332 BULLETIN 196, UNITED STATES NATIONAL MUSEUM 



breeding birds. This leaves us, so far as the above records are con- 

 cerned, with only the bird seen by Grinnell on June 14. This date 

 certainly suggests breeding. The bird was hunting among some 

 willow bushes and stunted spruces, and it was expected that it would 

 prove to be nesting, but eventually it flew right away, so that even 

 in this case there was no very strong evidence of breeding. Moreover, 

 it is not irrelevant to point out that apparently not one of the earlier 

 observers ever heard the song in Alaska, with the exception of Town- 

 send (1887), who records that one obtained by him in a thicket far 

 up on one of the highest hills of the middle Kowak River region was 

 still in song on August 1 . 



However, the question of breeding in Alaska seems to be settled 

 by the observations of Dixon (1938) in the Mount McKinley National 

 Park. He states that "Kennicott's willow-warbler was a fairly com- 

 mon breeding bird on the upper Savage River in 1926. Here, on 

 June 20, we found half a dozen willow warblers singing in one tract 

 of spruce woods. * * * Three specimens were collected in June 

 1926, and two proved to be adult males in full breeding condition." 

 Even Dixon, however, found no nests, and he records further that 

 "in 1932 I repeatedly visited the exact locality where these warblers 

 had been found in 1926, but I neither saw nor heard them. All 

 summer a continued search was carried on in the McKinley region 

 but not a single willow warbler could be found. The late heavy 

 snows had apparently prevented their reaching this inland district." 



Such erratic and inconstant breeding distribution from season to 

 season, dependent mainly on weather conditions in spring, is a char- 

 acteristic phenomenon with many Arctic birds: a species may be 

 present in numbers in a given district in one year and completely 

 absent in the next. Nevertheless, it may be suggested that the field 

 data are not particularly favorable to the existence of a distinct 

 Alaskan race. The writer speaks with some diffidence on this Amer- 

 ican problem, especially as there may be more recent data not acces- 

 sible to him, but the evidence does seem to raise the question whether 

 this warbler may not be still only in the process of colonizing Alaska 

 from Asia, a question that clearly has a bearing on the status of the 

 race kennicotti. Only more material from;«both sides of Bering 

 Strait and more field study in Alaska can finally settle the problem. 



Acanthopneuste, or, as the writer would prefer to call it, Phyllo- 

 scopus, borealis (for Ticehurst, already quoted, has shown that the 

 separation of Acanthopneuste has very little basis) is biologically note- 

 worthy on account of its extremely interesting migration. There are 

 well-established cases of migrants extending their winter range in 

 relation with an extension of the breeding area, but the present species 

 exemplifies the reverse situation, the adherence to ancestral winter 



