of certain species of Willow. — Part 2nd. 235 



iiiid similarly arranged, except that the 8th dorsal joint is larger and is 

 scarcely ever retracted, yet there are always 8 complete ventral joints, 

 i — 6, %S before, lying opposite to the dorsal joints 1 — G, while oppo- 

 site 7 and 8 there lie, not one but two joints, viz: joint 7, which is 

 nearly as large as those immediately preceding it, and joint 8. which 

 is very much larger and nearly as large as the large terminal or 7th 

 joint in % Tenthredo, kc. Evidently the typical number of ventral 

 joints throughout the whole family is 8; but in % Tenthredo, &c. 

 joints 7 and 8 are confluent, so as to become apparently one joint. 



In all 9 Tenthredinidse the abdominal dorsum is 8-jointed, 1 — 7 

 bearing a spiracle as in o . and 8 being rather small, yet very distinct; 

 but, as in all other Terebrantia, the venter has only six complete joints. 

 the ovipositor and its sheaths taking their origin from under the tip of 

 joint 6, so as to obliterate more or less completely the remaining ven- 

 tral joints, and being laterally fringed by the overlapping part of the 

 dorsal joints 7 and 8. This overlapping part is found % 9 in every 

 dorsal joint — being generally in Tenthredinidse distinctly separated by 

 an acute angulation from the dorsal surface and bearing the spiracle in 

 joints 1 — 7 — and has been called throughout in my descriptions u the 

 lateral plate." In reality, this part, as I have observed in Pseudoneu- 

 roptera, (Proc. &c, II. p. 250. &c.,) is homologous with the "pleura" of 

 the thoracic segments. Westwoood indeed describes and figures a 

 small piece (7- -), laterally attached to the tip of the Gth ventral in 9 

 Trichiosoma, as a true 7th ventral. (Jntrod. II, p. 94, figs. 12 and 13.) 

 But on the most careful examination I can detect no such piece in 9 

 Cimbex or any other Tenthredinidous 9 . though in 9 Cimbex there is 

 a hole or excavation in the spot occupied by his piece "7-j-." In Uro- 

 ceridse, it is true, there is a very distiuct, small, transverse lateral piece 

 corresponding to the Westwoodian u 7-j-," which is no doubt a rudi- 

 mentary 7th ventral, and is figured but not numbered or lettered by 

 Westwood. (Ibid. p. 115, fig. 13.) 13ut in the allied family Ichneu- 

 wonidx he neither describes nor figures any such piece, nor can I dis- 

 cover any such myself. Here, therefore, it might be inferred that this 

 author would describe the 9 venter as 6-jointed. No such thing. In 

 this family he obtains the additional 7th ventral in 9 j not at the tip. 

 but at the base of the venter. For in describing and figuring the 9 

 venter of the Ichneumonidous genus Pimpla as 7-jointed, not 6-joint- 

 ed, he has been deceived into considering the 1st ventral joint as two 

 joints, because its basal portion is enwrapped by the horny dorsal joint 

 1 so as to form a short robust peduncle, the whole of which, both 



