314 U. S. NATIONAL MUSEUM BULLETIN 217 



The variety of departure dates among tlie passeriform species is 

 conspicuous. Kobins are the latest recorded, except for shrikes. The 

 latter, I suspect, occasionally winter in Alaska. My records fail to 

 show many late tree sparrows. Although they are occasionally seen 

 in September I have no idea when most of the large summer popula- 

 tion of these sparrows leaves Anaktuvuk. The early start of the 

 movement of wheatears and yellow wagtails is not surprising if we 

 regard the great distance of their migratory course to Asia, but a few 

 wheatears have been seen on the first of September. It cannot be said 

 that migrants to the nearer wintering grounds leave Anaktuvuk late 

 in the season, for the early arriving horned lark has its winter range 

 nearer than those of many species which depart at later date. There 

 is no clear general relation between duration of residence or date of 

 departure and remoteness of wintering ground. 



The late birds remain in the arctic until ice and snow are likely 

 to cover their feeding places. At this season they encounter cold 

 weather, but it is not as severe as near the common date of arrival 

 in spring. The majority of the birds leave in August when the mean 

 temperature is above freezing and while even frost is of brief duration. 

 Evidently cold sets the latest termination of the stay of migratory 

 birds in the Arctic, but many birds leave just after the peak of sum- 

 mer's heat and while the weather is still warm. As it appeared at 

 the time of arrival, the departure of migration is not in phase with 

 the temperature changes of the seasons. 



Thus, it can be seen that the departure schedule of the various 

 species differs as to terminal date and as to approximate duration 

 of the process of departure. Although estimates of the duration of 

 departure are far from precise, there are evident large differences 

 among the species. In no case is the departure process short enough 

 to suggest that all individuals of a species act in concert. Considering 

 the homogeneity to be expected in populations breeding in the same 

 area, the duration of both arrival and departure seems rather long. 

 Neither part of the migratory process suggests that it is sufficiently 

 shortened at Anaktuvuk to indicate a strong, general, synchronizing 

 influence from the shortness of the arctic season. 



