THE GROWTH OF GROUPS 29 



their descendants have usually flourished. When we 

 consider how a lone island became inhabited with animals 

 it is enough to ask how a pair or two of the various kinds 

 arrived there on the first occasion, we do not suppose 

 that successive reimportations are necessary in order to 

 counteract the evil effect of inbreeding. 



The special question which will be raised in this book 

 is how far inbreeding is detrimental to rodents. Fortu- 

 nately this question has been answered by experiment. 

 I am not acquainted with the original description of the 

 experiments, but they have been summarized in the 

 following words : 



" Ritzema-Boz inbred rats for thirty generations ; 

 for the first four years (twenty generations) there was 

 almost no reduction of fertility, increase of mortality, 

 and decrease of size." In these experiments of course all 

 the descendants of the original pair were not preserved, 

 they were far too many. But since inbreeding for twenty 

 generations is harmless, we know that it is possible 

 for a pair of rats to give rise to many millions of healthy 

 and apparently fertile descendants. 



A group, such as this, would occupy a considerable 

 space in nature. There is no reason to believe that the 

 group, having reached such dimensions, would inevitably 

 deteriorate. In spite of artificial conditions in the experi- 

 ment, the descendants of the pair reached the twentieth 

 generation before deterioration commenced. This proves 

 that a pair in nature is able to establish a large group, 

 but the fact that deterioration eventually occurred does 

 not prove that such a group would deteriorate in nature. 

 Indeed, the history of rabbits in Australia and other 

 similar cases afford evidence to the contrary. 



