THE AMERICAN ENTOMOLOGIST. 



13 



less probable, hut reqirtring verification 

 by farther stud)'. 



The most fruitful peculiarity of the 

 inclhod used was the careful estimate, for 

 each specimen (after a critical microscopi- 

 cal examination of the contents of the 

 .stomach), of the relative amounts of all 

 the elements of the food, and the subse- 

 ([uent averaging of these ratios for the 

 species. By this means I determined the 

 hitherto unsuspected fact that the family 

 is inordinately destructive to predaceous 

 beetles {Harpa/iiii), seven per cent, of the 

 food of the 150 specimens consisting of 

 these highly beneficial insects. When we 

 remember that one predaceous insect must 

 destroy many times its own bulk of other 

 insects during its life, we see the import- 

 ance of this fact in respect to the econom- 

 ical value of these birds. 



Between the Turdida and other families, 

 I can make only the following crude com- 

 parison in this particular. Of the 150 

 Thrushes examined, forty-six per cent. 

 had taken Carabidce, while of 194 birds of 

 other families in whose stomachs insects 

 were found, less than five per cent, had 

 eaten these Coleoptera. 



The worst sinner in this respect was the 

 Hermit thrush ; while the Alice thrush 

 and the Wood thrush had eaten compar- 

 atively few. Curiously, the ratio of Cara- 

 bid(B continued undiminished during the 

 fruit season, when the total of insect food 

 fell away very rapidly. For example, the 

 Cat-birds ate in May, June and July, 

 eighty-seven per cent., sixty-four per cent. 

 and eighteen /(V cent., respectively, of in- 

 sect food, while the Carabidce for those 

 months averaged seven per cent., six per 

 cent, and ten per cent., the corsesponding 

 fruit record standing nothing, thirty per 

 cent, and seventy-one /f/- a'///. 



The following genera were distinguished 

 among the Carahidce : Scarites, Dyschirius, 

 Platynus, Evarthrus, Pterostic/ius, Amara, 

 Brachylobus, Geopiniis, Agonoderiis, Aniso- 

 dactylus, Bradycellus, Harpaliis and Steno- 

 lophus. The absence of all, or nearly all, 

 the specially protected genera is notice- 

 able (unless the obscure color of many is 



reckoned a special protection). A single 

 Cicindela [C. lecontei) was found in the 

 stomach of a Cat-bird. 



I was further interested in the apparent 

 specific differences in the food of allied 

 species, occupying the same ground at the 

 same time and drawing their food from 

 the same sources of supply. 



The Robin and the Cat-bird differed ma- 

 terially in the number of ants and myria- 

 pods destroyed, the former eating very 

 few of either (one per cent, and two per 

 cent., respectively), and the latter many 

 (viz.: tew per cent, and six per cent). The 

 Brown thrush departs from all the other 

 members of the family in its fondness (?) 

 (perhaps it is stern necessity which forces 

 him to this miserable shift), for insects and 

 fragments of grain picked from the drop- 

 pings of stock. Twenty-eight per cent, of 

 the food of those shot in April was derived 

 from this source, and another eight per 

 cent, consisted of carrion beetles (6'/'^//zV/ia,'). 



This bird was further distinguished from 

 the Robin (as is the Cat-bird also), by the 

 absence of the larva of Bibio albipennis Say 

 (kindly determined for me by yourself*), 

 which made over half the food of the Robin 

 in March. It is important to recall, as 

 throwing light on the question of fixity of 

 food habits over large areas, that Prof. 

 Jenks, now of Brown University, found nine- 

 tenths of the food of a large number of 

 Robins whose stomachs were examined by 

 him in Massachusetts, in March and April, 

 1858, to consist of this same larva. 



The above particulars and conclusions 

 will serve to give some idea of the interest 

 and promise of this subject, if it is studied 

 with as near an approach to the strict scien- 

 tific method as the circumstances will permit. 



For full details of the observations thus 

 far made upon this family and a much 

 more elaborate discussion of them, you are 

 respectfully referred to the forthcoming 

 Vol. 13, Trans. 111. State Hort. Soc. 

 Very truly yours, 



S. A. Forbes. 



111. State Lab. Nat. Hist. 



Normal. III., Dec. ii, 1879. 



* These larvae had been previously determined as Tipulid 

 larvae by the entomologist of the Department of Agriculture. 

 — Ed. 



