468 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1918. 



Watch the finch as be dances about his mate, fairly losing himself 

 in a frenzy of ecstasy, flashing his wings in a wild delight and 

 prancing about and chattering; the antics of the noisy street spar- 

 row, the prancing and cooing of the pigeons, and there is only one 

 evident conclusion. It is not for the purpose of mating but the 

 more immediate purpose of hastening the female to fulfill her nat- 

 ural function. There are times when two or more males are involved 

 in these antics, in which case there must be at least an unconscious 

 choice on the part of the female, or a battle royal which will drive 

 the competing males away, but in the vast majority of cases there 

 is only one ardent male bird in the presence of the female, and he 

 is often the bird with which she has already mated. 



A weakness of the sexual selection theory that has not been 

 given sufficient consideration is that the song of birds has been 

 treated too exclusively in connection with the mating season. Men 

 have riveted their attention on those rapturous bursts of song which 

 precede and continue through the mating time, and have given too 

 little attention to the fact that few birds are ever wholly voiceless, 

 that most birds speak the sign or voice language, at least to some 

 extent, all through the year. 



Most of our best singers have two distinct song periods. One 

 begins with the arrival of the advance guards of the migrating hosts 

 and continues until the broods of young birds are hatched. When 

 the young birds have left the nest and are able to care for themselves 

 there is a cessation of the full, joyous songs, September being gener- 

 ally the silent month. Then many of the birds begin to sing the 

 last of September or the first of October and continue until Novem- 

 ber. Bicknell has determined definitely the limits of these song 

 periods for many of our birds. The house wren begins to sing its 

 love song in April and continues to the last of July or the first of 

 August. After a period of comparative silence it begins its autumn 

 song which has none of the spontaneity of the spring song, but con- 

 sists of a " low rambling warble," which continues to the middle of 

 October. The black and white creeping warbler sings from April 

 to the late June. Its second period begins from the ninth to the 

 twenty-second of August and lasts only a few days. The first period 

 of the oven bird stops by the end of June. The second period be- 

 gins in August, at first haltingly, as though it had forgotten how 

 to sing, but finally bursts into full song by October. The wood 

 thrush sings from its arrival in late April or early May until the 

 middle of August. It is not heard again until October and then 

 only the call notes, never the full song. 



Bicknell attributes this period of silence to the moult of the bird. 

 Tn many cases the moulting periods of our song birds correspond 

 more or less closely with periods of silence, voice being renewed 



