274 THE VARIATION OF ANIMALS IN NATURE 



suggested (e.g. Russell Brain, 1927, pp. 18-23) that functional 

 adaptations, such as we have described in the case of bones of 

 the legs, may play an important part in allowing animals to 

 survive until the necessary inheritable variations turn up. 

 This argument has one definite limitation, in that the mutation 

 would have no selective advantage unless it produced a 

 greater effect than functional adaptation or unless it produced 

 it more economically. 



To whatever extent we establish parallelisms with inorganic 

 phenomena, we are only clearing our problem of superficial, 

 largely man-made, difficulties. We are not solving the problem 

 of adaptation so much as rationalising our outlook on the 

 facts. 



(1) Specific differences. 



The most striking impressions that a taxonomic survey of any 

 large group conveys to one's mind are the manifold diversities 

 of species, the distinctness of the majority of these groups, 

 the fact that they usually differ in several associated characters 

 and the apparent triviality of these distinctions. If the theory 

 of Natural Selection is correct, we must assume that all these 

 differences must have arisen either because at some time or 

 another in their owners' lives they are of adaptive value or are 

 correlated with adaptive characters, or because they are the 

 result of a general adaptive reorganisation. We cannot too 

 strongly insist on the point already made that it is no use to 

 attempt to smuggle these facts of specific differentiation into 

 the proof of Natural Selection by an appeal to ignorance, or 

 by an assumption of correlation, or by pointing out a few 

 cases that seem explicable on very slender and unverified 

 evidence. We ought to be prepared to show that at least 

 50 per cent, of specific differences are definitely adaptive. 

 How far we are justified in attributing the survival of ' useless ' 

 characters to their correlation with less obviously ' useful ' ones 

 is discussed elsewhere (Chapter VI). 



The substantiation of the selection theory has been at- 

 tempted mainly by the collection of numerous individual 

 examples of apparently useful structures or habits. It is sug- 

 gested that no other theory can account for the large body of 

 facts amassed, but this argument would carry more weight 

 if there did not remain an even more numerous series of 



