|Scientific Name:||Muntiacus rooseveltorum|
|Species Authority:||Osgood, 1932|
Muntiacus feae Osgood, 1932 subspecies rooseveltorum
|Taxonomic Notes:||This taxonomic status of M. rooseveltorum has been controversial for over 60 years, and the species was even relegated by Groves and Grubb (1990) to the synonymy of M. feae. However, genetic analysis of the holotype (Amato et al. 1999a, 1999b, 2000) and morphological investigations (R.J. Timmins pers. comm. 2008) confirm the species' validity with respect to all previously-named taxa. More complex is its relationship with several subsequently proposed names, M. truongsonensis, M. putaoensis, and M. puhoatensis.
M. rooseveltorum forms a clade with M. truongsonensis and M. putaoensis supported genetically (Amato et al. 1999b, 2000; James et al. in press) and morphologically, by the presence of the combination of well developed canines in females and small, generally unbranched, but regularly shed, antlers in males (R.J. Timmins pers. comm. 2008). The nominal taxon M. puhoatensis is also very likely to belong to this clade (R.J. Timmins pers. comm. 2008). It is still unclear how many species-level taxa there are in this clade, here referred to as the M. rooseveltorum species-complex (because rooseveltorum is the earliest available name within the complex). At present, diagnostic characters for individual taxa are limited to base sequences of four genes, although for M. puhoatensis even this is not available (Amato et al. 1999b, 2000); the dangers inherent in using mtDNA data to determine taxonomic and phylogenetic relationships among closely related populations urge caution in drawing firm conclusions without establishing congruence with variation of nuclear DNA (both autosomal and Y-chromosome) and morphological characters (Ballard and Whitlock 2004). No such analysis has been performed within this species-complex. Neither the congruence of various morphological features speculated to be important in identifying species within this complex, nor the variation in each character within each apparent species, have been analysed: too few specimens exist for such an analysis. Specimens are particularly few which allow the full set of skull and pelage characters to be assessed, and where skull and skin are known to be from the same original animal. The accounts for M. truongsonensis and M. puhoatensis (described from one frontlet [antlers and a bit of the skull] lacking any morphological characteristics likely to be necessary for species diagnosis within this species-complex discusses) consider some linked taxonomic and nomenclatorial ambiguities specific to the application of those two names which in turn affect application of the name M. roosveltorum. Synonymy of either or both M. truongsonensis and M. puhoatensis with M. rooseveltorum cannot be ruled out, although one or both could be valid species.
The animals of northern Myanmar taken to be M. putaoensis shows pelage similarities to M. rooseveltorum, specifically reduced dark frontal stripes, short rather than long crown hair tuft and generally rather uniform, and pale, body and tail pelage. There are several camera-trap photographs of animals from Lao PDR and Viet Nam (outside the presumed range of M. putaoensis) of M. roosveltorum-type animals which show pelage of pale sandy brown to darker orange similar to that shown by animals of northern Myanmar believed to be M. putaoensis (R.J. Timmins pers. comm. 2008). M. putaoensis is generally assumed to be geographically isolated from M. rooseveltorum, and current known localities are indeed widely disjunct. However, the lack of known specimens in intervening areas would be expected given the paucity of suitable surveys. The Mekong is a geographical barrier to some mammals (e.g. squirrels; Timmins and Duckworth in press), but not others, especially geographically and altitudinally wider ranging taxa in the northern highlands north to the headwater areas of China (e.g. the gibbon Hylobates (Nomascus) leucogenys). Given its altitudinal and geographic range, M. rooseveltorum is likely to belong in the latter category.
