SERPHOID AND CHALCIDOID PARASITES OF THE HESSIAN FLY 127 



HOSTS AND LIFE HISTORY 



The Girault record of this species attacking eggs believed to be 

 those of a species of Xiphidium is principally interesting as demon- 

 strating that this species, like all the other species of the genus, nor- 

 mally attacks the eggs of Orthoptera and Homoptera which are 

 inserted m stems and twigs. Its parasitism of the hessian fly, while 

 much more frequently observed, is believed to be either accidental 

 or an acquired habit. In this connection it is interesting to note 

 that, in one instance at least, specimens have been taken under cir- 

 cumstances indicating that the species may at times attack joint- 

 worms, as there are two specimens in the National Museum collection 

 which are said to have been found dead in the gall of Harmolita 

 tritici (Fitch) at Sparta, 111., by F. F. Dicke. 



According to McConnell, Centrodora speciosissima was first reared 

 in 1915 from hessian-fly puparia collected by P. R. Myers. Since 

 that time it has been reared many times, and it may be said to be a 

 common parasite of that insect. 



This Centrodora is a primary, internal, gregarious parasite of the 

 hessian fly. McConnell demonstrated that it was a primary para- 

 site by repeatedly getting it to oviposit in puparia known to be free 

 from parasites and rearing it through to the adult. On the other 

 hand, P. R. Myers stated in his unpublished manuscript that it 

 sometimes acts as a secondary parasite, specimens of it having been 

 reared from puparia. of the hessian fly which were afterward dis- 

 sected and found to contain dead remains as well as living adults of 

 Platygmter. 



According to McConnell, the eggs are deposited in the body cavity 

 of the host after the puparium has been formed, no distinction being 

 made between puparia containing larvae and those containing pupae. 

 In July and August a complete generation of the parasite may de- 

 velop in 18 to 20 days, indicating the probability of 3 or 4 generations 

 per season. Emergence of the adult is through a hole cut in the side 

 of the host puparium, and as many as 10 adults may mature in a 

 single puparium although the normal number is said by McConnell 

 to be considerably less than 10. The species is said to be capable of 

 reproducing parthenogenetically, unfertilized females producing only 

 males. 



DISTRIBUTION 



Records of the Bureau of Entomology now show a much wider 

 distribution for this species than that indicated by McConnell in 

 1916. At that time it was known only from the type locality, Ur- 

 bana, 111., and from six localities in Pennsylvania. It is now known 

 to occur in 10 States, viz New York, Pennsylvania, Delaware, Mary- 

 land, Virginia, West Virginia, Kentucky, Ohio, Indiana, and Illinois. 

 It has been obtained from numerous localities in Pennsylvania and 

 Maryland, but from relatively few localities in the other States 

 mentioned. This is probably due to the fact that the parasites of 

 the fly have been more extensively studied in these two States than 

 in the others. It is believed that equally careful investigations in 

 the other States would probably show it to be distributed quite as 

 generally there as in Pennsylvania and Maryland. 



