70 NORTH AMERICAN FAUNA 61, FISH 4ND WILDLIFE SERVICE 



obtained two specimens at Unalaska on June 20 and on August 

 7, 1932. These geese are known to winter on the following 

 islands: Unimak, Unalaska, Sanak, Umnak, Amukta, Seguam, 

 Atka, Adak, Tanaga, Kanaga (abundant), Amchitka, Ulak (longi- 

 tude 178 c W.), Ogliuga, Kavalga, Semisopochnoi, Kiska, and 

 Attu. The chief of Attu declared that they were in that locality in 

 "millions." These are the islands on which we have specific 

 information. Undoubtedly, emperor geese occur on many, if not 

 all, of the other islands; almost certainly they occur on Agattu 

 and Semichi, for example. 



As may be expected, there are many records of winter oc- 

 currences farther south, in Washington, Oregon, and California. 

 These records are numerous enough to suggest that some strag- 

 glers find their way into those southern localities quite regularly ; 

 however, the regular wintering area is confined to portions of 

 Alaska Peninsula, the Shumagin and Sanak Islands, and the 

 Aleutian chain. Apparently, they are rare on the west side of 

 Bering Sea during the winter. Stejneger (1887) records two 

 specimens taken at Bering Island, April 6, 1886. 



The spring migration varies according to the locality and 

 the age class. Natives declare that emperor geese leave Attu 

 Island in April; Turner (1886) gave the date as the "latter 

 part of March." He also stated that after the middle of Apri* 

 considerable numbers of geese begin to arrive on the north side 

 of Alaska Peninsula, particularly in the neighborhood of Ugashik. 



In 1924, I observed the spring migration at the nesting grounds 

 at Hooper Bay. The first migratory wave began about the 

 middle of May and continued to the end of the month. There 

 was another notable flight about June 5 and 6, which appeared to 

 end the migration of breeding birds. Nesting had begun at that 

 time. 



A second distinct migration at Hooper Bay took place from 

 June 21 to July 1. These were immature birds, probably all 

 nonbreeders. 



It was my good fortune to observe the other end of such 

 migration in 1925, at Izembek Bay and Unimak Island. On 

 April 29, 1925, and for several days following, flocks of emperor 

 geese were noted at Urilia Bay, on the north side of Unimak 

 Island, many of them flying northeastward. On May 17, a 

 flock of 250 was seen standing on an exposed sand bar in St. 

 Catherine Cove. On May 20, they were common in Izembek Bay, 

 and Donald Stevenson noted a flock of 300 there on June 2. We 

 saw a similar-sized flock on June 8, at Moffet Cove, where they 



