THE FOOD OF THE ROBIN. 119 
“Tf we compare the corn plant-louse, the gall stage of the grape phylloxera, the 
plum curculio, the small parasitic military microgaster, which lays its eggs in several 
kinds of cut-worms, the potato beetle and the chinch-bug with the large coral-winged 
grasshopper, bulk for bulk, the ratios will appear about as follows: 
1 Coral-winged grasshopper==12,000 military microgasters. 
“c 
1 * =3,000 phylloxera. 
1 * - ==1,500 corn-plant lice. 
1 hs «  ==750 chinch bugs. 
I te =60 plum curculios. 
I - bs =7 potato beetles. 
| _ s 1,000 young potato beetles. 
“ By a system of gauging bulk for bulk it is evident from the table that one coral- 
winged grasshopper eaten by a bird would give it a credit which would affect completely 
the destruction of 12,000 military microgasters, a proposition sufficiently absurd.” 
Having seen from the start that the ratios of the different food ma- 
terials could not justly be estimated according to bulk, and having seen 
also that a system based upon the number of insects, plant fruits, etc., 
found in the stomachs examined would be almost equally likely to intro- 
duce error, and that it would be a system particularly difficult to carry 
cut in consequence of the fragmentary condition of the food, I decided to 
combine these two systems of computing the proportions in a way which 
seems to me to represent justly all the elements of food. It would be ap- 
proximately true to say that 1 have estimated the proportion of animal 
food according to the number of the individuals, and vegetal food ac- 
cording to bulk. But all fruits which have a definite number of seeds 
have been estimated upon a numerical basis. It is evident that this 
would have been very difficult or even impossible in the case of black- 
berries or raspberries in which the number of seeds is so variable, 
It may be objected that the computation of the vegetable food on one 
basis and of the animal food upon another basis is a fruitful source of 
error. But I have exercised all care and diligence to avoid every possibility 
of error, and, in fact, an estimation of the relative proportions of the 
several kinds of food would not make the vegetal part appear larger 
than it really is, since a raspberry or blackberry is no greater in bulk than 
an earthworm or May beetle. It may as well be admitted that, in the 
present state of knowledge, only an approximation to the truth can be at- 
tained in a statement of the relative proportion of the various food ma- 
terials in a bird diet. 
But even after we have tabulated the numerous articles of food in 
their differing proportions in a more or less satisfactory manner, the task 
is by no means completed. In order that we may decide whether the 
robin is on the whole a benefit or an injury to farmers and gardeners, we 
must first determine the economic relations of the various species of plants 
