RELATIONS OP BIRDS TO PARASITIC INSECTS. 87 



up the order Diptera, to which belong the tachiiia parasites (such as 

 Avere found in the stomach of one bhie grosbeak), the data now avail- 

 able show that probably less than 3 percent of the food of birds in 

 general consists of members of this order. And there is no reason to 

 believe that parasitic forms of this or other orders are specially sought 

 for by birds. Since the dipterous parasites are confined to three or 

 four families of the three score in the United States, it is not likely 

 that the part of the 3 percent of bird food they furnish results in 

 much injury to agriculture. 



Turning to that most important group, the Hymenoptera, it is to 

 be noted that many of them are small, some being extremely minute 

 (as is the case with a large group of the most effective parasites, the 

 chalcis flies), and these are very rarely found in bird stomachs. A 

 few families of Hymenoptera (Chrysididse, Trigonalidse, Sapygidse, 

 and Mutillida^) are characteristically parasitic on other and useful 

 members of the order. The destruction of any of these by birds (the 

 cuckoo-flies are taken, at least occasionally) is a benefit. The same 

 must be said of the capture of many species belonging to other fam- 

 ilies which are guilty of the same practice. They parasitize their 

 beneficial relatives, often destroying a large proportion of them. 

 Moreover, even the species usually beneficial do not always stick to 

 their roles, but sometimes parasitize species having the same as their 

 own normal relation to the host. All of these things tend to mitigate 

 the injur}^ done by birds that feed on Hymenoptera. The facts at 

 hand show that, excluding ants (which are in no Avay concerned with 

 parasites) , Hymenoptera compose not more than from 3 to 5 percent 

 of bird food in general. Of the insects composing this percentage, ac- 

 cording to our identifications, nonparasitic forms outnumber the para- 

 sites in the proportion of about 4 to 1. Hence at most not more than 

 1 percent of the food consumed by all the members of our avifauna 

 consists of parasitic Hymenoptera, taken as free-moving forms. 



Proceeding now to the main point at issue, whether great numbers 

 of parasites in the egg or larval state are eaten by birds along with 

 the usual prey, the following statements can be made. The parasites 

 in insect eggs (which include very many of the large superfamily 

 Proctotrypoidea, and some others) are absolutely safe from the ma- 

 jority of birds, only the smaller titmice, Avarblers, etc., searching much 

 for eggs. Parasites also in the cocoons and chrysalids are com2:)ara- 

 tively secure, as only few birds consume these objects. The parasites 

 of all minute insects, like egg parasites, are exem]:)t from danger, ex- 

 cept from a certain contingent of the smaller birds. The parasites 

 living upon medium-sized caterpillars, beetle larva", and adult in- 

 sects, however, run considerable risk of becoming bird food. But 

 even memb(U"s of this group are in a measure favored, since a good 

 proportion of the caterpillars or other insects they infest are not 



