Cheyene Mountain Zoo, cb Cheyene. Outgroup sequences from GenBank An additional four sequences were used as outgroups in some analyses: [Hoolock], Hoolock (formerly Bunopithecus) hoolock (AF311725)[Roos and Geissmann]; [numt] a S. syndactylus nuclear insertion (numt) of the HVR-1 (AF035467); [Pongo] Pongo pygmaeus (D38115); [Homo] Homo sapiens (NC001807).
