88 FOOD HABITS OF THE GROSBEAKS. 



likely to be eaten by birds, both because they are usually discolored 

 or otherwise abnormal in appearance and because often they are in- 

 active, and hence escape attention. It should be stated, hoAvever, that 

 freshly parasitized individuals are of normal appearance, and 

 whether or not they are recognized by birds, it is certain that in the 

 stomach such caterpillars can not be distinguished from the unin- 



fested. 



However, it is certain that, in spite of this difficulty of recog- 

 nizing recently parasitized larvae, a well-defined tendency on the 

 part of birds to avoid these doomed insects is shown by the wholly 

 insignificant number found in which parasitism is evident. Only 

 the one jDarasitized caterpillar mentioned above, and one other eaten 

 by a hawk, have been found in more than 40,000 stomachs examined 

 by the Biological Survey. Hence it would seem that serious diminu- 

 tion in the ranks of parasites is much more logically explained by 

 some other cause, such as the well-known pernicious activities of cer- 

 tain hyperparasites, than by the attacks of predaceous foes. The 

 following extract « from "A Study of the Hymenopterous Parasites 

 of the American Tent Caterpillar," by W. F. Fiske, serves to show the 

 possibilities of this hyperparasitical activity. 



In 189G it [the tent caterpillar] was exceedingly common, so much so as to 

 attract attention all over the State, and it was consequently assumed that its 

 parasites would also have increased to such an extent as to be a considerable 

 factor in bringing about a more normal condition in 1897. But strangely 

 enough this reasoning was found to be diametrically opposite to the facts of 

 the case. In 1896 a great many caterpillars were destroyed by Llmneria fugi- 

 tiva [a parasite], and numbers of the characteristic cocoons of this species were 

 collected and bred at this station. These proved in their turn to be almost 

 entirely the prey of PimpJa inquisitor [a hyperparasite], and the breeding 

 cages in which the cocoons were confined becoming filled with the adults of the 

 hyperparasite, it was mentioned in the report of the year, and given credit 'for 

 the destruction of large numbers of tent caterpillars. Thus it was the Lim- 

 neria which proved to be the real victim, and as its cocoons were rare the next 

 year the primary host, Clislocampa americana [the tent caterpillar], was indi- 

 rectly but decidedly the gainer. 



As one season's observations succeeded another's this case from being an 

 exception, as was at first thought, was found to be more nearly the rule. The 

 comparative abundance of the tent caterpillars varied greatly, it is true, but 

 this was obviously due to other and distinct causes than this under discussion, 

 and only a certain percentage, varying considerably from one colony to another, 

 or to a less degree between different localities, -fell a prey to the hymenopterous 

 parasites. The parasites as a body were found to exact a certain tithe of blood, 

 so to speak, and this being obtained would proceed to fight over its possession, 

 passing it from one to another, from parasite to hyperparasite, until often, per- 

 haps, there would be little more than enough left of a single large caterpillar 

 than that sufficient to support a single small Chalcid. 



It is evident from these statements that we need look no further 

 for the enemies which so check the useful services of parasites. 



"Tech. Bull. 6, N. H. Agr. Exp. Sta., 1903, pp. 185-186. 



