THE CANADIAN ENTOMOLOGIST. 351 



below, and by the broader vertex at meson ; from Metopachia Westwood 

 by the usually shorter postmarginal vein, the smaller metathoracic 

 spiracles, and from the fact that in Arthrolyius the abdomen is not 

 strongly keeled, or acutely, strongly produced or convexed ventrad near 

 the base. Distinct, or not easily confused, with other genera of the 

 Pteromalini, excepting, perhaps, Coelopisthia Foerster,* from which it is 

 easily distinguished by cephalic characters, antennal characters, and those 

 of ihe abdomen. 



Host Relations of the Genus. 

 Of Thomson's three species, which were probably captured at large, 

 nothing concerning the host is known. Arthrolyius puncticollis Moller 

 was described from specimens reared from the larva of (Ajiobium) 

 Sitodrepa pa7iicea (Linnaeus), a coleopteron of the family Ptinidse. Ash- 

 mead's species, apatelcB, is recorded from a lepidopteron, Apatela popidi 

 Riley, of the family Noctuidae, but its relationships here are doubtful, 

 Fhogas inter7nedius Cresson also being present ; while ceneoviridis 

 Girault has doubtful relationships, having been reared from breeding-cages 

 containing the larva? of (Alceris) Pero?iea mmiita (Robinson), of the 

 lepidopterous family Tortricidae, the adults emerging after the 

 ichneumonoid Clinocentrtcs americana Weed, and before Horismetius 

 microgaster (Ashmead). It is therefore probably a secondary parasite of 

 the Peronea, its host being the Clinocentrus ; I think it more probably a 

 primary parasite of the lepidopteron, however. From these few records, 

 of course, it would be hazardous to generalize in regard to the genus as a 

 whole. It is indicated that the genus may attack the Hymenoptera, and 

 thus play a secondary or even tertiary role of parasitism for the Lepidoptera 

 and Coleoptera, somewhat like the role of Dibrachys. 



Distribution of tJie Geiius. 



The genus is represented in the continents of Europe and North 

 America. Of the European species, of which there are four, all are 

 confined, so far as is yet known, to the Scandinavian Peninsular (de 

 Dalla Torre, 1898), Sweden. The two North American species are 

 recorded fjom Ohio {apatelce Ashmead), and Iowa {ceneoviridis Girault.) 



a. No species was mentioned in connection with the genus by Foerster, but 

 as Thomson adopted the name and designated a type for the genus, the name 

 holds, and Foerster is held responsible for it. 



