Table 4 Species names proposed for each species recognized as putative ones by the method of Pons et al., [28].

R1

Rattus rattus

R1 specimens identified in [36] cluster unambiguously with R. rattus specimens identified by Robins et al. [25] (see Figure 4). It is worth noting that, during this study, this species was never sampled in the fields in Thailand, Laos and Cambodia.

R2

Rattus tanezumi

R2 specimens cluster unambiguously with R. tanezumi specimens identified in [25] (see Figure 4).

Medium-sized rat; fur light brown to reddish brown above, white below; dark tail, equal or longer than head and body length; caught in a large range of habitats, from houses, gardens, crops and rice fields to the edge of secondary forests.

R3

Rattus sp.(to be named)

R3 includes specimens identified as Rattus diardii in the study of Robins et al., [25] and rats referred to Malaysian house rat (i.e. Rattus diardii) by local populations in Indonesia (Andru, J., pers. comm.). Today, Rattus diardii has been placed as a synonym of Rattus tanezumi according to morphological criteria.

Urban rat or rat living near human habitations. Misidentified by us as Rattus tanezumi, R argentiventer and R. andamanensis in the Rattus rattus species group.

R4

Rattus losea or "losea-like"

Medium-sized rat; shaggy fur brownish grey above, white to geyish below; dark tail, shorter than head and body length; caught mostly in rice fields and sometimes in dry agricultural fields. According to Aplin [35] two distinct forms of R. losea may exist. True R. losea (described from Taiwan) would be distributed from Southern China to central Vietnam. The second form "losea-like" would inhabit the Mekong Delta region from Southern Vietnam, Cambodia, Thailand, to the North of Vientiane Province in Laos. Since our analyses did not include samples from the two putative groups, it was not possible to determine if they are genetically distinct. Until this taxonomic issue is resolved, we prefer to name R4 "losea-like".

R5

Rattus tiomanicus

R5 specimens cluster unambiguously with R. tiomanicus specimens identified in [25] (see Figure 4).

Medium-sized rat; fur brown above, white below; dark tail, slightly longer than head and body length; arboreal; caught in palm plantations. Morphologically very similar to Rattus tanezumi but with shorter guard hairs.

R6

Rattus argentiventer

R6 sequences cluster unambiguously with R. argentiventer sequences provided by O. Verneau and F. Catzeflis (see Figure 4)/identification of Verneau's specimen confirmed by G. Musser [64].

Medium-sized rat; fur yellowish brown above, grey-white below, with developed guard hair on the back, distinct orange fringe of fur just forward of the ear; dark tail, shorter than head and body length; caught in rice fields and plantations.

R7

Rattus andamanensis

R7 sequences cluster unambiguously with R. sikkimensis sequences provided by O. Verneau and F. Catzeflis (see Figure 4).

Medium-sized rat; fur orange brown above, white-creamy below, with very elongated guard hairs; dark tail, longer than head and body length; caught in evergreen forests.

R8

Rattus exulans

R8 specimens cluster unambiguously with R. exulans specimens identified in [25] (see Figure 4).

Small-sized rat; fur grey-brown above, pale grey below; dark tail, longer than head and body length; domestic species found in houses.

R9

Rattus norvegicus

R9 specimens cluster unambiguously with R. norvegicus specimens identified in [25] (see Figure 4).

Large-sized rat; fur dark-grey above, pale grey below; tail shorter than head and body length, dark above and paler beneath but not clearly separated; occurs in major ports and neighbouring cities.

R10

Rattus nitidus

Sister relationship with Rattus norvegicus evidenced by molecular data (see Figure 2) .

Medium-size rat with a soft woolly fur, dorsally brown and grey-based cream on belly. Pearly white feet. A nitidus/norvegicus sistership was proposed by morphologists. According to Musser and Carleton [11], both have "dense and soft fur, six pairs of teats, and an upper M1 in which the anterolabial cusp on the anterior lamina is missing or undetectable due to its coalesence with the adjacent central cusp".

B1

Bandicota indica

Only two Bandicota species have been described in the Indochinese region. Usually, B. indica specimens are unambiguously larger than B. savilei. Adult B. savilei and juvenile or immature B. indica may be confounded. A molecular test based on PCR amplifications with specific primers allowing discriminating between the 2 species (Chaval et al., in prep.) was used in such cases (data not shown).

Large-sized rat; fur dark above, grey below; tail shorter than head and body; aggressive and stocky; inhabits agricultural fields. The ratio of pes length to head+body length is used to distinguish B.indica from B.savilei [74].

B2

Bandicota savilei

Medium-sized rat; fur dark above, grey below; tail shorter than head and body; inhabits dry lands, grasslands, clearings in forest.

Be1

Berylmys berdmorei

Medium-sized rat; fur grey above, white below; tail shorter than head and body; inhabits secondary forests and fields close to forests.

