1920] Swnrth: Revision of Avian Genus Passerella 103 



same valleys ? Why should the birds of one mountain retire in winter 

 to far distant valleys and foothills, while a closely related subspecies 

 moves in from a remote mountain habitat to occupy the valley and 

 foothills most nearly adjacent to the summer home of the first men- 

 tioned race? We know that this happens, but it is hard to conceive 

 of circumstances whereby such results are brought about. 



The species Passerella iliacu is restricted to North America, and 

 is practically confined to that part of North America that lies north 

 of the United States-Mexico boundary line, I have, in fact, seen but 

 one published reference to the occurrence of the species south of 

 that line, a record for Guadalupe Island, Lower California (Bryant, 

 1887, p. 303). Passerella iliaca occurs in summer over a large part 

 of Boreal North America, extending from Newfoundland west across 

 northern Canada to western Alaska, and south on the summits of the 

 Rocky Mountains, the Sierra Nevada and the Coast Ranges, to Colo- 

 rado, Nevada, and southern California. The breeding area occupied 

 by the subspecies Passerella iliaca iliaca is mainly in the Hudsonian 

 zone. The western subspecies are mostly in the Canadian, the races 

 occurring in the mountains of California being markedly restricted 

 within the limits of this zone. Wliere the Hudsonian and Canadian 

 zones extend over vast distances uninterruptedly, as in northern 

 Canada and the Pacific coast region of Alaska, there is probably nearly 

 or quite uniform distribution of Passerella also, either with no differ- 

 entiation into distinguishable local races, as in the wide-ranging 

 Passerella iliaca iliaca, or else with gradual change by a series of steps 

 from one extreme to another, as is seen on the coast of Alaska. In 

 the latter case there are apparently other variable conditions which 

 constitute influencing factors within the limits of the zones inhabited. 



In the western mountains the Canadian zone extends southward 

 at increasingly higher altitudes and in constantly diminishing area; 

 Passerella accompanies this zone nearly or quite to its southern 

 extremity, but its colonies are naturally more and more widely sep- 

 arated and of lesser extent as the southernmost limits of the zone 

 are approached. These scattered breeding colonies are not necessarily 

 of distinguishable subspecies in every case, however, definitely isolated 

 tliough each may be. In some instances a distance of but a relatively 

 few miles of lower zones serves as a boundary between subspecies that 

 are widely different in appearance, as is the case with canescens and 

 mariposae; while again one subspecies may extend over a series of 

 disconnected Canadian zone islands, with no appreciable change in 



