Miss Haviland, Preliminary note on a Cynipid, etc. 235 



Preliminary note on a Cynipid hyperparasite of Aphides. By 

 Maud D. Haviland, Fellow of Newnham College. (Communicated 

 by Mr H. H. Brindley.) 



[Read 22 November 1920.] 



Aphides are liable to parasitisation by certain Braconidae of 

 the sub-family Aphidiidae. The larva develops in the haemocoele 

 of the host, which dies just before the metamorphosis of the para- 

 site, and the latter lines the empty skin with silk and pupates 

 within it. Throughout its larval life the Aphidius is in its turn 

 liable to parasitisation by certain Chalcids, Cynipids, and Procto- 

 trypids, which are thus hyperparasites of the aphid. I described 

 the development of one of these hyperparasites, a Proctotrypid, 

 Lygocerus cameroni, Kieff., in a paper read before the Society last 

 February. The following is a summary of some observations made 

 on the development of certain Cynipid endo-hyperparasites of the 

 genus Charips, formerly known as Allotria. The aphid used in the 

 breeding experiments was Macrosiphum urticae, Kalt. from the 

 nettle, and the primary, or host, parasite was Aphidius ervi, Hal. 



These Cynipids have long been known to be hyperparasites, but 

 at first I could not induce them to oviposit in captivity. The cause 

 of this failure was that the material offered them was in too ad- 

 vanced a stage; for, unlike most of the hyperparasites of this 

 group, which do not oviposit until the aphid is dead and the 

 primary parasite is in metamorphosis, these Cynipidae seek out an 

 Aphidius preferably in the third, or early in the fourth, instar, 

 though a second instar larva may also be chosen. At this time 

 the parasite is lying in the host's body cavity, and the aphis feeds 

 as usual. Until twelve hours or so before its death there is no 

 external sign that it contains a parasite, and yet the Cynipid un- 

 erringly recognises the presence of the latter and ignores unpara- 

 sitised aphides when they are offered to it. A necessary condition 

 for oviposition appears to be that the aphid host should be ahve, 

 and the Aphidius be still bathed in its body fluids. Larvae which 

 had emptied the Aphid's skin of its contents, and had already 

 begun to spin the cocoon, were never selected. 



In captivity, ovipositions were sometimes, but not always, par- 

 thenogenetic. The female Charips ran among the aphides, tapping 

 them excitedly with her antennae. When a victim was chosen, 

 she leaped upon its back, facing the head, and clung there in spite 

 of its struggles. Sometimes she was thrown off, but in such cases 

 always returned to the attack, until the aphid became passive from 



VOL. XX. PART II. 16 



