124 MODERN CLASSIFICATION OF INSECTS. 
Dahlbom, who thinks that Oryssus constitutes the connecting link 
between them); and 2dly, in showing the relation of Evania and Ichneu- 
mon. How far it would be natural to place the Cynipide at the 
head of these groups, making the aphidivorous Cynipide the con- 
necting link with the aphidivorous Adscitous Ichneumons, which 
should be succeeded by the genuine Ichneumons, and these by 
Aulacus and Foenus, Evania conducting to the species with the wings 
almost destitute of nerves, can only be ascertained by a more precise 
and general investigation of the structure of these insects than has 
yet been given to them. For the present I have arranged them in 
the following manner, provisionally keeping Evania distinct from 
Ichneumon : — 
I. Sprcutirera. * — Terminal segments of the abdomen not forming 
a retractile tube; ovipositor furnished with two delicate spiculee 
working in a horny semi-canal, which is defended at rest by two 
often partially exserted valves. 
Fam. 1. Cynipide.— Chiefly gall-feeding insects, having the 
ovipositor subspiral and internal ; antenne straight, 13- to 15- 
jointed. 
Fam. 2. Evaniide.—- Parasitic insects, having the ovipositor 
straight, the abdomen attached to the dorsum of the meta- 
thorax, and the antenne straight and 13. to 14-jointed. 
Fam. 3.. Ichneumonide. — Parasitic insects, having the ovipositor 
straight, the abdomen attached at the extremity of the meta- 
thorax; the antenne straight, and with more than 16 joints 
(except in a very few minute species). 
Fam. 4.  Chalcidide. — Parasitic insects, having the antenne 
elbowed, and generally thickened at the tips, 6- to 13-jointed ; 
the palpi short; the wings nearly destitute of veins; pupa 
naked. 
Fam. 5. Proctotrupide. — Parasitic insects, having the antenne 
more or less elbowed, 10- to 16-jointed; the upper wings vein- 
* The difficulty which opposes uniformity in the nomenclature of the higher 
sections of an order is well instanced in the groups of the Terebrantia. If we em- 
ploy terms founded upon the habits of the different families, we must introduce the 
gall-flies amongst the plant-feeders (Phytiphaga); if, on the other hand, we em- 
ploy terms founded upon structure, we shall have the Uroceridx separated from the 
saw-flies, and united with the Spiculifera, on account of the structure of their 
oviposit 
