354 U. S. NATIONAL MUSEUM BULLETIN 217 



a properly analytic approach. In the Arctic the situation is simpli- 

 fied. The number of species is small, none are passing through on 

 their way to higher latitudes, and the entire avifauna faces measur- 

 able and apparently irresistible external influences — a short summer, 

 extreme seasonal changes in temperature, and a resulting variation in 

 food supply. These factors impel arctic-nesting birds to regulate 

 their lives with strict regard for time and place. 



The known geological history of arctic Alaska, moreover, indicates 

 that the populations of migratory land birds now breeding there 

 could not have existed in their present ranges until after the last 

 extensive glaciation had subsided about 10,000 years ago. This im- 

 plies that their present migratory routes are recent and that the 

 characters by which many arctic nesting populations are distinguished 

 taxonomically must have developed within that time span. 



I have found no evidence, however, that the taxonomic characters 

 distinguishing these arctic population of birds are adaptive responses 

 to their arctic environment. Apparently these characters are the 

 result of isolation of the breeding populations. Such isolation could 

 lead to the production and establishment of characters which are 

 ecologically indifferent and not necessarily adaptive — ^which survive 

 because they impose no important hardship on their bearers. These 

 characters may result from variability in a limited gene pool, such 

 as would be available to a fragment of a species isolated regularly 

 during the reproductive season, and they would not have become 

 established except under the special condition of isolation. In each 

 of several interglacial periods since the Pleistocene, new arctic popu- 

 lations of the several species of birds must have been reestablished 

 without essential modification, and it would appear that adaptability 

 to arctic conditions may have remained latent in many species during 

 glacial periods when it could not be fully expressed. Such cases do 

 not argue against the general probability of adaptive evolution. 

 They do illustrate the fact that visible characters distinguishing geo- 

 graphical races do not necessarily demonstrate an ever-present 

 progress toward adaptive evolutionary modification. The develop- 

 ment by warm-blooded creatures such as birds of an adaptability to 

 the rigorous demands of arctic conditions must have taken place 

 long ago, or it would not have been possible for many of these species 

 dwelling in more temperate latitudes to have extended their ranges 

 northward in the last post-glacial period. 



Our studies have demonstrated in physiological terms the nature 

 of some of this adaptation to cold. Although the Arctic may seem 

 unfavorable to the human observer, arctic animals are adapted to 

 cold by insulation and behavior, so that they can, in most cases main- 

 tain their normal body temperature without expending special 



