THE BLACK OR SHIP RAT 593 
the group in three sub-groups or varieties; he used the name a/exan- 
drinus for all the Indian forms of vatéws, and stated that the “ typical 
form,” characterised by its large size, long tail, and coloration (dark 
rufous grey above, white below), inhabited Kashmir and the whole 
north-western region of India. 
A somewhat similar arrangement was later adopted by Bonhote 
(Fascic. Malay., Zool., i., 32, 1903; and Proc. Zool. Soc., London, 1910, 
653), who recognised three sub-groups of Oriental vattws as follows :— 
(1) The jalorensis sub-group, of which 4. jalorensis, Bonhote, 
from Jalor, Perak, and Siam, is the type. These are hill-rats repre- 
senting the zzttdus group of Thomas and other authors, and they 
have the hairs of the under parts white to their bases, and a hind foot 
length of about 30 mm. The sub-group has a wide but discontinuous 
distribution in India and Malaya. 
(2) The rufescens sub-group, of which &. rufescens, Gray, from 
Dharwar, is the type. These are tree-rats, and have the hairs of the 
under parts with white or yellowish tips, and slate-coloured bases, 
and a hind foot length of about 33 mm. This sub-group is found 
throughout the whole of continental India, except the north-western 
part, and in Ceylon. In some provinces and in Ceylon a variety with a 
pure white belly is found side by side with typical rufescens, but in 
Burma and Tenasserim the white-bellied form alone occurs (Wroughton, 
J. Bombay N. H. Soc. 23, 474 and 715). From Simla, Bonhote has 
described 4. vzcerex, a member of the group with striking bicoloration 
of the tail and white feet. 
(3) The griseiventer sub-group, of which Z. grzsezventer, Bonhote, 
from Perak, is the type. These are chiefly house-rats, and in them the 
ventral hairs are either entirely slate-coloured or else have fulvous tips, 
and the hind foot measures about 35 mm. This sub-group has a wide 
distribution throughout India, Ceylon, and the Malay peninsula and 
islands. 
Hossack (Mem. Indian Museum, i., 1-80, 1907) studied the rats of 
Calcutta in connection with plague; his material led him to regard 
the distinctions between the races of vattus recognised by Bonhote as 
quite inconstant and sporadic, and he therefore denied that such 
characters have any systematic importance. Bonhote (Proc. Zool. Soc., 
London, 1910, 653), in answer to Hossack, states that in the large 
towns, whence all Hossack’s material came, no order or classification 
is possible, because the varieties have become hopelessly mixed and 
crossed, but that in the country districts the varieties are much better 
defined, and apparently breed true; “all these varieties of the long- 
tailed rat belong to one species, Mus rattus.’ Meanwhile Lloyd 
(Rec. Indian Museum, iii., 1-100, 1909) studied a very large number 
of rats from India and Burma; his work was a continuation of that 
