248 FRANCIS H. HERRICK 



for their exclusive use, they simply go where food abounds, 

 and when not deterred by fear, fraternize with man to some 

 extent. 



The swift inherits the tendency to nest in a dark and cavernous 

 place with free entrance and exit, and to its mind the chimney 

 is a hollow tree with better ventilation in summer, but with 

 poorer protection from rain than that which the forest supplies ; 

 but so far as the immediate environment goes its instinct is 

 not quite precise or uniform, for the interior of a barn or of any 

 dimly lighted building, where no interference is offered, may be 

 chosen upon occasion. To conclude, so far as nesting and general 

 habits go both swift and eaves swallow are remarkably constant, 

 and far more so than either the robin or the song sparrow. 



Certain phases of the question of position, especially in refer- 

 ence to the immediate environment will be considered at this 

 point. 



Uniformity in the selection of nesting site. That certain birds 

 or their young, as in the case of the flicker or redheaded wood- 

 pecker, often return to the same nest site, — to the ancestral 

 tree, bird house or box, — is adequately accounted for on the 

 principle of association, and the proved tendency of the young 

 to return to the place of birth. On the other hand certain species 

 of birds, like the orioles and vireos choose their nesting sites 

 with great uniformity ; in our own minds we invariably associate 

 the Baltimore oriole with the elm, the red-eyed vireo with 

 maples, sapling pines with black-billed cuckoos, and tall ever- 

 greens with crows and many of the hawks. As in the other 

 cases considered so far as the birds are concerned, the associa- 

 tion is primarily due to the supports furnished by such trees in 

 relation to the food and the habits of the species in question. 

 The problem is precisely the same as in the robin, but with the 

 field of choice greatly restricted. While I have seen hundreds 

 of robins' nests in the apple, maple and elm trees, the number 

 found in the poplar (Populus tremuloides) , and gray birch {Betula 

 alba) during a period of over twenty years, is not more than one 

 or two in each instance, a fact which we should attribute to the 

 lack of proper supports which they furnish, when the require- 

 ments of cover and other needs are completely met. The build- 

 ing robin makes an equally good choice, as we have seen, whether 

 it takes to the pine, maple, apple, or oak, but the response of 



