LIFE HISTORIES OF NORTH AMERICAN DIVING BIRDS. 127 



disconnected and the edges ragged. All the birds taken on the nest, July 21, 

 were in that deplorable condition, only that the wings were spotted with slaty 

 black as the new coverts made their appearance now, the middle row being 

 complete already. Also the four or five inner primaries were shed, and the 

 new ones, in different stages of development — the three innermost full grown — 

 contrasted favorably against the faded-out remnants of the old ones. The tail 

 feathers are still unshed, but their condition plainly shows that they will 

 be molted before long. The ornamental feathers are worn down, the crests 

 are thin, and many of the long plumes have already disappeared. These 

 specimeiis prove beyond a possibility of doubt that the remiges and rectrices 

 are molted toward the end of the breeding season, and that the process com- 

 mences with the inner primaries. But not only are the wing feathers shed 

 now, but also the contour feathers ; all over the body protrude now the bluish 

 sheaths containing the new feathers, which in some places have already burst 

 through the tips. The postnuptial molt, therefore, is a complete one. At this 

 time the brightness of the bill has likewise faded away, the white tip gets 

 bluish, and the basal parts darken. The upper layers of the horny covering 

 scale off, but I feel satisfied that a regidar shedding of the basal parts, such as 

 in the Fraterculea\ does not take place. 



The complete postnuptial molt in adults seems to take place in 

 July and August, so that by September or earlier both old and 

 young birds have assumed the adult winter plumage and become 

 indistinguishable. Young birds in their first winter plumage, which 

 is worn for only about six months, show signs of developing white 

 plumes on the sides of the head, but they have no frontal crests. 

 In the adult winter plumage, which is practically only a fall plumage, 

 both the plumes and the crests are present, but are not so highly 

 developed as in the spring. The prenuptial molt during the winter 

 and spring is incomplete and the curious nuptial adornments which 

 it produces soon wear away. 



Food. — Judging from Doctor Stejneger's (1885) records of the 

 stomach contents of specimens examined, I should say that the princi- 

 pal food of the whiskered auklet is gammarids, with which many 

 of the stomachs or crops were filled; other amphipods were also 

 found and a few decapods and gastropods. Some of these animak 

 are probably found swimming on or near the surface of the sea, and 

 others must be sought for on or about sunken ledges. 



Winter. — Of the migrations and winter habits of the auklets we 

 know very little; probably they move out to sea and roam over the 

 open ocean. Doctor Stejneger (1885) writes: 



When the breeding season is over, they, like all the allied forms, I'etire to 

 the open ocean, part of them at least, going to more southerly latitudes to winter. 

 That many stay in the neighborhood of the islands is evident from the fact 

 that I obtained numerous specimens at Bering Island in December and Jan- 

 uary. A single female came near the coast on December 14, 1882, and was 

 shot ; but from the 29th of the same month until January 5, 1883, a few could 

 be met with every day. They could then be seen in small societies of two to 

 four, swimming along the rocky shores, alternately diving for food, which 



