THE ARRIVAL OF THE FITTEST 183 



very rare mutations ? No real attempt has been made to sur- 

 mount all these difficulties. Merely because some mutations have 

 been observed it has been assumed that all evolution proceeds 

 by steps and not up a smooth incline. Merely because some 

 mutations, the inheritance of which is alternative, are comparatively 

 stable, it has been assumed that all mutations are absolutely stable. 

 Merely because fluctuations cannot be fixed by a short course of 

 experimental selection, it is assumed that they cannot be fixed by 

 the age-long selection of nature. Merely because varieties which 

 have been evolved under artificial selection display when crossed 

 alternative reproduction in widely divergent characters, it has been 

 thought that all inheritance is alternative. All these inferences are 

 illegitimate expansions of single inductions illegitimate, because 

 there has been no rigorous deductive inference of consequences, 

 and no appeal to reality for confirmation. 



303. It is not to be maintained, of course, that no mutation has 

 ever been selected by nature. It is maintained merely that species 

 are so accurately adapted to their environments, including one 

 another, and that their structures nervous, circulatory, alimentary, 

 bony, muscular, glandular, etc. are so closely co-adapted that a 

 favourable mutation must be an extraordinarily rare thing so rare 

 that as factors in evolution mutations are negligible. We are 

 told sometimes that, whilst Natural Selection may account for the 

 survival of the fittest, it does not account for the arrival of the 

 fittest. This jingle of words implies, of course, that fluctuations are 

 too small to have selection value. 1 But we shall see in later 

 chapters, when we study the recent evolution of man, who, as I 

 say, is the wild type we are best able to observe, that fluctuations 

 have so real a survival value that, according as men fluctuate favour- 

 ably or unfavourably solely on that account they survive and see 

 their grandchildren, or perish in youth. The arrival of the fittest, 

 owing to the necessary co-adaption of parts and species, is an im- 

 mensely greater difficulty for the mutationist than for the selectionist 



304. We have noted that living beings are bundles of adapta- 

 tions. No phenomenon of life is more universal than the occurrence 

 of fluctuations. Presumably, therefore, fluctuating variability is 

 an adaptation, and a highly useful character. It has a function. 

 Mutationists at any rate many of them admit, for example in 

 the case of trotting-horses, that fluctuations are transmissible to 

 some extent, and may be the sources of some temporary evolution. 

 But we are asked to believe that opulent but parsimonious nature, 



1 See ' 4 826 (footnote). 



