GENESIS OF BIRD-SONG. 91 



of the reptile are indeed there, but they have 

 already received the stamp of the bird: 51 and 

 I may add that, as regards Odontornithes 

 collectively, the feathers are indeed there, 

 and the stamp of the bird, but the old reptile 

 character is still present, scarcely more than 

 dominated by the ornithic features. I have 

 said that it may be doubted whether any of 

 the Odontornithes were good flyers. By good 

 flyers I mean not merely strong flyers (like 

 the teals) , nor sailers ( like the hawks and 

 buzzards), but flyers whose movements in 

 the air are almost instantaneous, like the 

 highest type of oscines, say the mocking-bird, 

 or the cardinal grossbeak, a facility of flight 

 absolutely necessary to arboreal life, where 

 so many thorns, spikes, branches, twigs, vines, 

 and sprays have to be suddenly avoided in 

 the midst of the swiftest motion. Some of the 

 toothed birds of Marsh's smaller group may 

 have been as good flyers as our gulls, strong 

 and tireless ; but they could not dodge a dozen 

 twigs in a second, as I have seen a sparrow 

 do in full flight. 



The discovery of Palceospiza bella, a well- 

 preserved, almost complete skeleton of a spar- 

 row-like bird in the insectivorous shale of 

 Colorado, has given us the nearest approach 

 to a song-bird yet found in the old rocks ; but 

 the bill is lacking. Most probably Palaeospiza 

 was an oscine, in the ornithological sense, but 

 I think we may well doubt whether it could 

 sing, in the true meaning of the word. Its 

 position in the insect-bearing shale further 

 favors our classing it as insectivorous, an- 

 other characteristic of the true song-birds; 

 but this would not give it a song, for many 

 of the existing oscines have no song to sing, 

 chirp and pipe and squeak as they may. 



