96 JOURNAL OP ECONOMIC ENTOMOLOGY [Vol. 3 



offset by their larger numbers. Similar conditions may be expected to 

 prevail in the field. 



More than 50 per cent of parasitism by species which attack the 

 pupffi of any lepidopterous insect is rare, or appears to be so, and has 

 never been encountered in the experience of the writer. A higher rate 

 of mortality frequently results when the pupa of the host is also af- 

 fected by parasites which have attacked it during its larval stages, as- 

 for example in the case of the tussock moth which dies after construct- 

 ing its cocoons through parasitism by Tachinids, as well as by Pimpla 

 and Chalcis. If the parasitism by Tachina amounts to 50 per cent and 

 of Pimpla and Chalcis together to 50 per cent, it is safe to say that 25 

 per cent of the host will escape attack entirely, and that a consider- 

 able majority of the parasites, taken altogether, will be involved in a 

 conflict between themselves. This conflict, as already stated, may 

 result variously, but with rare exceptions to the disadvantage of one 

 or both of the participants. 



It is very largely on this account that it is deemed inexpedient to 

 depend upon the parasites, which confine their attack to any one stage 

 of the host, to effect the control of an insect like the gypsy moth, 

 which is subject to the attack of several distinct groups of parasites 

 at different stages in its career. It is recognized that the parasites 

 have their own enemies to contend with, and that they are subject to 

 all of the other and various controlling influences which go to main- 

 tain the stability of the natural balance. Add to these the losses 

 which they would suffer through undue super parasitism, and the 

 chances for their continued increase to the point where they could be 

 expected to exert effective control of an insect so fecund as the gypsy 

 moth are greatly reduced. 



The importance of superparasitism, in its relations to hyperpara- 

 sitism, is also considerable. It is probable, in the last analysis,, that 

 true tertiary parasitism will be found to be of rare occurrence, and 

 that instances in which a secondary parasite is attacked are, for the 

 most part, examples of superparasitism pure and simple. The destruc- 

 tion of host by parasite, and parasite by hyperparasite may be, and fre- 

 quently is, carried beyond " quarternary " and " quinquenary " para- 

 sitism, but in relatively very few instances are the species involved 

 incapable of attacking the original parasite as well as its secondaries. 



The general subject of parasitism, as a factor in the natural control 

 of insects, is of course the main, and almost the only subject under 

 consideration at the laboratory. Its phases are manifold and varied 

 in the extreme, but there is hardly a single one which is not involved 

 more or less directly with some phase of the subordinate topic of super- 



