374 THE VARIATION OF ANIMALS IN NATURE 



accuracy and closeness of correlation, or that the selection 

 of very rare mutants could produce that harmonious co- 

 ordination in which one part depends on the appropriate 

 appearance and degree of development of another part. 



In suggesting that group-divergence and local variation are 

 due to subordinate factors such as isolation of various kinds, 

 random spread and the reshuffling of heritable characters, 

 but that certain evolutionary tendencies may be referable to 

 an innate ' momentum ' and self- regulation, we ought not to 

 forget that after all one of the tests of an evolutionary theory is 

 its capacity to account for the spreading of variants and the 

 transformation of populations. In this respect, as we have 

 admitted, Natural Selection enjoys a strong theoretical advan- 

 tage. But it is only a theoretical advantage, and should not 

 influence our judgment of the theory if the more important 

 direct and circumstantial evidence is defective. 



Finally, we would point out that, if indeed group divergence, 

 specialisation and organisation are due to different causes, it 

 is by no means easy to assign to these factors their particular 

 spheres of influence with any accuracy. Some group 

 divergences are almost certainly void of adaptive significance ; 

 but in others we may discern the beginning of specialisation. 

 Organisation, in its more profound expression an attribute of 

 all living matter and independent of the temporary influences 

 that evoke specialisation, may sometimes be guided along 

 particular channels by specialisation. 



In arguing that an element of self-regulation and self- 

 organisation has had an influence in evolution we are aware 

 that we are touching certain profound and speculative issues. 

 If this organising activity is indeed an agent in producing the 

 main adaptive tendencies in evolution, it might be argued that 

 the gradual upbuilding and perfection of adaptations, because 

 they involve so large an element of design, must also involve 

 some reference to a purpose independent of survival value 

 and chance, and existing as an end in itself. We have to 

 admit that, if we were to relegate survival value to a sub- 

 ordinate role in the causation of evolution, the element of 

 design and purposefulness has to be explained. It is not 

 likely that the mere interaction of developing parts and their 

 reciprocal effects on one another could produce the ordered 

 and purposeful designs which we see in adaptation. For those 



