16 THE RELATION <>K BPABBOW8 TO AGRICULTURE. 
would eal on another farm where the food supply was identical. This 
line of research might be continued until it could be foretold with 
reasonable certainty which of the different objects in the accessible 
food supply of a locality a given bird would probably select. 
In recapitulation it may be stated that in the investigation of the 
food habits of any bird the lirst thing to be done is to examine enough 
stomachs to obtain a general idea of the bird's food, so that intelligent 
field work may be done. Then the observer should go to some favor- 
able spot, note carefully the different kinds of available food, watch 
the birds feeding for awhile, and collect stomachs for examination. 
In this way it will be possible to ascertain what a bird will eat, what 
it prefers, and what it will refuse. 
CLASSIFICATION OF BIRD FOOD. 
To aid in reaching final conclusions as to the economic position of 
a bird it has been found convenient to divide the food into three cate- 
gories according as its consumption- tends to produce a (1) beneficial, 
(2) injurious, or (.'J) inappreciable effect on agriculture. The bene- 
ficial part consists chiefly of insect pests and weed seed, the injurious 
part consists largely of insect enemies of insect pests and plunder 
from cultivated crops, and the neutral part comprises neutral insects 
and the fruits and seeds of plants of no economic importance. The 
relations which these three parts of the food bear to one another deter- 
mine in large measure the economic status of a bird. 
Of course, there can be no hard and fast rule in the matter. The 
beneficial and injurious elements of the food are marked by infinite 
degrees of gradation, and quality must be considered as well as quan- 
tity. The importance of the consumption of wheat varies greatly 
according to whether the grain is taken from the newly sown field, 
the growing crop, the stubble, or the shock. Insects may be either 
serious pests or insignificant in power to damage crops, with every 
shade of in juriousness between these extremes. 
The abundance of a bird has also much weight in fixing its value. 
A species must be numerous, must live among cultivated crops, and 
must take food that has a close connection with agriculture in order 
to produce any significant effect on the farm. 
The neutral element of the food must not be overlooked. The 
greater its proportion the more abundant must the bird be in order 
to produce any effect either one way or the other. With many shore 
and woodland birds it is so large, in comparison with the injurious 
or beneficial parts, that it is Likely that such birds have scarcely any 
effect in rural economy. 
These and many other like factors have to be considered in deciding 
a bird's economic status; but for a rough general estimate it is safe 
to assume that a bird that [\>vi\^ on insects, seeds, and fruit, and is 
abundant on a farm, will do more good than harm, and usually be 
