EVOLUTION OF THE COLORS OF BLRDS. 119 



natural selection is based upon the assumption that cer- 

 tain characters are useful, but we are not justified in 

 arguing from this that all characters have been devel- 

 oped by natural selection and are therefore useful. 

 Darwin himself freely conceded the inutility of many 

 characters, and, as Mr. Romanes says, " there is positive 

 evidence to show that the slight changes of form and 

 colour which chiefly serve to distinguish allied species 

 are often due to what Mr. Darwin calls 'the direct action 

 of external conditions,' such as changes of food, 

 climate, etc., as well as to mere independent variation 

 on isolated areas, and in some of our domesticated pro- 

 ductions, etc.; and in none of these cases do the specific 

 changes which result present a meaning of any kind." 

 In refuting Mr. Wallace's argument on the utility of 

 color Mr Romanes, in a footnote, quotes Darwin to the 

 effect that "each of the endless variations which we see 

 in the plumage of our fowls must have had some 

 efficient cause; and if the same causes were to act 

 uniformly during the long series of generations on many 

 individuals, all probably would be modified in the same 

 manner," and he adds: "The obvious truth of this 

 remark serves to dispose of Mr. Wallace's argument in 

 the Fortnightly, that ' the general constancy of colouration 

 we observe in each wild species,' of itself furnishes 

 sufficient proof that the colouration must be ' a useful 

 character.' Moreover, when using this argument Mr. 

 Wallace forgets that uniformity of colouration (whether 

 useful or unuseful) is preserved in wild species by free 

 intercrossing. Where this is prevented as by isola- 

 tion or migration variations of colour very frequently 

 do take place, just as in the then analgous case of our do- 

 mesticated strains." 



Concerning the swamping effects of intercrossing, Mr. 

 Romanes shows that the variations cited from Mr. 



