58 THE GROWTH OF GROUPS 



All of them possessed the same peculiarities. No other 

 kind of rat entered the traps set in those particular houses 

 during the time. There can be no doubt that the ten 

 rats represented only a part of the total number which 

 infested those houses. It is scarcely possible to clear a 

 building by trapping. In any house several rats may be 

 caught on the first night and perhaps one or two subse- 

 quently, but after such losses their comrades become wary 

 and avoid the traps, at least for a time. In all proba- 

 bility there were as many as thirty rats in those two 

 houses. The fact that the ten were remarkably alike, 

 each possessing definite peculiarities by which it could 

 be recognized at a glance from the other rats of the town 

 makes it almost certain that if the whole group had been 

 captured and examined all would have been found alike. 

 The subsequent capture some months later of two more of 

 these peculiar animals, both showing the same special 

 characters, confirmed this opinion. 



Although it may seem unfair to assume that this 

 family group contained as many as thirty members 

 judging solely from the fact that ten of them were 

 captured, yet those who know the extent to which houses 

 in tropical cities are infested with rodents will consider 

 that the assumption is moderate. The particular houses 

 in question were of a kind very favourable for rats. 

 The partition walls were composed of double layers of 

 wooden planking separated by a space two or three inches 

 wide. The floors were of broad tiles laid upon the earth. 

 Circumstances of this kind are very suitable for rats, 

 especially for those of a species which is well known for its 

 burrowing powers. 



There is supporting evidence in the fact that those 



