Junes. i9i6 Hessian-Fly Parasites 379 



mounts where five pairs could not be found, each of those present cor- 

 responded to some one of the pairs in the complete series. No more 

 than five pairs were found in a single mount. As in the two species 

 of parasites previously discussed, the head shield of the newly hatched 

 larva was more heavily chitinized than those of later instars. The 

 mandibles appeared to be more powerful for their size than those of 

 any later instar, and in some cases they were actually larger than the 

 second-instar mandibles. The approximate sizes of the respective pairs 

 of molted mandibles follow. The measurements represent the distance 

 from the tip of the mandible to the shoulder where it suddenly enlarges 

 into the broad base. 



Molt No. Length of mandible. 



I o. 012 mm. 



2 012 mm. 



3 016 mm. 



4 020 mm. 



5 028 mm. 



Larvae of Micromelus subapierus do not seem as capable of moving 

 around and reattaching themselves to the host as are the larvae of Eupel- 

 mus allynii and Merisus destructor. Larvae reared in glass cells crawled 

 about a little immediately after hatching before they settled down to 

 feed, but they usually completed a large part of their growth without 

 leaving the original feeding point on the external surface of the host. 



This species not only developed on Hessian-fly lar\^ae in puparia, but in 

 some instances fed on the lar\^£e of other parasites. One egg of Micromelus 

 subapierus was placed on a full-grown lar\-a of the same species in a glass 

 cell. The egg hatched and the little larva became full grown on the large 

 larva, almost completely devouring it. Another egg of M. subapierus was 

 placed on a full-grown larva of Merisus destructor and the little larva 

 hatching from the egg became full grown on the larva of Merisus destruc- 

 tor. Experiments like these, however, usually resulted in the destruction 

 of the egg or young larva of M. subapterus and the survival of the full- 

 grown larva of the same or the other species as the case happened to be. 

 Larvae of M. subapterus apparently could make their growth on the 

 Hessian-fly pupa as well as on the larva unless the former had partially 

 developed. Where the host pupa had already completed a large part of 

 its development, both the host and the parasite generally died, the latter 

 apparently for lack of sufficient suitable food. Larv^ae of M. subapterus 

 appeared to be the least able to defend themselves where the larv^ae of 

 more than one species occurred in the same flaxseed. They also seemed 

 the least capable of successfully establishing a feeding point on the host 

 larva, at least when reared in little glass cells. They seemed more delicate 

 in structure and less vigorous. 



The respective periods required for 36 larvae to make their growth 

 varied from 7 to 10 days. A large proportion of the larvae after finishing 



