TAXONOMIC STUDY ON THE LARGE INDIAN CIVET , VIVERRA ZIBETHA ( LINNAEUS ) FROM THE INDO-BURMESE SUBREGION ( MAMMALIA VIVERRIDAE )

The colour, textur~. pattern and length of dorsal fur of the body are some of the characters on which a number of subspecies of the Large Indian Civet, Viverra zibetha (Linnaeus) are differentiated (Pocock. 1939). While cataloguing the collections of Viverridae present in the Zoological Survey of India, the authors had an oppor­ tunity of examining a good series of skins of V. zibetha from the Indo-Burmese sub­ region, and faced difficulty i~ the subspecific identification on the basis of the characters provided by Wroughton (1915, 1918), Thomas (1927) and Pocock ( 1933, 1939). Hence, it was felt 'necessary to study this species afresh, based on the material present in the Zoological Survey of India~ the Bombay Natural History Society and the data available in the extant literature. The results of this study are presented in this paper.

dorsum ranges fronl pale grey, Smoke Grey, dark grey, Raw Umber, Tawny Olive to Mars Brown, and grizzled with white, buff or yellow, particularly on the sides.Hairs on chest and belly ranging from Mars Brown to Raw Umber on base and white at tip.Lower half of limbs Seal Brown.It was noted that the colour varies irrespective of locality or season.
Body markings: Body markings indistinct and cloudy, except on the upper half of limbs where small stripes may be present.Pattern on the head and shoulders distinct in majority of specimens (Table 2).In six out of 15 specimens from the distributional range of the nominate subspecies, pattern on shoulders feebly developed.Whitish patch on either side of the front of muzzle, and greyish patch behind it, show various degrees of development, irrespective of locality and/or season.A black spinal stripe runs from behind the shoulders to root of the tail, which in majority of specimens flanked on either side by a narrow but distinct white stripe from posterior half of the back.However, no seasonal or geographic distinctness could be noticed in this character.
Pattern on flanks highly variable, distinct in 50-67% of the summer-specimens and 0-40% of the winter ones of the different localities (Table 2).
Tail encircled with white and black rings, the latter much broader than the former; white rings vary from 5 to 9, usually 6 in nunlber; first one being incomplete dorsally in all the -specimens; second, third, fourth and fifth being incomplete in two, one, two and one specimens respectively.
Skull-structure: Author's observations agree with the descriptio~ given by Po~otk (1939) for the genus Viverra, and found more or less similar' for all three subspecies of V zibetha.However, it was noted that the posterior margin of palate varies individually, straight or semi-circu lar, and smooth, serr.ated.or medially inUndated.Similarly, structure of auditory region varies from sp,ecirnen to specimen.In one specimen each from Pyaunggaung and Hkamti (N.Burma), Gohlghat (Assam) and CharHam (Tripura), zygomatic arch was found to be poorly developed and narrow (zygomatic width less than 46% of greatest length of skull).
In other specimens Zygomatic arch is well-developed {48%-53% of GL), irresp~.ctive of geogra phical distribution (Table 3).DISCUSSION Linnaeus (1758) described the species Viverra zibetha from Bengal (probably present northern West Bengal), without mentioning the exact locality.Later, Wrotl• ghton (1915) described two subspecies, namely V. z. picta from upper Chindwin, northern Burma, on the basis of distinct markings on the body, and V. z. pruinosa from Tenasserim, southern Burma, by the' absence' of yellow' tinge in the ground colour of the body and presence of white-tipped •hairs.Three years later, based on the study of additional material, Wroughton (1918) remarked that the distinctness of the pattern is not a racial character.Hence, lfe withdrew the name pieta, and relegated it to the synonymy of V z. zibetha.Pocock (19~3) on e~amination of a number of skins collected during winter noticed that the pattern on shoulder is more distinct :in specimens from Assam, Meghalaya and N agaland than those from Sikkim and n~r thern West Bengal, and the resurrected the subspecies ••picta.He. als~o maintained pruinosa on the basis of the fur being short, and pattern on the body distinct in winter specimens.
The present study of material from the' geographical ranges of zibetha, piela and pruinosa reveals that the differences in coat-colour and the pattern on sho~'ders and flanks in all the three populations vary from specimen to specimen (Table 2), hence appear to be of no systematic value.As regards the length•, of fur, .it is definitely shorter in two winter specimens from southern Burma than in specimens from Nepal, Sikkim and northern West Bengal, but are well within ; the range of length of fur from other localities (Table 1) ...Moreover, the sample-size from southern Burma is too small to come to any conclusion as to the shortness of the fur.Lekagul and McNeely (1977) although maintained the subspecies piela and pruinosa on the basis of material present in Thailand, but regarded them as poorly differentiated.An analysis of the various measurements of the body and skull from different subspecific zones of the Indo-Burmese subregion reveals that there is no significant difference between them in any of the body and skullineasurements (Table 3).The structure of the posterior margin of palate, auditory region and the zygomatic arch though differ to a certain extent, but are of no systematic importance because they vary from specimen to specimen.
Hence, in authors' opinion, the separation of picla and pruinosa from zibetha is taxonomically not justified.It would be appropriate to treat Viverra zibetha picta and Viverra zibetha pruinosa as synonyms of the nominate subspecies.Since' the authors could not examine the specimens of another subspecies, V z. ashtoni, from southern China due to non-availability of the material, they are not in a position to comment on the status of that subspecies.However, according to Pocock (1939, footnote on page 351) the southern Chinese ra ce has longer coat.A winter specimen has crest and flank-hairs about 90 and 50 mm long respectively.

Table 1 .
-Length (in mm) of hairs on crest and flank in three populations of Viverra zibetha Linnaeus (NE India and S. West Bengal) V. z. pruinosus Dorsal hairs relatively thicker and coarser than ventral hair~: length of dorsal hairs highly variable, generally longer in winter specimens than in summer ones (Ta ble 1); seasonal variation less evident in specimens from Nepal, Sikkim and West Bengal (Darjiling district).In winter specimens, hairs of crest from Tenasserim (S.Burma) relatively smaller than those from Nepal, Sikkim and northern West Bengal but fall within the range of length of crest-hairs from southern West Bengal, north-eastern India and northern Burma, thus forming a cline.

Table 3 .
-Measurements of body and skull (in mm) in different populations of Viverra zibetha (Linnaeus) with range, mean and standard deviatian.Sample size in parenthesis.