Vol. XXIJ Grinnell, Chestnut-backed Chickadee. 365 



lateral brown areas suddenly weaken to pale rusty ; while from 

 San Francisco south past Monterey (Santa Cruz District), adult 

 birds have the sides pure smoke gray without a trace of rusty. 

 (See Map II.) 



The species thus presents geographic variation within itself, and 

 three distinguishable forms have been named, respectively, the 

 Chestnut-sided Chickadee {Pariis rufescens rufescens), the Marin 

 Chickadee {Farus rufescens neg/ee/us) ,a.x\d the Santa Cruz Chicka- 

 dee {Pariis rufescens barlotoi). But all three subspecies are unmis- 

 takably the Chestnut-backed Chickadee {Farus rufescens). (For 

 detailed descriptions, distribution and synonymy see beyond.) 



This southward paling of the lateral feather tracts seems to be 

 parallel to the relative decrease in the humidity of the regions 

 occupied. But still, even the Santa Cruz District with its gray- 

 sided barlotvi has very much greater rainfall and cloudiness than 

 regions immediately to the southward and interiorly. The too 

 abrupt aridification with accompanying sudden floral changes 

 apparently forms the present barrier to further distribution in 

 these directions. 



The paling of the sides in the southern bird seems to be a sec- 

 ondary condition, as I hope to show further on by age comparisons. 

 We can reasonably infer that Farus rufescens rufescefis was the 

 ancestral form from which Farus rufescens ncgkctus and then Farus 

 rufescefis barlowi successively arose through exodus distally from 

 its point of differentiation further north, where the faunal condi- 

 tions were doubtless then as now most effective. 



First, as to the origin of the species, Farus rufescens. Can we 

 find a chickadee now occupying a faunal area which can be con- 

 sidered as nearer the common ancestral form than rufescens now is ? 



An affirmative answer seems plausible when we come to consider 

 Farus kudsonicus, which occupies the interior of Alaska and Brit- 

 ish Columbia east to Labrador and Nova Scotia. This wide- 

 ranging boreal species also affects coniferous forests, and according 

 to my own experience possesses life habits quite similar to those of 

 Farus rufescens \ in fact to me indistinguishable. The latter differs 

 from Farus hudsonicus in smaller size and particularly in shortness 

 of tail. The color areas on the two species are coextensive, but 

 the colors themselves are different in intensity. The top of the 



