1907.] Dr. Hossack: The Rats of Calcutta. 3 



not only of specific but even of subgeneric and generic value , but after what Thomas 

 has to say on the matter, it seems pretty clear that it is simply, in the main part, a 

 seasonal variation, spines taking the place of hairs to a greater or less extent in all 

 tropical rats in the hot season. With reference to the relations of Mus alexandrinu's; 

 Mus rufescens, and Mus nitidus, he arrives at the definite conclusion that they are 

 three intergrading varieties of the same rat, alexandrinus typically found in Kashmir 

 and North- West India, rufescens in Southern India, nitidus in the Nepalese District, 

 while the skuUs of all three show no difference except in size. With regard to the 

 relationship of this little group of varieties to the European Mus rattus he is by no 

 means so clear, but he tends to the opinion that they will be found eventually to belong 

 to the same species. 



We now come to the third epoch, one characterized unfortunately by more or 

 less the same state of confusion as prevailed during the first. Since Thomas did his 

 work the boundaries of India have been enlarged to include Burma, the rats of which 

 are linked up with those of Borneo and the Malay Archipelago, Formosa, Java and 

 Celebes. From all the countries named a vast number of new species have been des- 

 cribed, the specific differences between which are often minute, so minute in fact that 

 no description can convey them, and comparison with existing types is absolutely 

 necessary. 



Some observers seem to have gone on the principle that every island should have 

 its own named rat, a point on which I shall say more, when considering lOcal variations. 

 The result is that in a recent conspectus of some of the Oriental Rats, Mr. Lewis Bonhote 

 has collected 8 groups, 10 subgroups and 95 species.^ The distinctions drawn between 

 the groups and subgroups are sufficiently unsatisfactory ; but to find included such 

 names as M. asiaticus, M. decumanoides , M. sladeni, M. caudatior, etc., is more than 

 disappointing. They were extinguished by Mr. Thomas more than twenty years ago, 

 on the ground of identity with other species of incomplete description, or, in the most 

 extreme instances, of complete absence of description. No less than fifteen such ex- 

 tinguished varieties have to be marked off Bonhote' s list ; five of them extinguished 

 not only by Thomas, but also by Sclater after examination of the types in the 

 Indian Museum ; they were made mere synonyms of M. rufescens. It is impossible 

 to pass over the paper of Mr. Bonhote without calling attention to certain slips 

 which are liable to cause needless difficulties. Thus he makes M. bukit similar to 

 M. cremor IV enter but distinguished by being larger, being i2-i cm. in length of head 

 and body. But M. cr emoriv enter , according to Miller, is 14-6 cm. in length of head 

 and body. 



M. jalorensis he distinguishes from M. rufescens by its colour and its short tail ; 

 but in the contrasted measurements he records, M. jalorensis has a tail 113 per cent, of 

 its body-length (145 cm.), while M. rufescens has a tail only 108 per cent, of its body- 

 length (17 cm.). Though the latter is defined as a short-tailed rat, the type has a tail 

 of 122 per cent, of its body-length, measuring 14-4 cm. 



' In Annandale and Robinson, Fasciculi Malayensis , Zoology, part i, Oct. 1903. 



