THE AMERICAN ENTOMOLOGIST. 



217 



that all the localities and all parts of the 

 month were equally represented in each 

 group ; and then averaged each ten sep- 

 arately, and compared the averages. In 

 the first group beneficial insects composed 

 29 per cent, of the food, and injurious in- 

 sects 59 per cent. ; in the second group 

 beneficial insects composed 27 per cent, 

 of the food, and injurious insects 61 per 

 cent. The close correspondence of these 

 averages shows that, on this question, ten 

 specimens would have given as accurate in- 

 formation as twenty, and indicate that ten 

 birds a month will usually afford a fair basis 

 for an opinion. 



APRIL. 



The food of April, as shown by the thir- 

 teen specimens of that month (from Nor- 

 mal, Evanston, Waukegan, and Elizabeth, 

 in 1876 and 1880), was remarkable for the 

 number of ApJwdii (dung beetles) it in- 

 cluded ; 21 per cent, of the food of the 

 month was Aphodius inquinatus, 9 per cent. 

 A. fimetarius, and i per cent, undeter- 

 mined Aphodii. This peculiarity is ac- 

 counted for, in harmony with what has 

 been said above respecting the feeding 

 habits of the Bluebird, by the fact that 

 this is the month when the Aphodii fly 

 most actively in the latitude of Northern 

 Illinois. CarabidcB now stand at 8 per 

 cent., including Carabus pahistris, Pteros- 

 tichus, Evarfhrus, and other Fterostichi, 

 Fla/ynus, C/ilcenius tomentosus, Anisodac- 

 tylus rustic us, Ainphasia intcrstitialis, and 

 Harpalus ; 4 per cent, of Hemiptera in- 

 clude Coriscus* and Hymenarcys nervosa, 

 while spiders rise to 9 per cent. Cater- 

 pillars are 21 per cent. (17 per cent. Noc- 

 tuids), June beetles [Phyllophagd) 2 per 

 cent., Curculionidx i per cent., and grass- 

 hoppers {Teiiigidea sp. and Tetiix ornatd) 

 8 per cent. \ a total of ^2 per cent, of in- 

 jurious insects against 21 per cent, of pre- 

 daceous species. Among the neutral ele- 

 ments we find a sprinkling of ants (2 per 

 cent.), larvae of a Tenebrionid {Meracantha 

 contracta\) 4 per cent., and thousand-legs 



* Kindly identified for me by Mr. Uhler. 



t For the determination of this species and most of the 

 other larvae which have been identified specifically, I am 

 under obligations to Prof. Riley, without whose most gene- 

 rous assistance this paper would have lacked much of what- 

 ever value it may be supposed to have. 



lulidiz I per cent. Long strips of grass, 

 in pieces much too large to have been 

 eaten by any of the insects present, were 

 found in the stomachs of two of these 

 birds, and also occurred during each of 

 the three following months. I am in 

 doubt whether these were taken as food ; 

 but, since I have found them in no other 

 bird, and since a species which feeds so 

 largely on cut-worms and grasshoppers 

 may have acquired the power of digesting 

 the very considerable quantities of grass 

 contained in the intestines of these insects, 

 I have thought it best to include them in 

 the percentages of food. It is probable, 

 however, that they were swallowed acci- 

 dentally with insects taken from the ground. 

 It will be noticed that the excess of Co- 

 leoptera in April is largely compensated 

 by the diminished quantities of orthoptera 

 and caterpillars. 



MAY. 



In this month nine birds were taken, 

 from six localities in central and northern 

 Illinois, in 1876-80. The Lepidoptera, 

 Coleoptera, and Orthoptera return to 

 about their normal ratios, but spiders rise 

 to the excessive figure of 21 per cent. 

 This ratio is, however, partly misleading, 

 as, although six of the nine birds had 

 eaten spiders, yet 1 1 per cent, is due to a 

 single bird, which had eaten nothing else. 

 In such a case a larger number of speci- 

 mens is required to restore the balance, so 

 violently disturbed. Two birds of this 

 month had eaten moths, and five had 

 eaten cut-worms. The averages stand 55 

 per cent, of moths, caterpillars, June bee- 

 tles, curculios and orthoptera, opposed to 

 35 per cent, of Carabida, soldier bugs and 

 spiders. The Carabid(c include Crata- 

 canthus dubius, Agonoderus comma, Aniso- 

 dactylus, and Harpalus. Other details may 

 be obtained from the table at the close of 

 this paper. 



JUNE. 



In June, ten birds — one from Mount 

 Carroll, the others from Normal — had 

 taken a somewhat unusual diet. The ratio 

 of spiders (18 per cent.) falls little short 

 of that for May ; but an examination of 



