FAUNA OF THE ALEUTIAN ISLANDS AND ALASKA PENINSULA 163 



known that the eastern glaucous gull nests on St. Matthew and 

 Walrus Islands, we would expect it to visit the Aleutian district 

 at times, though the western glaucous gull would be most pre- 

 valent. Oberholser (1918) also mentions Diomede Islands for 

 this form. 



Larus hyperboreus barrovianus 



Though the status of barrovianus has been belabored repeatedly 

 by able ornithologists, certain specimens obtained in the Aleutians 

 induced me to examine the whole question again. More than 200 

 specimens were examined in the U. S. National Museum and in 

 the American Museum of Natural History in New York. 



It is useless to deny the difficulties in recognizing barrovianus 

 as a distinct form. In studying series from a given locality, one 

 is confronted with specimens that do not fit a given description. 

 Gulls are variable, and one must be cautious in arriving at con- 

 clusions. On the other hand, if one is careful to give due weight 

 to breeding territory, and to allow for migration to explain cer- 

 tain irregularities, many of these difficulties disappear. 



All gulls of the species hyperboreus are pale mantled, but true 

 hyperboreus is noticeably paler than barrovianus. Furthermore, 

 hyperboreus is definitely larger and has a decidedly larger and 

 more massive beak. Listed measurements do not adequately ex- 

 press the difference. Depth of bill of the two forms overlaps, or 

 meets, at about 23 mm., though most of them are above, or below, 

 this figure, and a difference of even 2 mm. makes a considerable 

 difference in appearance. 



A good series of specimens from Point Barrow and the east 

 shore of Bering Sea are remarkably uniform in the characters 

 assigned to barrovianus — the darker mantle, the smaller size, 

 and especially the smaller bill. Available specimens from eastern 

 North America are confusing, but it is notable that when winter 

 specimens are eliminated, and apparent breeding birds are used, 

 they fall more generally into the group of hyperboreus. This was 

 especially true of Greenland, where a good series of breeding 

 birds presented a clear picture of Larus h. hyperboreus, as here 

 described. 



The confusing aspect of the distribution of these two forms is 

 the considerable number of small-sized birds found along the 

 Atlantic coast in winter, which apparently agree with barro- 

 vianus, but which are far from the type locality. Possibly we 

 should expect this. Oberholser gave the breeding range as 

 extending along the Arctic coast as far east as "the territories of 



