196 THE MIGRATIONS OF BIRDS 



few have been recorded in April in Haiti, apparently 

 in northward migration. We may suppose that 

 their winter home is somewhere in the northern part 

 of South America, though this remains to be defi- 

 nitely ascertained. Identification of swifts in 

 regions of tropical forest is difficult, as the birds fly 

 above the trees, where they are usually beyond gun- 

 shot. There are in the Tropics in addition, related 

 swifts, which can not be distinguished from our 

 northern species except with the birds close at hand, 

 which still further complicates the case. In the 

 Tropics there are few chimneys, so that we must 

 picture the chimney swift in winter returning to 

 former habits of roosting in hollow trees, as it did 

 universally before the advent of civilized man. For 

 six months of each year then it frequents settlements 

 in the north, where it nests and roosts in chimneys, 

 or, casually, in barns and outbuildings, while during 

 the winter season it must revert to the ancient cus- 

 toms of its race, truly a curious mixture of habit. 



Migrations among the strong-winged nighthawks 

 of North America are of considerable interest, since 

 these birds show a tendency to follow the general 

 direction of river valleys more closely than there 

 would seem to be need. Formerly these birds were 

 far more abundant than at present, and for many 

 years it was usual to shoot them in wanton sport, 

 especially during their autumn flights, a custom 



