P''>ipl(f-] BRITISH ICHNEUMONS. 53 



basally attenuate antennae and obsoletely tuberculate bodv, and differs 

 from Tromatobia in the shape of the spiracles and its indistinctly lobate 

 onyches. This is certainly ver}' closely allied to Polysphincta in the often 

 externally pellucid areolet and smoothish bodv, which is far less deeply 

 punctate than in Kphialtis, Epitirus and Piinpla proper ; moreover they 

 also prey upon Arachnida, though upon the eggs and not the imagines, in 

 the egg- sac and not ektoparasitically. 



^Marshall, in the Ent. Ann. 1874, p. 126, makes the sweeping statement 

 that " /'/W/A7 undergoes its transformations without any other covering 

 than the skin of its victim." But this is only true in part and especially of 

 the restricted genus. I ha\e shown that P. ociilaloria spins for itself 

 cocoons within the egg-sac ; Bignell states that the cocoons of P. rufi- 

 pkurae are very like those of the Braconidous genus Macrocentrus and 

 woven upon one another ; 1 have pointed out that those of P. didyma (of 

 which the last is not, as I had first thought possible from the figure, a 

 variety) are arranged in a similar manner, without possibility of error, as 

 also are those oiP. similis and P. detriia lives in galls, constructing no 

 cocoon ; but we know too little of the earlier stages at present to form an 

 opinion as to how far the mode of pupation will prove Thomson's genera 

 to be natural ; we can but surmise that Epiiirus and Scattdms construct a 

 stout, bro\\n, papyraceous cocoon, Tromatobia a flimsy, white, translucent 

 cocoon, Piinpla and Apcchtis and Itoplectis no cocoon at all, simply l}ing 

 free in their host's chr\salis. 



Pimpla is represented in the palaearctic region by some seventy-seven 

 species, excluding nearly twenty of doubtful distinction described by 

 Ratzeburg, and docs not appear to be rare in tropical countries. Of these 

 ^Marshall presented 21 from Britain in 1872, and I found a total of 37 in 

 1 90 1. Many of these have proved to be synonymous, or, like P. abdomi- 

 iialis, to have been incorrectly ascribed to this genus ; with the result that, 

 after adding five species to our fauna and describing three new ones, our 

 total stands at 40 species, of which 1 do not possess P. pictipts and 

 P. aeihiops. 



Table of Species. 



(50). I. Clypeus apically emarginate ; ner- 

 vellus intercepting ver>' rarely 

 above centre ; metathoracic spira- 

 cles always circular. 



(5). 2. Abdomen very strongly punctate ; 

 terebra longer than body [Exer- 

 istes, Forst.) 



(4). 3. Tarsal claws basally dentate ; thorax- 

 black .1. RORORATOR, Fab. 



(3). 4. Tarsal claws not basally dentate ; 



thorax red . .2. RvricoLLis, Grav. 



(2). 5. Abdomen normally punctate ; tere- 

 bra rarely as long as body. 

 (49). 6. Lateral metathoracic areae wanting; 



cheeks black. 

 (38). 7. Front femora of J entire ; abdomen 

 of $ distinctly tuberculate {EJ>i- 

 urus, auctt.). 

 (27). 8. Nervellus intercepting not below the 

 centre ; antennae longer than half 

 body. 



