372 Clark, The Classificatioji of Birds. \^^ 



cretory or central nervous system. It may be said that these sys- 

 tems are too uniform throughout the class to be of any value in 

 classification, but it is very possible that a more intimate acquaint- 

 ance with their structure will make them of very great value. 



The question now confronts us, What characters are of the 

 most importance in determining the relationship between two 

 birds, and what are of the least value ? There can be little ques- 

 tion that the least valuable characters are those connected with 

 the form and external characters of the bill and feet. Somewhat 

 more valuable, but still very uncertain, are the characters of the 

 wings and tail, and hardly more valuable is the nature of the plum- 

 age, such as the presence or absence of down, aftershaft, tuft on 

 oil-gland, etc. The arrangement of viscera, muscles, and blood- 

 vessels, are of some value, but probably less than the characters 

 offered by the skeleton. It must be borne in mind, however, that 

 the skeleton, and especially the skull, would be very liable to 

 marked changes, accompanying changes in the bill, feet or wings, 

 so that skeletal characters are by no means as indicative of real 

 relationship as many writers assume. Certain characters con- 

 nected with breeding, as the condition of the young when hatched, 

 are of considerable importance ; but there is reason to believe 

 that even these are rapidly modified under changed conditions. 

 As already suggested, the structure of the urino-genital organs, 

 and the central nervous system would probably be very slowly 

 modified, and ought therefore to furnish some very valuable funda- 

 mental characters. It is the purpose of this paper to show that 

 the arrangement of the contour feathers, that is, the pattern of 

 the pterylosis, is a similar character, in that it is only slowly modi- 

 fied, and therefore serves as a most important clue to the relation- 

 ship of the various groups of birds. 



The general opinion among ornithologists at the present time 

 is that the pterylosis offers us little assistance in determining the 

 relationship of birds, because it is believed that the arrangement 

 of the feathers is governed largely by the shape of the body, and 

 that the distribution of the tracts depends to a considerable degree 

 on the habits of the bird. It is hard to see how this opinion has 

 arisen, for there is much evidence to the contrary. If one will 

 compare a plucked Swift and Swallow, a Colie and Cuckoo, an 



