1114 U.S. NATIONAL MUSEUM BULLETIN 237 paet 2 



aries edged and their greater coverts mostly the same color as the 

 back!" 



Another variation within dorsalis is the color of the lower mandible, 

 which occasionally is dark, but probably never as dark as the black 

 maxilla. Seven birds Philhps banded at Flagstaff, all but one in winter 

 and hence not juveniles, had lower mandibles ranging from "pinkish 

 lavender" through "bluish," "dark lavender" (a juvenile banded in 

 June), "dark blue," "blue black," and "black." 



F. M. Bailey's (1902) comment on the plumage of dorsalis applies 

 equally well to caniceps: "The coloration of most of the juncos is not 

 particularly protective except as the color pattern disguises the bird's 

 form, but the red-backed on the pine plateau of San Francisco 

 Mountain, Arizona, spends a large part of its time about the fallen 

 pine-tops, where the red of its back and the red of the dead pine 

 needles and old bark make a protective combination that, added to 

 the gray of the body, which offsets the gray of the branches, results 

 in a most effective disguise." 



Food. — Juncos are mainly terrestrial and obtain practically all of 

 their food on or very near the ground. Unless hard-pressed by 

 hunger they seldom feed on a sheK or other raised feeding station, 

 but prefer to pick up the seed spilled on the ground by other species. 

 F. M. Bailey (1928) mentions dorsalis in July "feeding among the 

 dead leaves, * * * scratching much like chickens" in the Capitan 

 Mountains of New Mexico, and mentions Major Goldman's finding 

 them in the Mogollon Mountains in late October, when "Small 

 parties were every\\'here hopping about, scratching among the leaves 

 through the thin snow, and when startled rising and alighting in the 

 lower branches of trees." 



Unusual feeding behavior of a "red-backed" junco is reported by 

 E. C. Jacot (MS.) in the White Mountains of Arizona on Mar. 13, 

 1935 "feeding from the bark of a pine tree 20 to 30 feet from the 

 ground, as painted redstarts often do. Perched on a Hmb and flew 

 to the trunk keeping tail somewhat fanned, then back to the limb." 



No detailed study has been made of the food of the gray-headed 

 junco, which probably differs little from that of the other, better- 

 known northern junco species. Of caniceps in the Sangre de Cristo 

 Mountains of New Mexico, F. M. Bailey (1928) says: "Among the 

 insects fed to the young birds were a caddice [sic] fly and a green cater- 

 pillar." A. R. Phillips notes that the dorsalis young he studied at 

 Flagstaff, Ariz., when 4 days old were fed green seeds and "apparently 

 a green caterpillar." One of Miss Bartlett's Flagstaff notes says 

 that at the "vine" nest previously mentioned Milton A. Wetherill 

 observed that the female parent "fed young birds with soft parts of 

 grasshoppers, carrying hard parts away." There, also, Hargrave 