These uncertainties mean that M. rooseveltorum. can be considered to be currently known only from the holotype. One further specimen with skin and skull is very likely to belong to this species (R.J. Timmins pers. comm. 2008 based on material collected by R. Steinmetz and Tanya Chanard). A number of other specimens (poorly documented with respect to provenance and morphological features) which have been reported to share the same mitochondrial DNA sequences as the holotype (Amato et al. 1999a) are also likely to belong to the species, but two closely related species might not be diagnosable with respect to partial analysis of their mtDNA (Ballard and Whitlock 2004), as exemplified by the analogous case of M. gongshanensis and M. crinifrons (see 2008 Red List account for M. gongshanensis).
Würster-Hill and Seidel (1985) claimed to have analysed the karyotype of M. rooseveltorum, but given the lack of information on the source of their specimen and the means by which it was identified as M. rooseveltorum, and the similarities in karyotype with northern red muntjac M. vaginalis, it has generally been assumed that their tissue originated from a specimen of M. vaginalis (e.g., Groves and Grubb 1990).
Ma et al. (1986) listed M. rooseveltorum for China (seemingly for the first time) with the following statement 'M. lachrymans teesdalei described by Lydekker (1915) as having relatively small preorbital pit and premaxillae separated from nasals should be M. rooseveltorum'. Although Lydekker's animal is definitely M. reevesi, Ma et al. (1986) were correct in M. rooseveltorum having the said features. Thus, some records identified as M. reevesi teesdalei in southern China may in fact not be of that taxon and could, given the features of this 1986 description, refer to animals of the M. rooseveltorum complex. Presence of the species-complex in southern China is extremely likely as specimens and photographs of animals in the M. rooseveltorum complex in both Lao PDR and Viet Nam have been obtained close to the Chinese border (R.J. Timmins pers. comm. 2008). Ma et al. (1986) also identified four purchased skins of this species in the Kunming Institute of Zoology as this species, but the identification was revised to M. reevesi by Groves and Grubb (1990) who apparently examined the four skins in the Kunming Institute of Zoology. There is no description of the skins in the above sources, and given the current understanding of muntjac systematics it would likely be hard to diagnose the specific identity of the skins (especially if heads and tails were incomplete or missing) without molecular analysis, their species identity is thus probably best treated as indeterminate at present.
The English name Roosevelt’s Muntjac is often used, but the scientific name is in the genitive plural, and the original account (Osgood 1932) makes clear that both Theodore and Kermit Roosevelt are to be commemorated. Hence, the appropriate English name is Roosevelts’ Muntjac.
|Red List Category & Criteria:||Data Deficient ver 3.1|
|Assessor(s):||Timmins, R.J., Duckworth, J.W. & Long, B.|
|Reviewer(s):||Black, P.A. & Gonzalez, S. (Deer Red List Authority)|
This species is listed as Data Deficient because current taxonomic uncertainties prevent clarification of status and distribution of this species within the many specimens and photographs of animals undoubtedly belonging to this species-complex (see taxonomic note). As a result there is no reasonable way to estimate its species-specific range or conservation status. General hunting levels in the region are high but there is no information about how tolerant the species may be to hunting (northern red muntjac is able to withstand high hunting, but there is no real reason to believe that trait this is shared with the rest of the genus, but equally nor is there any reason to believe that it is not). Habitat loss is probably a threat in some regions, but levels of tolerance of fragmentation and degradation cannot yet be assessed.
|Range Description:||The distribution of this species is very unclear (see taxonomic note). The type was collected in 1929 at Ban Muangyo (=Muong Yo, now locally known as Ban Yo) in northern Lao PDR, in Phongsali Province (Osgood 1932). The only additional localities pertain to a few specimens proposed by analysis of mtDNA to refer to M. rooseveltorum, found at 19°49’N, 103°45’E in northern Lao PDR (Amato et al., 1999a, 2000), and even these cannot be taken as certain, in case more than one species shares similar mtDNA. A female found in the Xe Sap NPA in southern Lao PDR by R. Steinmetz and Tanya Chanard in the late 1990s shows many morphological similarities to M. rooseveltorum (R.J. Timmins pers. comm. 2008). Published range descriptions (e.g. Groves and Schaller 2000; Grubb 2005) are hypothetical because under current uncertainty, there is no way objectively to determine the range of any taxon within this species-complex, including of M. rooseveltorum itself (see taxonomic note).
