ICO DwiGHT, Variatiofis of the American Goldfinch. I Apr 



that a better understanding of individual and seasonal variation in 

 recognized species will do much to remedy a growing evil that, 

 whatever its cause, is greatly to be deplored. If, eventually it be- 

 comes both necessary and proper to recognize by a name every 

 minor variation due to environment, it seems desirable, first of all, 

 to learn how much of it is individual and how much seasonal, so 

 that the value of the geographical element remaining may be better 

 characterized. This is no simple matter, but one involving more 

 tedious comparisons of plumages and measurements than are 

 required in the mere naming of races. 



In order to show how great may be the variation in a single 

 species, I have selected the American Goldfinch {Astragalinus 

 tristis) as being peculiarly suitable for the purpose. It is widely 

 distributed over North America, and being to a certain extent 

 resident wherever found, its races, of which two have been, de- 

 scribed, ought to show marked characters,while individual variation 

 ought to be very little. This does not seem to be the case, for the 

 subspecific characters are slight, while individuals differ widely in 

 both color and size. The seasonal variation in plumage is consid- 

 erable, there being no less than six plumages easily recognizable 

 in the male, and although most of the feathers are renewed by 

 moult twice in the year they are subject to a large amount of fading 

 and actual loss of substance. 



The two geographical forms 2.XQ. palUdiis, a large pale bird from 

 Arizona described by Dr. Mearns (Auk, VII, 1890, p. 244), and 

 salicamans, a small dark bird from southern California described 

 by Mr. Grinnell (Auk, XIV, 1897, p. 397). A small series oi pal- 

 lidas^ including the type, has been available at the American Mu- 

 seum of Natural History and Mr. Grinnell has kindly loaned me 

 a series of eighty specimens of salicamans. These have been com- 

 pared, plumage by plumage, with over one hundred specimens of 

 tristis in my own collection. The variations in dimensions have 

 been tabulated, and those of color may be found under the differ- 

 ent plumages, which I have designated in numerical sequence. 



Several new facts are brought to light by the study of this 

 material. It appears that adults have, on an average, appreciably 

 longer wings and tails than young birds and that each loses in 

 the course of a year through wear an appreciable amount of this 