Be2a

Berylmys bowersi

Large-sized rat; fur grey above, white below; tail slightly longer than head and body; inhabits secondary forests and fields close to forests.

Be2b

Berylmys sp.

Berylmys mackensiei has been described in the Indochinese region by Marshall [33]. However the skull of B. mackenziei he studied was identified by Musser and Newcomb [75] as B. bowersi. Populations of Berylmys bowersi in peninsular Thailand were reported to be geographically isolated and to differ in some ways from those elsewhere (here speculated as to be Be2,a) [67]. Be2b specimen came from the Kanchanaburi locality, North to the isthmus of Kra and could consequently belong to this former particular population. Because of the lack of additional information about this specimen, no species name could be convincingly assigned to Be2b.

L1

Leopoldamys edwardsi

Large-sized rat; fur red-brown above, white-cream below; very long tail, longer than head and body; inhabits secondary forests.

L2

Leopoldamys neilli

Genuine sequence obtained from the holotype specimen of L. neilli was assigned to L2 without ambiguities.

Large-sized rat (but the smallest Leopoldamys species); fur greyish -brown above, white-cream below; tail longer than head and body. Until now, the species has been recorded from a few locations in limestone areas of northern and South western Thailand, North of the peninsular region [76]. Our specimens were also trapped on tower karst in northern and northeastern Thailand (Phrae and Loei provinces).

L3

Leopoldamys sabanus

Large-sized rat; fur red-brown above, white-cream below; very long tail, longer than head and body; inhabits secondary forests. Caught in secondary forests. Often misidentified as Leopoldamys edwardsi. The two species of Leopoldamys sabanus and Leopoldamys edwardsi are indeed morphologically very similar. The species name we proposed for L3 is based on geographical evidences from Marshall (1977). Based on his work, the only Leopoldamys species that has been described in Kanchanaburi province is Leopoldamys sabanus. The L3 specimens were caught in this province.

N1

Niviventer fulvescens

Medium-sized rat; spiny fur red-brown above, white-cream below; tail longer than head and body, sharply bicoloured from base to tip; absence of terminal pencil and smallest length of bulla make us exclude Niviventer confucianus as species name.

N2

Niviventer sp. 1

Marshall [67], Musser [77] and Corbet [34] documented the occurrence of Niviventer bukit in Kanchanaburi, where representatives of N2 and N3 species were caught. One of the two could be N. bukit. However, bukit is today considered as conspecific with Niviventer fulvescens [16]. Consequently, we prefer to refrain from giving a species name to these 2 species.

N3

Niviventer sp. 2

N4

Niviventer langbianis or Chiromyscus chiropus

N4 is placed at the base of the Niviventer group. It could thus belong to the genus Niviventer or to a sister genus to Niviventer. According to Musser and Carleton [16], Chiromyscus is presumed to be one of the closest phylogenetic relatives of Niviventer. Based on morphological criteria, this specimen could be a Chiromyscus chiropus representative. However, Chiromyscus chiropus is morphologically very closed to N. langbianis. Thus, N4 could be one of these two species. At the end of this work, we have just received N. langbianis samples from the AMCC. Our preliminary work based on mitochondrial DNA suggests that N4 may be N. langbianis rather than C. chiropus.

Identified in the field as Nu-deng because of its reddish fur (in Lao, "red rat"). Further considerations of pictures of one of the two specimens included in this study show that legs, feet and head are buffy orange as described by Musser [77] regarding Chiromyscus chiropus. However, the wide dark brown rings around the eyes are not visible and the tail is not bicoloured as expected for Chiromyscus. Chiromyscus is morphologically very close to Niviventer langbianis [77] and easily confused with it. Other criteria to discriminate between the two species such as the presence of a nail on each hallux instead of a claw for Chiromyscus are not obvious on our pictures. Morphological identification is thus questionable. However, molecular data are tipping the balance for N. langbianis assignation.

M1

Maxomys sp.

Identified by us as Maxomys surifer in the field. Could be assigned to Maxomys rajah but this species has never been reported in this area. This result could be to a bias of the branching-length method that could have some difficulties to deal with strong phylogeographic pattern. The phylogeography of Maxomys surifer was investigated using mt DNA but focusing on the large Sunda shelf area [78]. A structuration between the North-eastern Vietnam and the Southern Vietnam seems to exist but this finding is based on only four sampled (for which sequences are not available in databanks). As a greater sampling and more additional data are needed to assess the phylogeographic pattern of this species, we prefer to refrain from giving a species name to this cluster.

M2

Maxomys surifer.

Medium-sized rat; spiny fur red-brown above, white-cream below; tail slightly longer but nearly equal to head and body length, sharply bicoloured with a white tip. This is the only Maxomys species described in this area