JUNCO CINEREUS VAR, CANICEPS. 145 



habits it. resembles hycmalis ; its twitter is the same, but its song is different. Of all the 

 varieties that are found in this region, this is the most constant in its n]arkin<TS. 

 Among scores of specimens, I never met with auj- that exhibited in any marked de<'Tee 

 the characteristics of any of the other races combined with those of the Chestnut- 

 backed species. 



J. (iinievteiia. — Chestnut-sided Snow-bird. This race or species is supposed by some 

 ornithologists to be a hybrid between orefjoniis and cu)iiceps, a supposition showing the 

 danger of generalization on too narrow grounds, for there is nothing to support the 

 idea esceiit the combination of certain characteristics belonging to those two races; 

 and on the same principle Colaptes clirysoides might be supposed to be a hybrid between 

 C. aiiratus and C. mcx'icanns. The Chestnut-sided Snow-bird is as well marked, and 

 almost as constant in its coloration, as J. eaniceps, and far more so than orcr/oiius, 

 hi/emalis, or ailiCni. It is common only in spring, associating with other varieties, and 

 remaining till April, being the last to leave of the migratory species. It ranges rather 

 higher than either the Oregon or the Black Snow-bird, occasionally extending as high 

 as 9,500 feet, or higher. In winter I have not noticed it, but am informed by Mr. Aiken 

 that it is very abundant on the plains from early in October till May, and that it out- 

 numbers canicrps and oref/onus together — a singular fact, if it is a '■ hybrid." Undoubt- 

 , ediy it does occasionally mix with both these varieties, as specimens sometimes sho v 

 a tendency to one or other of them, usually toward the latter ; but I have never seen 

 any that I could not refer, without hesitation, to one or the other of three species — 

 eaniceps, ainiecteiis, or that which embraces hjieinalis, oregonus, and aikeui as sub- varieties. 



J. aiken'i. — White-winged Snow-bird. Abundant; does not breed. This race is evi- 

 dently the most northern of the live, as it arrives latest (in November) and departs 

 earliest (in March), ranging higher also than any of the others, up to 10,000 feet, and 

 probably further still. In its notes and habits the White-winged Snow-bird diflers 

 somewhat from its congeners. Its song is louder and sweeter ; it is less gregarious in 

 its nature; and it frequents brushy hill-tops and mountain-sides, high up above the 

 valleys, and rarely visited by the other species during the winter. It is the only Snow- 

 bird at all common during winter, choosing as its favorite haunts the bushy ravines and 

 hollows, as well as the vallej's of the larger streams, and wandering thence far up on 

 the mountains, associating in small parties only, more than six or eight being rarely 

 seen together. During the coldest weather, only the well-marked typical birds are 

 seen, among which are both males and females, the former being most numerous ; but 

 toward the close of winter the females become more abundant, and among a large 

 series of specimens, obvious approaches to both hi/cmalis and orcgoiins may be distin- 

 guished, especially toward the former. The intergradation, however, is by no means 

 as perfect as that between the two latter races, and a specimen that cannot be de- 

 cidedly referred to either aikeiil or orcyoiius-lniemalis is unusual. The white wing-bands 

 vary very nuich ; in the largest malt'S they are almost always broad and well-detined ; 

 in the small males and females they are narrower, sometimes almost obsolete, occa-' 

 sionally wanting on one wing and present on the other, and sometimes wholly want- 

 ing or indicated by the faintest trace. In the latter case, the other characteristics of 

 the bird are those of hi/cmalis, while the peculiar featuies of the latter bird or ot'ore- 

 yoiim are never to be seen wliere the white wing-bands are well defined. Well-devel- 

 oped males frequently have white lores or wbiti; spots on the head and neck. 



Without a very large number of sitecimens from ditierent localities, it is difficult to 

 discuss, intelligently, the relations of these races or si)ecies; but from a careful study 

 of their haljits, notes, distribution, and migration in this region, as well as the com- 

 parison of a large number of spircimens, I cannot help thawing the conclusion that at 

 leiist three, or jjerhaps four, of the various forms have, in the words of Dr. Cones, 

 " passed the nuMely varietal stage and become nascent species, though still unstable ia 

 character." These may be arranged as follows : 



I. JiDico caiikrpn. Sp. ch. Interscaymlar region bright rufous. Breeds abundantly 

 in the mountains of Colorado, from 9,000 feet up to timbcr-liiu'. Retreats south 

 of the Territory in winter. 

 II. J. (iiiiucteiis. Sp. ch. Sides clear cinnamon-fulvous. Abundant in Colorado during 



winter and spring, but breeds north of the Territory. 

 III. J. kijenudiH. With three varieties : 



a. aikcni. Var. ch. Two broad, well-defined white bands on the wing-coverts. 



Abundant in winter, ictiring noith earlier than the other races, and 

 evidently the most northern in its distribution. Intergratles witli th«' fol- 

 lowing form, though not as intimately as that does with the succeeding. 



b. hut-mdliH. Var. ch. IS'o wing bands; upi)er parts with no rufous, except a 



rusty tinge in autumn. Less abund.int than tiie others, reaching hero its 

 wc'siern limit; breeding north of the Territory. Intergrades iniiiercei)t- 

 ibly with tlu; next. 



c. orefidiiiin. Var. ch. Hack and wings with a deci<led rusty-brown wash. 



Al)und.int, n-aehing here its eastern limit. Hreeds northward, retiring, 

 from Colorado with the preceding, or a little later. 



10 



