426 MIGRATION 



in the twenty-four hours, they overtook and passed the 

 thaw, and finding no food had to turn back. The 

 records of migration which have been kept on the British 

 coast seem also to show that similar blunders are com- 

 mitted in autumn, and that many birds which ought tO' 

 reach our northern and eastern shores have apparently 

 in like manner overshot the mark, and have had to turn 

 back, some from the sea and others from the continent, 

 and consequently arrive on our western or southern 

 shores. 



It has been remarked in this country that migration 

 takes place in autumn in greater flocks or "rushes " than 

 in spring. This is probably caused by the birds lingering 

 at some favourite feeding-grounds, and accumulating in 

 increasing numbers until a sudden frost warns them that 

 they are overstaying their time, and they " rush " off en 

 masse, helter-skelter, for summer climes. A somewhat 

 similar accumulation of birds apparently takes place on 

 the skirts of the frost in spring, for when the ice broke 

 up we had a "rush" of various sorts of birds, which 

 suddenly swarmed on all sides. 



In the valley of the Yenesei the stream of migration 

 follows the course of the river from north to south, instead 

 of from east to west as at Heligoland. Very few, if 

 any, birds appear to cross the deserts of Mongolia. In 

 South Siberia the stream of migration divides, part of the 

 birds probably following the Angora, and part the smaller 

 stream which retains the name of Yenesei. Among the 

 birds which take the eastern route are the yellow-browed 

 warbler, the Arctic warbler, Blyth's grass-warbler, the 

 pintailed snipe, the Petchora pipit, and many other birds ; 

 whilst amongst those which appear only to take the 

 western route are the willow-warbler, the sedge-warbler, 

 the great snipe, the fieldfare, and many others. Occa- 