Many incomplete specimens, predominantly skulls and partial skins (largely not mutually associated), and camera-trap photos, all of which present characters apparently diagnostic or characteristic of the M. rooseveltorum species-complex, have been collected over a large range of northern Indochina highlands (Lao PDR and Viet Nam), the northern and central Annamite range of Indochina (Lao PDR and Viet Nam), and northern Myanmar. A few specimens at the Viet Nam National University, Hanoi, suggest a presence east of the Red River. Records of M. putaoensis from northeastern India (Datta et al. 2003), initially identified morphologically, have been confirmed by recent genetic analysis of five specimens as M. putaoensis (James et al. press). There has been insufficient work in the southern Annamites of Viet Nam and in northern parts of Thailand to rule out presence of the species-complex there, which is possible (R.J. Timmins pers. comm. 2006). Ma et al. (1986) listed M. rooseveltorum for China (seemingly for the first time) with the following statement 'M. lachrymans teesdalei described by Lydekker (1915) as having relatively small preorbital pit and premaxillae separated from nasals should be M. rooseveltorum'. Although Lydekker's animal is definitely M. reevesi, Ma et al. (1986) were correct in M. rooseveltorum having the said features. Thus, some records identified as M. reevesi teesdalei in southern China may in fact not be of that taxon and could, given the features of this 1986 description, refer to animals of the M. rooseveltorum complex. Presence of the species-complex in southern China is extremely likely as specimens and photographs of animals in the M. rooseveltorum complex in both Lao PDR and Viet Nam have been obtained close to the Chinese border (R.J. Timmins pers. comm. 2008).
The camera-trap and specimen evidence suggests that in some, perhaps many, areas of Indochina two (or more) taxa within this species-complex are sympatric, or at least parapatric in a mosaic pattern. The situation is less clear in Myanmar, although the presence of multiple taxa there is also possible (see account for M. putaoensis). Contrary to various statements in the literature (e.g. Amato et al. 1999b) animals of this species-complex are not highly localised, but rather appear to be widespread, although this does not rule out that there might be individual species with small ranges. M. rooseveltorum may yet be found to occur throughout the range of the complex, but equally it could be much more localised. Specifically, despite the molecular genetic analysis of a relatively large number of specimens from Myanmar and adjacent India (Amato et al. 1999a; James et al. press), there is no suggestion of animals with DNA similar to the M. rooseveltorum holotype in these countries.
There have been various unsubstantiated claimed localities for species-level identifications of both M. rooseveltorum and M. truongsonensis in the grey literature. Notably, an animal observed in captivity at ‘Lak Sao’ (= Ban Lak-20), Lao PDR, 18°20'N, 106°00'E and identified by Amato et al. (1999a) as M. rooseveltorum, should not at present be identified; this animal presented several key external characteristics at variance with the holotype of M. rooseveltorum (Timmins 1996; Amato et al. 2000; R.J. Timmins pers. comm. 2008). However, M. rooseveltorum may well occur in this area, because there are camera-trapped animals and one sight record, with distinctive reddish pelage (rather than the darker, less red, and more contrasting pelage assumed to be shown by M. truongsonensis; but note that the pelage of the nominal taxon M. puhoatensis is unknown), from a number of localities in the Northern Annamites of both Lao PDR and Viet Nam (R.J. Timmins pers. comm. 2008).
Ma et al. (1986) listed M. rooseveltorum for China, but the identification was revised to M. reevesi by Groves and Grubb (1990) who apparently examined four purchased skins of this species in the Kunming Institute of Zoology (thus extending this species range significantly to the west). There is no description of the skins in the above sources, and given the current understanding of muntjac systematics it would likely be hard to diagnose the specific identity of the skins (especially if heads and tails were incomplete or missing) without molecular analysis, their species identity is thus probably best treated as indeterminate at present.
