34 
The Food Relations of the Carabidce and Coccinellidce. 
entomologists, is largely due to a hasty generalization, based upoii 
insufficient data. Observations of the food of these beetles have 
hitherto been left almost wholly to chance, and have nowhere 
been systematically pursued — from which it has resulted that we 
know their habits only in the most conspicuous situations, and 
have not a fair idea of the general average of their food. 
Neither have observations of any kind been numerous enough to 
enable us to detect clearly differences of food habit in different 
species or genera of these families; but, with slight occasional 
exceptions, all Carabidse and Coccinellidse have been classed 
together as essentially carnivorous. 
Besides insufficient observation, a tendency to reason too confi- 
dently from structure to function is responsible for many mistaken 
notions — a tendency particularly liable to mislead when applied 
to the habits of animals. It is frequently assumed that the most 
prominent and peculiar adaptive structures are necessarily indica- 
tive of the most important and customary habits, and that struct- 
ures especially fitted for one function are thereby incapacitated for 
every other. 
The first of these assumptions ignores the fact that many 
adaptive structures are acquired for the sake of the advantage 
derived, not in ordinary, but in extraordinary circumstances. The 
struggle for existence is one of greatly varying intensity, and the 
really decisive moments of the conflict are often only brief and 
occasional. The time spent in actual combat by very belligerent 
and very powerful animals, is doubtless but a small fraction of 
their whole lives; and yet by far the most prominent and import- 
ant of the structural peculiarities which serve to distinguish them 
from their more peaceful allies, may be those which enable them 
to triumph in these occasional but critical instants. Likewise the 
pinch of starvation must commonly be felt only at rare intervals, but 
no structures will be more thoroughly elaborated or carefully pre- 
served than those serving to give the animal the advantage 
during these brief periods, since the continued existence of the 
species depends on these no less than on those of constant use. 
From the prominent adaptive structures we may safely infer, *as a 
general rule, what the animal will do in the stress of a life and 
death struggle, but not necessarily what are its ordinary practices. 
The second of the above assumptions is also negatived, occa- 
