ORIGIN OF SONG 167 



faculties, there is at the same time evidence 

 which demonstrates that such variations exercise 

 no influence on the course of mating ; and 

 inasmuch as it is difficult to conceive of 

 any voice departing more from the normal 

 type in these particular qualities than the 

 immature does from the adult, if there be 

 degrees of suggestive influence, we must seek 

 it in some other direction. There remain the 

 two other characteristics which we found to 

 be constant under all circumstances, namely, 

 loudness and specific distinctness ; and if, in 

 addition to serving the" purpose of disclosing 

 the positions of the males, they serve to 

 evoke some emotion in the female, which 

 helps to further the biological end of mating, 

 so much the more reason is there for their 

 survival. 



There can be no question that this ingenious 

 and attractive theory, if it were true in its 

 special application to song, would immensely 

 simplify interpretation, and moreover that 

 preferential mating would 'contribute not a little 

 to the success of the whole territorial system. 

 No one can deny the strength of the argument : 

 that the sexual instinct, like all other instincts, 

 must require a stimulus of an appropriate kind ; 

 that the effect of the sexual call upon the 

 female cannot be neutral ; and hence the prob- 

 ability that stimulation varies too ; no one, I 

 say, can question the strength of this evidence, 

 and, one might add, of the evidence derived 

 from the analogy of the human voice. But 



M2 



