ROUTES OF MIGRATION. 99 



enters by such channels. Bad weather occasionally 

 keeps all birds close to the coast-line, and thus 

 modifies their direct line of flight. The coast 

 migration of Eastern America, from what evidence 

 I can glean on the subject, appears to cross from 

 Florida to the West Indies, rather than follow the 

 continental coast of the Gulf of Mexico to South 

 America. I might give many instances of birds 

 •on their way south from the Eastern States to South 

 America that cross the West Indies and Trinidad. 

 Another very interesting fact connected with coast 

 migration, is the remarkable way in which certain 

 birds suddenly alter their course by leaving a coast 

 and starting directly across the sea. These birds 

 are supposed to be following old migration routes — 

 ancient coast-lines now submerged beneath the 

 waves. The known great attachment of birds to 

 their fly-lines makes this explanation feasible, and 

 it is still further confirmed by direct geological 

 evidence. There is some evidence, for instance, 

 that birds follow an ancient coast-line once reaching 

 from Spurn Point in Yorkshire to Denmark or 

 Holland, in the fact that several species are known 

 to migrate along the East coast of England up to 

 this point, and then to strike across the sea, seldom 

 •or never being observed further north : the Knot 

 {Tringa caniitus), the Bar-tailed Godwit {Liinosa 

 rufa), and the Gray Plover [Charadrius he/veticus), 

 may be mentioned as examples. Another ancient 

 coast-line followed by the Knot, the Asiatic Golden 

 Plover {Charadnns fn/vKs), and the eastern form of 



