94 
The Food of Birds. 
for the solution of a few separate questions. A far more 
useful method, and, in fact, the usual one, is that of 
watching birds while taking their natural food in the free 
state. Now and then a fact may be learned in this way 
which would escape detection in any other — such as the 
perforation of cocoons of Cecropia by the downy wood- 
pecker reported by F. M. Webster* — but usually this 
method is of wholly secondary usefulness. The difficulty 
is very great of telling with certainty, in the great 
majority of cases, just what a bird is eating, even 
if one watches it with a glass. The notion of 
the food resulting must be distorted, as the species will 
be seen much more frequently and clearly in some of its 
haunts than in others. It is impossible by the use of this 
method, even to guess intelligently at the ratios of the 
different elements of the food — a matter of the first im- 
portance to an understanding of the subject. It yields 
very few facts for the time expended, and these, in nearly 
every instance, could have been learned in much less 
time, with far greater certainty, and in far greater detail 
by the following method. Finally, it affords no means of 
reviewing observations, but the impressions received 
from the hasty and imperfect glance of a moment must 
either be rejected wholly or must stand as verified ob- 
servations. 
By the third method, however, that of examining the 
contents of the stomachs after death, each bird usually 
affords a large number of objects which can be studied 
critically, and in detail, and can be indefinitely pre- 
served for reference. These objects give a nearly or quite 
complete and impartial record of the food for some hours 
past — those elements taken in a thicket or a tree-top be- 
ing as evident as those taken on open ground. They are 
usually identifiable by the skilled student. Even very 
minute fragments will tell as much as the out-of-door ob- 
server can learn under the most favorable circumstances. 
Tn the great majority of cases it is possible so far to fix 
the kinds of food as to bring every element clearly into 
one of the three classes, beneficial, injurious or neutral. 
* In an unpublished paper read at the meeting of the Illinois State Nat. 
Hist. Soc., at Bloomington, Feb. 1880. 
