46 
The Food Relations of the Carabidce and Cocciuellidce. 
ordinary situations, consisted of animal food, of which a little less 
than half was insects. Fungi made thirteen per cent., and the 
remaining vegetable food was about equally divided between 
grasses and exogenous plants. Three specimens of H. caligino- 
sus and H. pennsylvanicus , taken among the canker-worms, had 
derived one-third of their food from those caterpillars, while the 
other two-thirds consisted of vegetation, sixteen per cent, being 
Peronospora, and the remainder chiefly seeds and exogenous 
# tissues. Four specimens of H. herbivagus , collected in the cab- 
bage field, in April, had eaten none of the cabbage-worms, and 
only ten per cent, of insects (Diptera). The remainder of the 
food consisted apparently of fragments of seeds, as indicated by 
the contents of the cells of the fragments and by other micro- 
scopic characters. A piece of the epidermis of grass was noticed 
in one of the beetles. Taking the genus Harpalus as a whole, as 
far as these nineteen specimens can be supposed to indicate its 
food, we find that only about one-eighth of it consisted of animal 
substances. Insects stand at nine per cent., two-thirds of them 
caterpillars, — ants and Diptera making up the balance. Among 
the items on the vegetable side of the account, we find fungi and 
pollen of Composite each eleven per cent, and seeds and other 
tissues of grasses, fourteen per cent. 
Genus Patrobus. 
Two specimens of P. longicornis , one from Central and the 
other from Southern Illinois, had eaten nearly twice as much 
vegetation as animal food. The latter consisted chiefly of cater- 
pillars, and included in fact nothing else but traces of plant-lice, 
eaten by one of the two. A little of the vegetation was derived 
from utass, but the source of the remainder could not be satisfac- 
torily traced. 
The Family as a Unit. 
We have now to treat the various collections of Carabidm upon 
which this paper is based, as distinct and unbroken groups, with- 
out reference to the genera of which they are composed. The 
eighty-three specimens of all the species obtained in miscellane- 
ous situations, are found to have derived forty-two per cent, of 
their food from the animal kingdom, while the seventy specimens 
