COMMON SCOTER. 603 



the Black and Caspian Seas, but in neither of these localities has it been 

 obtained by modern collectors. It is probable that Pallas was correct, as 

 Bogdanow records it on migration from the valleys of the Kama and the 

 Volga. 



The American form of the Common Scoter breeds in the Kurile Islands 

 and across Arctic America to Hudson's Bay. It winters in Japan, on the 

 Pacific coast of North America down to Southern CaliforniaJ the Great 

 Lakes, and on the Atlantic coast as far south as the Gulf of Mexico. 



There are few Ducks more exclusively marine in their habits, or more 

 uniformly gregarious, than the Common Scoter. In winter they are gene- 

 rally seen in very large flocks, which rarely if ever wander inland from the 

 shore. At their breeding-grounds in the Arctic regions they seldom ascend 

 the rivers far from the ocean ; but many of them reach their summer- 

 quarters by migrating across country, following for the mest part the 

 great river- valleys. On the tundra they are seen in pairs on the lakes as 

 soon as they arrive, but they are so very wary that few Ducks are more 

 difficult to shoot. Either some of them must breed very far north, or, 

 what is more probable, few of them breed during their first spring; for 

 large flocks are to be seen during the whole of the short summer at the 

 mouths of the great rivers, on the banks of which other individuals are 

 busily occupied with the duties of incubation. In the valley of the 

 Petchora flocks of Black Scoters were seen flying north down the river 

 long after other Ducks had eggs. In the middle of July we were lying at 

 anchor in the lagoon of the river waiting for the disappearance of the fog 

 which had come down from the arctic ice and concealed the Golievsky 

 Islands, which divide the lagoon from the ocean. The sun was shining 

 brilliantly overhead, and when the fog lifted, the island was revealed close 

 to us, with a flock of ten thousand Black Ducks circling in a cloud over it. 

 It seems scarcely possible that these were all males whose mates were 

 scattered on the nests over the tundra. More probably they were the accu- 

 mulation of the late flocks that we had seen migrating down the river, and 

 which most likely consisted of the previous year's birds, not yet adult 

 enough to breed. They appear to arrive at their summer-quarters very 

 late, and to leave them again very early, probably before the autumn 

 moult takes place. If this be so, it will explain the statement of Naumann 

 that the adult males arrive in the Baltic in August, but the young not until 

 two months later, and also that of Cecil Smith, who found them on the 

 Devonshire coast moulting their quills and unable to fly in the middle of 

 November. 



Kriiper found the Common Scoter in Iceland, breeding in small numbers 

 on the islands in the Myvatn See or Musquito Lake, and more abundantly in 

 the willow-scrub on the mainland. In the valley of the Petchora we never 

 found the nest on the islands in the delta, but either near a lake on the 