The type localities for available names in this species-complex are:
1. M. rooseveltorum: Ban Muangyo, Phongsaly Province, Lao PDR, ca. 21°31'N, 101°51'E (Osgood 1932).
2. M. truongsonensis: Viet Nam, "in the west of Quang Nam province"; "collected from four houses in three locations in Hien District, West Quang Nam Province, Viet Nam. The three locations are: Hien, the district capital, A Tin village, and A Plo village (15°56'59''N, 107°34'18''E)" (P. M. Giao et al. 1998). Although this information is not given in the original description (Nguyen An Quang Ha 1997) it can reasonably be used to restrict the more general indications in the original description because there is no doubt that the two sources refer to the same collected material.
3. M. puhoatensis: the Pu Hoat area in Que Phong District, Nghe An Province, Viet Nam (Binh Chau 1997).
4. M. putaoensis: “purchased … at Atanga village, 30 km east of Putao (27°21'N, 97°24'E), northern Myanmar” (Amato et al. 1999b).
Native:Lao People's Democratic Republic
|Range Map:||Click here to open the map viewer and explore range.|
|Population:||Over 50 specimens (not all now extant) and close to 100 camera-trap photographs of the species-complex are known to R.J. Timmins from Indochina (pers. comm. 2008), suggesting it is relatively common in both Lao PDR and especially Viet Nam. Camera-trapping in northern Myanmar suggests that the species-complex is also at least locally common there (R.J. Timmins pers. comm. 2008, based on WCS unpublished data). Results from camera-trapping in Lao PDR (Nan Et–Phou Louey NPA: R.J. Timmins pers. comm. 2008, based on unpublished data A. Johnson/ WCS; Nakai-Nam Theun NPA: R.J. Timmins pers. comm. 2008, based on Watershed Management Protection Agency unpublished data; Robichaud et al. in prep.) and Viet Nam (Hoang Lien Son mountain range: R.J. Timmins pers. comm. 2008, based on unpublished data S. Swann/FFI) show that the species-complex is at least locally common at these sites, and it could be so over much of its range. In other areas, mostly south of the previous list of areas, such as Pu Mat NR; Thua-Thien Hue province; and Quang Nam province (Viet Nam), and Bolikhamsai province (Lao PDR), the species-complex has been camera-trapped less frequently than M. vaginalis in each area, but this may reflect ecological factors of sampling areas, such as habitat and altitude range (R.J. Timmins pers. comm. 2008, based on various unpublished datasets; SFNC 2000; Robichaud and Stuart 1999).|
|Habitat and Ecology:||
In Lao PDR this species-complex is probably restricted to higher elevations (over 1,000 m asl) and forests at the evergreen end of the spectrum; it may be absent from forest even above 1,000 m asl in drier western areas. In Viet Nam it occurs also down to lower altitudes (perhaps to the foothills) especially on the eastern slope of the Annamites and probably also in northeastern areas (east of the Red River) (R.J. Timmins pers. comm. 2006). In Myanmar the species-complex appears to occur between at least 700 and 1,250 m asl (see M. putaoensis 2008 account). Social ecology is likely to be similar to other species of muntjac, which are generally solitary and territorial.
The complex is widely sympatric with northern red muntjac M. vaginalis, although at high elevations over about 1,500 m asl that species appears to become scarce or naturally absent within Indochina, whereas the rooseveltorum complex ascends to over 2,000 m asl in at least northern Viet Nam (R.J. Timmins pers. comm. 2008, based on photographs and other unpublished data of S. Swann/FFI).
