236 THE CANADIAN ENTOMOLOGIST. 



original female type there preserved is very different from any females of 

 O. Carolus I have seen, especially in the form of the occiput (see Plate 5, 

 figs. 10 and 19). The occiput is variable, to be sure, but I have shown 

 the full extent of variability exhibited by a very large series of females of 

 0. Carolus in these pages (Can. Ent., XXIX , PI. 7, figs. 1-4), and have 

 found nothing approximating the conformation of the type O. Mainensis. 

 Since it is possible that the males associated with this female type may 

 not belong with it, one must show before uniting the species either that 

 the normal variation of the occiput includes such forms, or else that the 

 female type is a freak. The specimens in the Lintner collection, deter- 

 mined by Hagen as O. Mamefisis, agree entirely (both males and females) 

 with O. Carolus. 



Of the three closely allied far-western species, O. severus, Hag., O. 

 inonianus, Sel., and 0. Morrisofii, Sel., I have seen very few specimens: of 

 f/wntatius, none at all. Montanus was first described as a variety of O. 

 severus, but was ranked as a species by De Selys in his Revision des 

 Ophiogomphus (C. R. Ento. Soc. Belg., 1879, P- Ixiv.), and so listed 

 by Kirby in his Catalogue of the Odonta. These three species constitute 

 a group within the genus characterized by De Selys by the simple (horn- 

 less) occiput of the female — a thing not distinctive, as we have seen, but 

 apparently entirely characteristic of these species. 



The remaining species constitute a troublesome lot, among which 0. 

 aspersus, Morse, seems pretty sharply defined ; but variation in the form 

 of the accessory genitalia is very considerable. Only two of the species, 

 O. rupinsulensis and O. occideniis, are known from more than a few 

 specimens. The figures herewith given fur these two species seem 

 distinct enough, yet the specimens in the Hagen collection show them to 

 intergrade almost completely. O. bison was dropped by De Selys frojii 

 the list given in his Revision (op. cit.) — whether intentionally or not, I do 

 not know — but the female in the Hagen collection is certainly very much 

 like O. ru[)insulensis. (See Plate 5, fig. 32.) I desire at this point to 

 correct a very serious error of my own : Misled by the upturned inferior 

 appendages of the male, and having too great faith in the constancy of 

 genital characters, I described as Herpetogomphus pictus (Can. Ent., 

 XXIX., 181, 1897), some exceptionally finely coloured males of O. rupin- 

 sulensis. Since studying a large series, I do not retani the name even 

 for a reliable variety. 



I figure here for O. occidentis (PI. 5, figs. 4, 13 and 22), the bred 

 specimen in the Hagen collection, which must be consideied the type, 

 since its cast skin is described (Trans. Amer. Ent. Soc.^ XH., 259). 



