THE BIRDS OF THE PRIBILOF ISLANDS. 371 



justified in tlie belief that migration does take place across these extensive wastes of 

 waters. The barred-tailed godwit [Limosa Icqjjpoiiica haxieri) ' is another example of a 

 Pacific migrant. They pass through the Aleutian Islands in spring in small flocks, 

 reach St. Michaels before the end of May, and breed along that coast of Bering Sea. 

 They return in the fall, pass through the Pribilofs and the Aleutians, and winter on 

 the islands of Polynesia and Southeastern Asia, Australia, and New Zealand. They 

 are absent from North America during our winter, there being but one record — in 

 Lower California.^ Their course is probably through the Aleutian chain, on many 

 islands of which they have been taken, thence to the Japanese coast, aud southward. 

 Other species of Asiatic wintering birds occur, breeding in Alaska; also indiviauals 

 of eastern American species have been taken with them on the islands of Bering Sea, 

 thus pointing conclusively to a migration route over the western Pacific Ocean. An- 

 other group of migration movements from the Alaskan breeding grounds is shown in 

 such species as the Pribilof sandpiper [T. ptilocnemis), and the emperor goose (P. cana- 

 gica) and many other species. These winter on the shores of the southern islands of 

 the Aleutian chain and along the Aliaska coast southward. 



The fact that many species of Alaskan birds boldly launch themselves into the 

 wide expanse of ocean between their summer and winter habitats naturally leads us 

 to expect that at sometime in the remote geological past their ancestors had a more 

 happy course over contiguous land areas which have since been submerged. No other 

 solution seems possible, and many probabilities point to such a conclusion in spite of 

 the fact now of considerable deep water intervening. Birds have no inclination to 

 explore unknown regions lyiug at such great distances apart. Their movements must 

 necessarily have begun gradually over contiguous or narrowly separated areas which 

 have been widened during the operation of geologic changes. The Pribilof group is 

 some 250 miles from the nearest Alaskan coast; they are nearly 200 miles north of the 

 Aleutian chain. They are 200 miles south of the next island to the northward, St. 

 Matthew. The Commander Islands lie 750 miles westward. The islands of the Aleutian 

 chain are generally visible from each other in clear weather, but gaps of 40 to 60 miles 

 are frequent. Attn, the most eastern of the chain, is about 180 miles from the nearest 

 of the Commander Islands and about 500 from the nearest coast of Asia, Cape Shi- 

 punski in Kamchatka. California is some 2,000 miles southeastward of Unalaska, 

 and Sitka is about the same distance almost directly eastward. The Hawaiian 

 Islands are also 2,000 miles directly southward from Unalaska, without intervening 

 land. Bird migration between the the Pribilofs and the Aleutians aud between the 

 Aleutians and Asia, Polynesia, and the Hawaiian group undoubtedly occurs with 

 many species and in enormous numbers. 



It has often been asserted that migrating birds take advantage of geographical 

 objects, as mountains, valleys, rivers, etc., in directing their course during their long 

 journeys from their summer to their winter habitat. However plausible this may be 

 as applied to land-migrating birds it fails completely upon consideration as a factor 

 of the movements of the Pacific birds noted above. I am not even sure that it is a 

 necessary factor even to land-migrating birds, but however that may be it most 

 certainly can not be considered when one is endeavoring to account for causes that 

 enable these birds to continue in a direct line; for instance, from the Aleutian to the 



■' ^^Limosa novce-zealandica. 



2 La Paz, No. 86418, U.S.N.M., 1882, L. fielding (iu spring, head). 



