■JqS Grinnell, a New Race of Spintis tristis. foct' 



salicamans is dull greenish j-ellow on the throat, becoming still duskier 

 anteriorly. Even juveniles of the Willow Goldfinch just from the nest 

 are deeper and darker colored than those of 6'. tristis proper. 

 Habitat. — Pacific Coast, from Washington to Southern California. 



This geographical race seems as well pronounced as many others 

 of the dark Pacific Coast forms. As might be expected, speci- 

 mens from Oregon and Washington in winter plumage are slightly 

 darker and browner than Southern California birds ; but as the 

 Willow Goldfinches are of a wandering disposition, and at least 

 slightly migratory in the northern part of their habitat, this vari- 

 ation is not great. 



The habitat of salicamans seems to be perfectly divided southerly 

 from that of pallidus of the Great Basin. The Colorado Desert 

 and the high Sierras form a broad territory from which I have 

 seen no records, and which cannot be regularly crossed. The 

 only specimens of an intermediate character are from the northern 

 part of the Great Basin, where they seem to combine in part the 

 characters of tristis, pallidus and salicamans., and are hardly refer- 

 able to either. Several specimens of the latter from California 

 show a considerable extension of the white on the inner w^ebs of 

 the tail-feathers, but so do occasional eastern birds. 



I have examined 175 Goldfinches from nearly every State; the 

 winter birds vary greatly in respect to the tone of coloration, but 

 as this species is more or less migratory throughout its range, it is 

 hard to draw any lines. Specimens from Texas (Fort Clark, Coll. 

 E. A. Mearns) in particular, are light colored and grayer than any 

 others, though showing no pronounced tendency toward pallidus. 



If I judge rightly from specimens at hand, the spring moult of 

 the Eastern Goldfinch takes place during two or three weeks in 

 April. In Southern California, however, the spring moult in 

 many individuals of 6". t. salicamans begins in January and 

 extends far into May. Thus a series of specimens taken from 

 January to May, presents every gradation of plumage from winter 

 to full summer, but none entirely in one or the other. They fre- 

 quently breed early, before the spring moult is completed. I 

 have adult males, taken with sets of eggs in April, that possess 

 but a few of the bright yellow feathers, and the black crown is 

 but imperfectly indicated. In certain tracts of feathers, I have 



