28 COLLETT AND NANSEN. ACCOUNT OF THE BIRDS. [NORW. POL. EXP. 



As long as the travellers were in the drifting ice itself, far from the 

 land, the roseate gulls could not be said to be numerous. They appeared, 

 as a rule, only singly or in pairs, seldom several together. On July 29th, 

 for instance, they were seen altogether four times, but only single birds; on 

 the 30th (one pair in the morning and one pair in the afternoon) and 31st 

 July, and the 3rd August, only single, full-coloured pairs were seen each time. 



Only off Hvidtenland did they become at all numerous. On the 7th and 

 8th August, they were seen repeatedly (near Eva and Liv Islands), whole flocks 

 of them flying about or sitting on the edge of the glacier, which was about 

 50 or 60 feet above the sea. Young and old were together. Whether 

 there was any nesting-place in the neighbourhood, however, could not be 

 ascertained. 



Eva Island is completely covered with glacier, and does not appear to 

 have a single patch of bare ground. From the top of Adelaide Island 

 (August 9th), a strip of low, bare ground could be discerned on the NW 

 coast of Liv Island, possibly an old beach-line, about 20 or 30 feet high, 

 and perhaps 1 kilometre long, apparently forming a low beach, partly covered 

 with rock debris. With this exception, Liv Island was also snow-white, and 

 covered with a dome-shaped ice-cap sloping down into the sea on all sides, 

 without any abrupt glacier edge. Of all the land that was in sight, this was 

 the only spot free from snow, where birds could possibly breed; but whether 

 the locality would be considered suitable as a nesting-place for this species, 

 and whether it was large enough, is doubtful, although it seemed to resemble 

 the places where nests of the ivory gull have been found. 



The last specimens were seen near the Coburg Islands between the llth 

 and 14th August. After that they vanished completely, and no specimen 

 was seen during the rest of the journey; and there was thus no sign of 

 their inhabiting any locality in the region between Dickson Sound and 

 Cape Flora. 



In their movements they recalled Rissa tridadyla. They appeared to 

 find their food exclusively among the pelagic organisms on the surface of 

 the water. They never came near the provisions (like Pagophila eburnea 

 and Larus glaucus) to take any of them. They generally flew rather low, 

 and were on the wing night and day. They often flew close to the kayaks, 

 but were never seen swimming. They were extremely graceful in their move- 



