DistributionRead full entry
Range DescriptionTragulus williamsoni is only known from the holotype which was collected at Meh Lem, Muang Pre, Song forest of Thailand (18Â°25'N, 100Â°23'E); also referred to as Me Song (Meijaard and Groves 2004). The species' true range cannot be determined until more material of Tragulus is examined from the northern highlands of South-east Asia and adjacent southern China; this will require serious further collecting. The lack of subsequent records of T. williamsoni may largely reflect the paucity of collecting in recent decades, the fact that most surveyors and other biologists in the region have been unaware of the taxonâs distinctiveness, and so have not sought it and may well have overlooked it when they came across it. The collections in China and Thailand (and Viet Nam), where additional specimens of T. williamsoni would be most likely to be found, were not studied by Meijaard and Groves (2004) and remain to be critically examined. Thus, the current restriction of records to the type locality cannot be seen as even weakly indicative of a genuinely restricted distribution. The type locality is west of the Mekong and this river is a biogeographic barrier for some species (Meijaard and Groves 2006, Timmins and Duckworth in press) but it should not be assumed to be so for T. williamsoni. While all specimens (about four) examined from east of the Mekong from localities lying along the southern edge of the northern highlands of Indochina have all proved to be T. kanchil (E. Meijaard pers. comm. 2008), this is a small sample size and does not represent the actual highland area. Most significantly, two specimens from Yunnan province (China) here provisionally assigned to this species (below) come from east of the Mekong. Specifically, given the known presence of Tragulus across a large area of the northern highlands of Lao PDR far from any specimens validated to either T. williamsoni or T. kanchil (Bergmans 1995, Duckworth et al. 1999: 269; Johnson et al. 2003), it is an open question whether T. williamsoni occurs in Lao PDR. Records from the area of north Lao PDR west of the Mekong (Xaignabouli province), based on village interviews but surely reliably indicative of the genus given the strength of the distribution pattern in Lao PDR from this nation-wide dataset (Duckworth et al. 1999: 269), should be priorities to determine to species. Two specimens from Mengla, Xishuanbanna, southern Yunnan province, China (21Â°32â N, 101Â°36â E) appear to be T. williamsoni on the basis of some skull measurements (the skulls are damaged and knowledge of intraspecific variation in this taxon is obviously weak given only a single specimen from the region of the type locality) (Wang Yingxiang in litt. to E. Meijaard pers. comm. 2008). Tragulus, nominally identified as T. javanicus (=, here, T. kanchil) occur in northern Viet Nam (Dang Huy Huynh et al. 1994), but the distributional map suggests records are likely to have been from the lowlands.
The type locality and the data from China suggest that this species might be associated with highland terrain and higher altitudes than T. kanchil, but this is no more than speculation at present. The genus Tragulus has an odd northerly distribution within South-east Asia, being entirely unrecorded from Myanmar (except the far south) (Tun Yin 1967, Lynam 2003, Meijaard and Groves 2004; Shan state, adjacent to highland areas of northern Lao PDR and Thailand which hold the genus, has not been well investigated and may yet be found to hold the species), apparently absent from the east of Lao PDRâs northern highlands (and the central northern highlands remain to be investigated; Duckworth et al. 1999: 269), present in northern Viet Nam but quite probably restricted to the lowlands (see above), and penetrating China only in the far southwest, in Xishuangbanna prefecture of Yunnan Province (Smith and Xie 2008). The T. williamsoni holotype appears to be the only Tragulus record critically identified to species from this extensive area, making the absence of confirmed T. kanchil records biologically uninformative.