38 MISCELLANEOUS PAPERS. 



HABITS IN THE OPEN. 



Ovi position as observed in the field is not very different from that 

 in the laboratory. The following notes were taken while watching 

 the parasite on foliage of turnip infested with Thrifs tahaci: 



The female was oliserved cravling slowly along with the antenna? in motion and. 

 turned first to one side and then to the other; the body also was turned to one side or 

 the other. At other times she ran very rapidly across the leaf. Finally she discovered 

 a larva of tabaci and immediately placed her anterior legs on it and lowered her 

 antennae until they rested on the body. She then oviposited in the host in the same 

 manner that was observed so many times in the laboratory. In nature, however, the 

 thrips larvse were more scattered, so that this insect succeeding in finding and oviposit- 

 ing in only 4 larvae during 40 minutes, and to do this slie covered nearly 6 square 

 inches of leaf surface. 



In other cases where this insect was oljserved in the open, a slight jar would frighten 

 her and instantly she would take wing. 



EFFECT OF OVIPOSITION IN DIFFERENT STAGES OF HOST. 



In the course of the work in rearing the thrips parasite it was found 

 that this insect would oviposit in all stages of the host. In order 

 to determine if the parasite would develop in all stages, a number 

 of experiments were performed in which the ]:)arasite was placed in a 

 vial with ])repupfe and pupse of Ileliothrips fasciatus; and in every 

 case the i)arasite did not hesitate to oviposit freely in these stages. 

 In spite of this not a single form ever developed the slightest sign of 

 parasitism, and the host adult developed in all cases. From these 

 experiments, it appears, we can safely say that, although this insect 

 will ovijiosit in all forms of this thrips, it will develop successfully 

 only when the eggs are deposited in the larval stage of the host. 



NUMBER OF EGGS AND LENGTH OF OVIPOSITION PERIOD. 



Because of the extremely artificial conditions that were necessary 

 in rearing this very minute hymenopteron, it is im])ossible even to 

 estimate with any degree of correctness the egg-laying capacity of a 

 female under normal conditions. Tliis is due not only to the fact 

 that the parasite is apt to die before the normal number of eggs is 

 deposited, but also largely to the artificial method of rearing, by 

 reason of which many of the host larvae that have been parasitized 

 die before they show the usual indications of parasitism. In one 

 case a female was observed, under laboratory conditions, to oviposit 

 in 36 larvae of Ileliothrips fasciatus 38 times in 1 hour. From the 

 oviposition record of another female, 515 parasitized ])repup8e devel- 

 oped, while a third, that undoubtedly ])arasitized a still larger number, 

 gave a total of 91 parasitized forms tliat later developed. 



The parasites in the vials were observed to ovi])osit in thrips, as a 

 rule, during a period of 5 days, although in one instance a female 



