130 THE VARIATION OF ANIMALS IN NATURE 



indeed states that, if in the course of divergence a point is 

 reached after which it is impossible for the diverging form to 

 coalesce with the parent stock, we are given by this point a 

 definite means of distinguishing varieties from species. 



The changes in animals themselves which make inter- 

 breeding actually impossible form the second or permanent 

 type of isolation. Permanent isolation may be the result of a 

 variety of factors, and an important consideration is to determine 

 whether it can ever be developed in the absence of some degree 

 of geographical separation. The establishment of geographical 

 isolation might often be due to geological changes within the 

 area of a widely ranging species, but we must also recognise 

 the importance of the wanderings of the animals themselves. 

 The continual invasion of all countries and habitats, however 

 apparently uncongenial, is a commonplace of natural history. 

 Where the invaders have to overcome great difficulties, we 

 usually find the formation of isolated colonies, as in oceanic 

 islands. 



Permanent isolation may arise frequently from ' accidental ' 

 changes in the structure and habits of populations no longer 

 in a position to eliminate or assimilate the variant individuals 

 by free intermixture. The actual mechanism which prevents 

 allied species from interbreeding is rarely understood in detail, 

 but very often there seems to be a great difficulty in explaining 

 how the mechanism can have been perfected, since the charac- 

 ters on which it depends appear to be of little use to in- 

 dividuals or even to the species as a whole. 



Although we now suspect that some measure of permanent 

 isolation may be developed amongst individuals inhabiting 

 a continuous area, yet it is probable that geographical isolation 

 is more often than not a necessary preliminary. The temporary 

 nature of the latter type of isolation makes it important for 

 us to examine the rate at which topographically isolated 

 populations diverge from one another. It may be admitted 

 that the degree of permanent isolation is only very roughly 

 correlated with that of the resulting morphological divergence, 

 but in so far as the latter is likely sooner or later to entail 

 permanent isolation, the rate of divergence under geographical 

 separation becomes relevant. We shall therefore digress to 

 consider the available evidence as to the time necessary for 

 the establishment of a new species or subspecies. 



