MUSEUM OF COMPARATIVE ZOOLOGY. 199 



were taken in the breeding season ; while in still other instances 

 (Tables II to P) the species were purposely chosen from among 

 such as find their northern limit of distribution near the locality where 

 all were taken. Of ten species of the latter class, twenty perfect male 

 specimens have been carefully measured,* the measurements embra- 

 cing a series of eighteen to twenty distinct parts; uuder such circum- 

 stances the variation in general size, in length, in alar extent, in the 

 length of the folded wing, the tail, the tarsus, the head, the bill, etc., etc., 

 commonly ranged from twelve to eighteen per cent. 



In respect to the differences in the general form of the body, two 

 leading styles of variation from the average form may be recognized in 

 nearly all species, namely, a relatively robust form, in which the stout- 

 ness extends to all parts, and a relatively slender form, in which the 

 slenderness is equally general. Variations of this general and sym- 

 metrical character are remarkable only for their extent, since in such 

 cases there are no marked discrepancies between the relative size of 

 different parts. Contrary, however, to our usual notions of exact sym- 

 metry in animals, the unsymmetrical variations are by far the most 

 frequent and important. 



Variation in the Relative Size of Different Parts. — In specimens 

 of average size of any given species, considerable differences exist 

 in the relative size of different parts. In individuals of the aver- 

 age alar extent of their species, for example, the length of the folded 

 wing may vary very considerably, in consequence of a difference in the 

 length of the primary quills as compared to the length of the bones of 

 the wing. The length of the folded wing or the alar extent may vary 

 with reference to the whole length of the specimen, in consequence of 

 differences in the relative length of the tail, the neck, or the body. The 

 tarsus also varies independently of variation in the general size, as do 

 also the toes to the tarsi, relatively short toes being found to accom- 

 pany tarsi of ordinary length, and, conversely, long toes short tarsi.f 

 The wing varies in its form in consequence of the different relative 

 development of the primary and secondary quills. \ The tail varies in 

 respect to its form, especially in regard to the degree of its emargination 

 or graduation, and, in some groups, in respect to the number of its 

 feathers. The bill also varies greatly in size and form. The variations 

 in these various parts will be considered separately and in detail. 



* See below, Tables H to P, pp. 210-219. J See Table E, p. 204. 



f Seo Table F, p. 205. 



