652 FROCEEDINGS OF THE NATIONAL MUSEUM. 



due to the jireater size of the former, and we may asanme that the latter 

 represent more nearly the primitive condition. This assumption is 

 based on the greater simplicity of the dorsal tract and the cervical tracts 

 in the quails and the small number of rectrices. At the same time it 

 must be remembered that it is a i)ure assumption adopted only for cou- 

 venien(;e in pointing out the relation of the genera to each other. The 

 common bob white and its allies will serve, then, as a starting point from 

 which to develop the other genera. Lophortyx is nearest to Colinxs, 

 having the sanu'. number of rectrices and resembling that genus closely 

 in other ways. But some of the feathers of the crown form a distinct 

 crest tract. From Lophortyx may have been derived, ou the one hand, by 

 increased size and greater specialization of the crest, the genus Oreoriyx; 

 and on the other hand, by revei'se changes in the crest and in(;rease in 

 the number of rectrices, the genus CaUijypla. Th(5 degeneration of the 

 crest has gone further in Cyrionyx than in CaUipcpla, but the 12 

 rectrices have been retained, though they have greatly degenerated in 

 size and importance. This arrangement of the genera may be seen at 

 a glance from the accompanying diagram: 



X CivUipepl.a. 

 Oreortyx. ^^ 



\, ^.- '--.^ 



^Loi-hortyx. "^^^ Cyrtouyx. 



X 



X 



• s 



N 

 \ 

 N 



^ Colinus. 



Which genus of grouse to use as a starting point is not so easy to 

 decide, but for convenience we will take (Janave. It must not be sup- 

 posed, however, that this is meant to im])ly that that genus is nearest 

 to the quails. But it has the smallest number of rectrices and the 

 simplest pterylosis, and it is easy to show its connection with most of 

 the other genera. Dendrayapus has developed from (Janaoe by increase 

 of size, accompanied by greater development of the femoral tracts, a 

 marked increase in the number of rectrices, and some changes in the 

 dorsal tract. LayopuH has been modihed from Canace only in the 

 great«!r amount of feathering on the feet and the greater develoi)inent 

 of upper tiiil coverts. Tympmnichus, Pedioewtes, and Boh((S((, form 

 Still another branch, of which the first is perhaps nearest the ancestral 

 form, and J>oii<(sa the most moditied. All three of these genera have 

 an increased numl)er of rectrices and modilied cervical tracts or apte- 

 ria. ]n Bonasa there has been a marked decrease in the amount of 

 feathering on the feet, and the special pteryla' on the branches of the 

 lower cervical tract are very noticeable. The position of (Jeutrocerciis 

 is not easy to determine, ;is it shows greater specialization tlian any 

 other genus. This is indicated by the changes in the arrangement of 

 the cervical tracts, in the greater size of the dorsal and femoral tracts, 

 and in the increased number of rectrices. Wliether it is the descend- 



