igog.] R. E. Lloyd : The Races of Indian Rats. 85 



they even confine themselves to particular houses or groups of 

 houses, and are much given to breeding within the family circle. 



To obtain sure evidence about such a question is a matter 

 of some difficult}'. It might be obtained experimental!}' by means 

 of marked rats. Fortunately cases have occurred in which Nature 

 has herself provided the mark. Captain Davys has contributed 

 a good example from the Amritsar district, and pointed out its 

 significance. Among a mixed collection of Mus rattus from any 

 large Indian town there is often a small number of white-bellied 

 ones. Among the large number of rats which we received from 

 the Punjab was a small sprinkling of such forms, and it was de- 

 finitely ascertained by the sender that out of sixty-nine villages, 

 in which rats had been collected, only three provided these white- 

 bellied forms, and that they made up about ten per cent, of the 

 total rats of those villages. Whether we regard the white-bellied 

 forms as a separate race or not, there is clearly not much inter- 

 course between the rats of those villages. 



Two other cases have been already described whicl^ show con- 

 clusively that a group of rats may identify itself with certain ad- 

 joining houses. The first of these is the case in which ten black 

 mole-rats were caught in two adjoining houses in Rangoon during 

 three nights, and no other kind of rat was caught at the time in 

 those houses, although black mole-rats are very rare indeed. This 

 shows that a " family group " ma}' establish itself within very 

 narrow limits, keeping without those limits all others who are not 

 of the group, A precisely similar case was recorded from Naini Tal 

 where the house rats found on a part of Ayapata Hill could always be 

 distinguished from those which dwelt at either end of the lake, 

 although the three places were separated by less than a mile. It 

 may be objected that these examples do not constitute fair evidence 

 in favour of the view that the rats of one species in a town do not 

 freely intermingle, because in these cases the rats have been marked 

 off by their peculiarities from the majority, and held themselves 

 aloof. 



While watching large numbers of rats brought in by towns- 

 folk, the observer is compelled to recognise that the rats of any 

 one species are split up into a great number of ' ' family groups ' ' 

 or clans, each with its own limited domain. If, for example, at a 

 collecting station rats of the long-tailed kind are superficially ex- 

 amined and set aside, the accumulation, which consists perhaps of 

 some hundreds of individuals, has a truly heterogeneous appear- 

 ance ; some few of its constituents are black or nearly so, a few are 

 pure white below, others are white below with a gre}' breast stripe. 

 Among them there is much variation in size. The length of one 

 may be 150 mm., the length of another 200 mm., both being ob- 

 viously adult. Tail length will vary from 105. % to 135 % of length. 

 In respect to length or tail length, the rats can be laid out side by 

 side in an unbroken series, the mediocre in both respects being 

 in the majority. In respect to colour it is much more difficult to 

 arrange them in an unbroken series. The collection will perhaps 



