4 Records of the Indiati Museum. [Vol. Ill, 



to the validity of such an acknowledgment, and the probability 

 that similar variations may arise in two separate localities appears 

 very great. 



In illustration of this the following example may be given : — 

 The mole-rat Gunomys bengalensis was found to be common 

 in Rangoon, occurring not only in granaries, warehouses and 

 stables, but frequently in the rooms of dwelling-houses situated 

 in the heart of the city. This rat, usually of a greyish brown tint, 

 does not as a rule show much colour- variation. However, within the 

 narrow limits of two adjacent houses, in a certain street in Rangoon, 

 a number of pure black specimens were found. (This occurrence 

 will be described in detail further on.) These may be regarded 

 as a new variety which may or may not persist ; but there seems 

 to be no reason why a similar localized race should not appear in 

 any other town in India. Indeed some evidence will be brought 

 forward to show that melanotic mole-rats have arisen in two 

 separate parts of Rangoon city itself, independently of one 

 another. 



Further, it can be shown that such peculiarities of colour- 

 variation are by no means confined to the phenomenon of 

 melanism. The following example may be quoted : — A race of 

 white-bellied rats of the Mus rattus type was met with in Rangoon. 

 The tails of these rats were uniformly dark greyish brown, being 

 of exactly the same tint above and below. Among some hundreds 

 of these rats which were examined, one was obtained which 

 showed a pure white line along the lower surface of its tail, — in 

 other words, the tail was obviously bicoloured. It appears that 

 great stress is laid on this feature in defining species of rats ; while 

 it is admitted that within the limits of one species the lower sur- 

 face of the body may be dark or pure white, the same admission is 

 never extended to the tail. The following difficulty, therefore, 

 presented itself : — Should this single specimen with the bicoloured 

 tail be regarded as a separate species from the others which, 

 except in the one respect, it so closely resembled ? The probability 

 that it was merely a sport from the common type appears too 

 great for this step to be taken with any confidence. If now, it 

 is granted that this particular specimen may have arisen as a 

 sport, there appears to be no reason why, being prepotent, it 

 might not have established a family with bicoloured tails within 

 the limits of a few adjacent houses, as in the case of the melanotic 

 Gunomys. Such a family would probably die out in a few months, 

 but on the other hand, the strain might become firmly established 

 as a new species. 



These remarks have been made in order to explain the paucity 

 of species and varieties recognised, and to defend the practice 

 observed of not defining new species from single individuals or even 

 from small local groups of individuals, which may show well- 

 marked peculiarities of coloration. The literature of Oriental 

 rats frequently reveals species, defined from three or four indivi- 

 duals caught within the narrow limits of a single house, all of which 