The main threat to the species-complex is hunting. Especially within Lao PDR and Viet Nam, there is indiscriminate snaring throughout its known range, for local meat consumption and a thriving bushmeat trade (see Timmins et al. 2007). However, the best protection from hunting at present is the ‘cushioning’ effect of large forest blocks, where the costs of extracting wildlife of low to medium commercial value from the centre are higher than the income from selling it, and where habitat is so extensive it is simply more difficult to hunt out cryptic, non-gregarious species, than it is in smaller forest blocks. Because most such blocks within the complex’s geographical range are centred on rugged terrain, with higher elevations forming the core of such areas, these muntjacs are likely to be significantly less threatened than are sympatric ground-dwelling mammals with altitudinal ranges limited to lower elevations.
In localized areas, especially of northern Indochina, the species-complex is likely to be affected by habitat loss. Here, most ethnic groups practice shifting cultivation and in recent decades there has been a clear net loss and fragmentation of forest: such shifting cultivation is not now in a balanced cycle (for rotational systems), or never was (frontier systems). The adaptability of the species-complex to degraded or fragmented habitats is not well known, although evidence from northern Lao PDR suggests that animals there persist in landscape-scale mosaics including extensive secondary, degraded forest fragments and cultivation (R.J. Timmins pers. comm. 2006). In general it appears that muntjacs (particularly M. vaginalis and M. reevesi) adapt well to secondary and degraded habitats and in fact may reach higher densities in such areas than they do in unencroached habitats. More serious than losses to shifting cultivation, in the most recent years and the immediate future, is likely to be the large-scale replacement of natural forests (albeit mostly heavily degraded) with rubber plantations to supply the burgeoning Chinese market. This is affecting large areas of the northern highlands of Lao PDR and elsewhere, and will have three linked negative effects on ground-dwelling mammals: direct loss of habitat (it is unlikely that rubber plantations will support even mediocre densities of this species-complex, although this is not yet known), decline and destabilisation of the forest resource-base (including wild-meat sources) for subsistence-level communities, leading to higher hunting levels in remaining wildlife areas (C. Wood pers. comm. 2005 to J.W. Duckworth, based on emerging patterns across several provinces in Lao PDR’s northern highlands); and shrinkage of the size of natural habitat blocks and thereby reduction and, locally, loss, of the cushioning effect.
Taxonomic uncertainties within this species-complex forestall any analysis of conservation needs for the individual species. Populations of this species-complex occur in probably most protected areas of the Annamite range and northern highlands of Lao PDR and Viet Nam (R.J. Timmins pers. comm. 2006). The species-complex also occurs in several protected areas in Myanmar (Schaller and Rabinowitz 2004). If there are localised taxa within this species-complex then conservation measures additional to support and consolidation of the existing protected area system may be warranted.
Whatever actual taxonomic pattern is found within the species-complex, management of hunting will need to recognise that hunters cannot be expected to recognise muntjacs to species in the field: some methods are non-selective (e.g. snares) while active projectile hunting (e.g. guns, bows, blow-pipes) would be severely reduced in their harvest efficiency if identification had to precede shooting. The only realistic site-based management tool for this species-complex is likely to be the declaration of suitably-sized ‘core areas’, site in ecologically appropriate areas, where all hunting using methods suitable for muntjacs, including under hitherto traditional rights, is prevented. Lao protected area legislation apparently provides for ‘core areas’ within each national protected area although few if any have yet been declared. Each protected area within the species-complex’s range should therefore have at least one large (several hundred km², in the absence of any empirical knowledge of muntjac spatial ecology) core zone declared and enforced. This measure implemented in conjunction with well considered hunting management in peripheral forest areas might also benefit the ongoing dietary needs of local people, who could harvest in a controlled manner animals outside core zones, and who, in parts of montane Lao PDR that have been effectively studied, are being forced into involuntary veganism because the lack of effective wildlife management and sensible hunting quotas, has meant wildlife stocks have fallen severely (Krahn and Johnson 2007).
|Citation:||Timmins, R.J., Duckworth, J.W. & Long, B. 2008. Muntiacus rooseveltorum. The IUCN Red List of Threatened Species. Version 2014.2. <www.iucnredlist.org>. Downloaded on 25 July 2014.|