586 Mr. A. Trevor-Batty e on the 



King-Eiders flew with each low tide from the sea to the 

 marshes at the head of Advent Bay, where they fed among 

 the Common Eiders, who also resorted there. If disturbed 

 they would fly round mixed in a .flock with the Common 

 Eiders, but would presently separate and settle by them- 

 selves. On June 21st Studley and I walked a long way up 

 the south shore of Advent Bay till we reached the marshes at 

 the head. We saw there '^many hundreds of Common 

 Eiders and a fair sprinkling of King-Eiders " ; perhaps we 

 saw a hundred King-Eiders in all. They were hard to 

 approacli because of the creeks, but I missed badly a splendid 

 old drake which flew round within shot while I was trying to 

 balance myself on the only sound lump of grass in a bad 

 place. On June 25th Studley killed a drake King-Eider off' 

 the sea in beautiful plumage, and on the same day I killed a 

 duck of this species. I find no entries about the species 

 after June 28th. 



10. Harelda GLACiALis (Liun.) . Long-tailed Duck. 



Reference to the writings of other observers goes to show 

 that this species is more abundant in Spitsbergen than my 

 own observation has led me to suppose. Von Heuglin 

 says : ''We have seen it on Dun Island, Ice Island, and Ice 

 Fjord." He adds that ''it frequents the rocks of fresh- 

 water lakes, and flies in dense flocks, chiefly of males, low 

 over the sea.^^ The first remark is probably a general refer- 

 ence to the habits of the species ; and the last no doubt refers 

 to Malmgren's observation, in 1864, of a family-party of five 

 " on a small pool of fresh water on one of the Horn Sound 

 Islands." Malragren reports four others, including a pair 

 as far north as 80°, of which the male went to the Stockholm 

 Museum. This I take from Prof. Newton's paper (Ibis, 

 1865, p. 515). Mr. Cocks says (Zool. 1884, p. 15) : " Three 

 ducks flying in Recherche Bay on Sept. 22nd were without 

 much doubt of this species, but they did not come close 

 enough to identify with certainty." Mr. Eaton says (Zool. 

 1874, p. 3816): "This duck occurred in King's, Wijde, 

 Neurenberg, and Lomme Bays. In the first of these localities 



