PASSERES FKINGILLIDAE JUNCO. 265 



light ash, and also, as Mr. Aiken informs me, not infrequently shows a decided 

 tendency to the white banding of the Avings. This is well shown in a 

 specimen taken at Fountain, Colo., in December, which has two well defined 

 bands, though not quite so conspicuous as in typical examples of aikeni. 

 Junco caniceps of the central Rocky Mountains of the United States is at 

 once distinguished from any of the above by the bright, reddish, chestnut- 

 brown of the interscapular region. 



In the southern Rocky Mountains, in New Mexico and Arizona, is 

 found var. dorsalis, which seems to combine certain features peculiar to 

 both caniceps and cinereus, and also in certain other points to differ from 

 either. In the restriction of red to the interscnpular region, it is like 

 caniceps ; but, in quite a number of specimens collected in New Mexico 

 during the past season the tertiaries are strongly tinged with rufous, show- 

 ing in this respect an approach to cinereus, where the chestnut of the back 

 extends over the wing-coverts and inner secondaries. 



The bill above is brownish-black, below whitish, thus differing from 

 caniceps, which has a flesh-colored bill, and apparently approaching cine- 

 reus, where it is black above, below yellow. Like cinereus, also, the pale 

 ash of the tlu'oat fades gradually into the white of abdomen, instead of 

 being, as in caniceps, abruptly defined. 



Of quite a large series of specimens collected by myself the past 

 season, and others in the Smithsonian collection, I have seen none which 

 are not readily assignable to one variety or the other by the distinctive 

 features pointed out. The theory of hybridization, which might be admis- 

 sible were only one or two specimens known possessing intermediate char- 

 acters, seems wholly inadequate as an explanation in the case of either 

 annectens or dorsalis, where the forms extend over very extensive regions, 

 and preserve their distinctive characteristics intact. Whether cinereus of 

 the table lands of Mexico, with a local variety, alticola, of the mountains 

 of Guatemala, may not justly be entitled to specific rank, is a matter of 

 considerable doubt. While the typical forms of caniceps and cinereus are 

 widely different, dorsalis, intermediate in its habitat, seems also intermedi- 

 ate in its characters, and it therefore may be best to treat the two (caniceps 

 et cinereus) as only separable as varieties rather than as distinct species. 



