56 The Food /tflafioiis of tin- (\ir(ihid<r and Cocci iirtlidir. 



especially as no coniparisoii wliatever was made ol' the tw(i 

 sets of data until tlie tables were eoiiipletecl in thcii- present form. 

 When, therefore, vvti hnd that the one hundred and seventy-five 

 specimens of the present paper, belonging to thirty-fight species, 

 were estimated to have taken fifty-seven per cent, of animal food 

 and thirty-six of insects, and that the ratios of cryptogams, gram- 

 inaceous plants and exogens are respectively five, eleven, and five, 

 we must conclude that those figui-es are a fair average of the 

 ordinary food of the family. 



Rki>atioxs to Birds. 



The forecroing pages have set forth the relations of the C'arab- 

 id;e and the Coccinellid;e to the species upon which th^y feed, 

 and a few general statements will now be proper coiu;erning the ani- 

 mals which prey upon them in turn. Predaceous groinid-beetles 

 are peculiarly exposed to lairds wliich commonly seek their food upon 

 the grovnid, and we need not be surprised to find that they enter 

 largely into the food of such species as the thrushc^s and the blue- 

 bird. Carabida^ were found to furnish about five or six per cent, 

 of the food of four hundred and twenty-three specmiens of these 

 birds, as stated in a paper on that subject in the third Bulletin (^f 

 this series, Imt Coccinellid.'B did not occur at all. Indeed, in the 

 food of more than four hundred other birds, of various families, 

 Coccinellidie were found only in Regulus, Avhere a single species 

 was reckoned at one per cent, of the food. 



The o-reat difterences in tlie food of the Carabida', disclosed by 

 tliis paT)er, give considerable importance to the questif)n of the 

 kinds of these beetles most freely eaten by birds, and tlie follow- 

 ing list of species and genera recognized in the food of the col- 

 lection of thrushes and bluebirds above mentioned is given as an 

 answer. 



It will be seen that there is a very wide difference between the 

 number of Carabidie proper taken by these l)irds, and the number 

 of Harpalida% I'epresentatives of the former group occurring in 

 only six specimens, and of the latter in one hundred and sixteen. 

 On the other liand, fifty-nine of the birds had taken Harpalids 

 which may l)e fairly classed with the second group established in 

 this paper, and fifty-seven had taken those belonging to the third 

 group, or pliytophagous Carabida-. The genera most preyed 



