INTRODUCTION. . xlv 



specialised a group, constituting exactly one of the two great divisions of 

 Linne. But it is quite necessary to here set forth what does and what does 

 not belong to the family Ichneumonidae, since this appears to be often 

 imperfectly understood, among even distinguished general entomologists 

 of the present day. The other division of Linne was the Terebrantia of 

 Latreille, embracing all those insects whose females were possessed of a 

 terebra in place of a sting; and it was not till 175S that the former 

 grouped the social species under Sphex, and separated the remainder by 

 the sessility of the abdomen, dividing the Ichneumons into four purely 

 artificial families. Consequently by this method a male specimen was, 

 from the very outset, undeterminable. 



Forster^ proposed the division of Hymenoptera by the joints of the 

 trochanters ; those with two joints were designated Ditrocha, those with 

 but one were Monotrocha. Here difficulties also arose upon application, 

 since no hard line could be drawn, instances occurring in which a second 

 joint was distinctly present, though connate with, and apparently forming 

 part of, the femur ; and the ensuing divisions disseminated obviously natural 

 affinities, giving rise to the most incongruous grouping. De Geer had, in 

 1752, employed the comparative extent of the connection between thorax 

 and abdomen for his minor groups of the Terebrantia, and this feature is 

 now acknowledged to be as essential and natural for main divisions as 

 is to be obtained in so heterogeneous and vast an Order, of which more 

 than four ttiousand seven hundred species are said to inhabit Britain 

 alone. 



Ashmead primarily divides the Hymenoptera into Phytophaga, in which 

 the abdomen is sessile, with the trochanters two-jointed and the larva 

 legged, and Heterophaga, in which the abdomen is more or less petiolate, 

 with the trochanters either one- or two-jointed, but the larva apodous. This 

 latter is sub-divided roughly into the Aculeata, with the Oxyura, emitting 

 the terebra from the apex of the abdomen, and the Ichneumonidea, with 

 the Cynipoidea and Chalcidoidea, emitting the terebra from the ventral 

 surface of the abdomen. Further evidence is, however, I think, needed 

 before mingling the Proctotrypidae with the Aculeata, and, in fine, it 

 appears most desirable to follow the classification set forth by Konow.- 



This, in effect, differs little from that of Westwood,-"' though it raises the 

 Phytophaga to equal dignity with the Entomophaga and Aculeata. We thus 

 have not a dichotomous but a three-fold classification, in which each section 

 possesses some feature common to both the remainder. The new element 

 introduced is centred in the number of cubital wing-cells supposed to have 

 been originally present. The first of these sections is the Vespoidea, bear- 

 ing one-jointed trochanters, petiolated abdomen and, at least in the ancestral 

 type, four cubital cells. The second is the Ichneumonidea, bearing two- 

 jointed trochanters, petiolated abdomen, and but three, often suppositious, 

 cells. The third, the Tenthredonidea, bearing two-jointed trochanters, 

 sessile abdomen, and four, often suppositious, cells. Thus we have the 

 first and second sections constituting the Clistogaster ; the third the 

 Chalastogaster. The second and third the Ditrocha ; the first the Mono- 

 trocha. The first and third the Tetrastega, and the second alone the 

 Tristega. 



1 Dr. Arnold Forster, Synopsis der Faiiiilien und Gattungen der Braconcn, 18G2, and dcr Iclinou- 

 monen, 1868. 



2 Ent. Nachr. 1897, p. 148 ; cf. E.M.M., 1899, p. 124 ; et Deut. Ent. Zeit. 1890, p. 228. 

 s Modern Classification ot Insects, 1840. 



