i89 



and the parasite almost immediately proceeds to investigate an- 

 other, and to lay therein, it is probable that the eggs in the first 

 one have been already parasitized, though* of course it may be 

 simply that the egg of the leaf-hopper has advanced beyond a 

 proper stage of development for being parasitized. In eggs 

 examined I found no advanced stage of development of the 

 leaf-hopper embryo, when they contained a larva of the para- 

 site. However this may be, a suitable egg-chamber having 

 been found, the ovipositor is kept inserted for a considerable 

 time, sometimes for several minutes. Both in the preliminary 

 examination of the egg-chamber and during the process of ovi- 

 position, the parasite may sometimes be seen by a very rapid 

 turn through half a complete revolution, to exactly reverse its 

 position, and this sometimes takes place more than once. Chal- 

 cids of other families, with habits similar to the Mymarids, 

 behave in precisely the same way, and are even more rapid in 

 their movements, so that these can only be appreciated by our 

 eyes owing to the change of position, the head and tail end of 

 the parasite having as it were, changed places instantaneously. 

 The number of eggs of leaf-hopper in a single chamber varies 

 from one to twelve or more, six or eight being common, and all 

 these eggs may be stung, or some may escape. I have not exa- 

 mined the egg of the parasite which presumably is thrust into the 

 leaf-hopper's egg, and in that case is likely to be hardly or not at 

 all visible amongst the opaque contents of the latter. The ma- 

 ture larva is the usual apodous grub, and it may be seen per- 

 forming its curious rotatory movements amongst the fluid con- 

 tents of the egg that contains it. It has also a habit at times 

 of violently throwing forward its anterior segments, so that the 

 head of a half grown larva, which reaches to the middle of the 

 hopper's egg, will, at its greatest extension, nearly attain the 

 operculum. According to authors mPolynema, there is a very 

 remarkable second form or 'HistriobdcUa' stage of the larva. I 

 have not yet observed this in Paranagnis, but the third form (as 

 figured by me) greatly resembles that of Polyncma. The smallest 

 larve examined by me measured {\ mm., the largest |f mm. 

 so that the main growth must be in this stage. Except in size 

 the smallest and largest larvae did not noticeably dififer. 



The head of the larva has a cjuite remarkable armature as is 

 represented under very high magnification on PI. XIII, fig. 5 

 and 5a, from a half grown example extracted from the egg of its 

 host. The pupa almost or quite fills this egg (excluding the 

 opercular portion) and before it is fully developed is generally 



