and Caprification. 387 



terebra itself. This organ, affixed to a chitinous plate 

 within the ventral region of the 2nd abdominal segment, 

 effects its exit from within the ventral valve of the 5th, 

 and enters the narrow channel that underlies the elon- 

 gate tubiform segment, leading to the sheaths and their 

 overlapping segment beyond ; but, when separated from 

 these sheaths, it may be readily released from below 

 both retaining segments alike, as far as the ventral 

 valve. This is acutely pointed at its corneous projecting 

 apex, thus coinciding with Hasselquist's definition of 

 " Acute us alius abdomen terminans" &c* : as in Prof. 

 Westwood's description of this part in the GynvpidcB 

 (Mod. Classif., &c, vol. ii., p. 127), " the venter being 

 terminated by a pointed piece having a canal running 

 along its middle, which is also produced considerably 

 beyond its front margin in the shape of a spine; this is 

 the terminal ventral segment of the abdomen." The 

 analogy is sufficiently obvious, although the figure here 

 referred to belongs to a different family. 



In the elaborate descriptions and figures of various 

 ovipositors in the several allied families, exemplified in 

 the same assiduous work, " as typically represented " in 

 Pimpla instigator (loc. tit., p. 139 ; fig. 75, 8—13), the 

 abdomen of the female exhibits " eight dorsal arcs, the 

 eighth furnished at the tip with two minute styles. On 

 the under side of the abdomen there only exist seven 

 ventral arcs, from the last of which arises on each side 

 a corneous elongated plate, which is the basal portion of 

 the outer sheaths of the ovipositor ; the apical portion 

 of these sheaths varies greatly in length in different 

 species, but the articulation always takes place near 

 the extremity of the body." Thus the position of these 

 sheaths in Idarnella, and their articulation towards the 

 base of the overlapping segment, serve to indicate this, 

 together with the elongate antecedent segment, as 

 integral parts of the abdomen, the more especially as, 

 exclusive of such prolongation, its dorsal arcs would be 

 reduced to six ; the ventral arcs being continuous beyond 



* Hasselquist's full description is as follows : — " Cynips Cakic^e. 

 Partes omnes tot in antecedente (C. Ficus !). Abdomen oblomjum, 

 tenuius quam in antecedente, utrinque pa/rum acuminatum, a 

 tlwrace distinctissimum. Spatium inter thoracem et abdomen 

 angustissimum, longius. Aculeus caudce nnicus, corpore duplo 

 longior, capellaris, versus caudam subtus carinatus, crassior, 

 parumque pilosus, reliqua parte tenuis, glaber, cequalis. Aculeus 

 alius abdomen terminans, minimus, crass iusculus, subrigidus." 

 {Iter, p. 425). 



