106 
It follows, of course, from the above, as a general rule, that every 
natural agency originating or stimulating oscillations of numbers 
among insects is to that. extent an injury, and every agency tending 
to prevent such oscillations, or to limit and reduce them aftei they 
have arisen, is a benefit in that particular. In fact, so overshadowing 
is the importance of insect injuries due to what we may call a disturb¬ 
ance of the balance of plant, and insect life, that the point oi view 
from which all natural checks on insect multiplication should be con¬ 
sidered is that of their effect on such disturbances. Concerning a 
predaceous or parasitic insect, an insectivorous bird, or a parasitic 
plant, the main question of interest to the economic entomologist is, 
what is its effect on insect oscillations? 
In truth, however desirable total extermination of any insect may 
be, it is evident that we cannot expect this result from the depre¬ 
dations of those of its enemies which are dependent upon it for 
food. The adjustments of nature are not so clumsily made. The 
best that we can expect from any predaceous or parasitic organism, 
is that it shall hold the species which it infests, or upon which it 
preys, steadily down to a fair average number. 
Concerning every such organism, we have therefore three ques¬ 
tions to ask: 
1. Does this bird, insect or plant originate any oscillations among 
the species of insects which it affects? That is, are its numbers or 
habits so inconstant from year to year, that insects which are at 
one time vigorously attacked by it, are at other times relatively free 
from its injuries, and allowed to multiply without restraint? 
2. Does it prevent or restrain any oscillations of insects now noxi¬ 
ous, or capable of becoming so, if permitted to increase more freely? 
That is, does it bring to bear upon any species a constant pressure 
so great, that those insects would increase unduly, if this pressure 
were removed by the destruction of this enemy? 
3. Does it do anything to reduce existing oscillations of injurious 
insects? Does it sometimes vary, either in numbers or habits, in 
such a way as to affect injuriously to an extraordinary degree those 
species which for any reason become superabundant for a time? 
When these questions are answered for any beneficial species, or one 
whose economic relations are in doubt, we shall be able to estimate 
intelligently its usefulness, while without this such an estimate will 
evidently be impracticable. 
The present paper is an attempt to answer these questions, in 
part, with reference to some of the most important families of car¬ 
nivorous insects. 
The two most important families of insectivorous beetles, are the 
ground beetles (Carabidee) and the ladybugs (Coccinellidse), the 
latter noted for their destruction of plant-lice, and the former mak¬ 
ing a variety of insects their prey, and feeding, also, to some extent, 
upon vegetation. 
The view of the functions of these two families which is common 
among entomologists, is certainly based upon insufficient data. 
Observations of the food of these beetles have hitherto been left 
almost wholly to chance, and have nowhere been systematically pur* 
