i9i9 j Allen, Evolution of Bird-song. 529 



theory. As a theory it seems to be open to the fatal objection that 

 it fails to explain the relative uniformity of bird-song within the 

 species. If every variation has a chance of being perpetuated, 

 what is it, precisely, that decides for or against it and reaches the 

 same or a similar decision in all individuals of the species? Can 

 conservatism alone do this and thus permit progress in a definite 

 direction? 



It seems to me that something more positive in the way of an 

 evolutionary process is needed to account for the multifarious dis- 

 tinctive songs of birds than the unregulated inhibition of variations. 

 Granted that the song-impulse is rooted in the superabundant 

 vitality of the male, there must still be some process that selects 

 the variations to be preserved — whether it be sexual selection, 

 natural selection, or some other agency, or a combination of two or 

 more such agencies. 



As Mr. Aretas A. Saunders has pointed out (' Auk,' January, 1919, 

 p. 149), Mr. Hawkins has failed to make careful distinction between 

 call-notes and song. Song probably originated in the rapid repeti- 

 tion or special adaptation of call-notes, as Mr. Charles A. Witchell 

 has shown in his interesting book on 'The Evolution of Bird-Song' 

 (London, 1896), but it has assumed an entirely different function 

 in the bird's life, and, as Mr. Witchell and others have shown, it is 

 as a rule transmitted from generation to generation in an entirely 

 different way. Dr. Chapman, in his comprehensive discussion of 

 'The Voice of Birds' in the Introduction to his 'Handbook of 

 Birds of Eastern North America,' indorses "the theory of the 

 mimetic origin of bird-song," and says, "Birds inherit at least the 

 calls they utter when in the nest, just as a child cries instinctively, 

 but they apparently do not inherit their songs any more than the 

 child inherits the language of its parents." 



Call-notes are means of identification between individuals of a 

 species and, being necessary in order to bring the sexes together 

 and to prevent the separation of families, they have been evolved, 

 whether through natural selection or otherwise, to meet the needs 

 of the several species. No one thinks of attributing them to a 

 surplus of sexuality. The songs are similarly differentiated for 

 purposes of identification. Doubtless some, and perhaps many, 

 songs were evolved either through the ordinary processes of evolu- 



