330 



the larva, which are merely a chitinous secretion from a part 

 of the mass of cells which will later develop the mandibles 

 of the adult, are produced from a short mandibular imaginal 

 disc, which grows forwards from the second segment and 

 t-erminates close beside the mouth; secondly, in the late larval 

 stages there is a considerable shifting forwards of the lower 

 surface of the head, the anterior portion of the second seg- 

 ment being pushed into the cuticular sheath of the first. (I 

 may draw attention here to a fact from which an important 

 deduction can be made later, viz., that the antennae and 

 mandibles of the larva, though so absolutely distinct from 

 those of the adult, are yet developed in close connection with 

 the same group of cells — the antennal and mandibular 

 imaginal discs — as produce the corresponding structures in 

 the adult wasp.) 



Of the ''post-oral" appendages four pairs may be recog- 

 nized. The most anterior is a pair of short outgrowths, 

 which I shall call here the second antennae (fig. 13). Their 

 homology will be considered below. Immediately behind these 

 arise a pair of long outgrowths, which end close to .the 

 larval mandibles — they are the mandibular rudiments ; close 

 behind these, and nearer the mid-line, is a pair of short 

 stout maxillary rudiments; and behind these, and still nearer 

 the middle, are the rudiments of second maxillae, quite dis- 

 tinctly paired at this stage (fig. 13). 



Of these appendages the mandibles are the largest, and 

 I have seen larvae, slightly before defaecation, m which each 

 is provided with a palp, which, at this stage, is even longer 

 than the mandible itself (fig. 3). I have also observed larvae, 

 in the same stage of development, in which no mandibular 

 palps were visible. In order to be certain that I was not 

 confusing the mandibles with the first maxillae, I examined 

 the mouth appendages of defaecating larvae, cut in serial 

 sections ; under which conditions no error could be made in 

 determining the various mouth appendages, and the mandi- 

 bular palp could be clearly seen (fig. 48). A mandibular palp 

 has not, so far as I am aware, been found hitherto in insects. 

 Of special interest, however, is the fact that it does not appear 

 to be present in all larvae, its occurrence being perhaps a 

 frequent ''abnormality." 



The first maxillae are rather short thick outgrowths at 

 this stage, and each has a short palp on its outer side. The 

 second maxillae are small, and each has a very distinct palp, 

 which twists around the maxilla from beloM', and embraces 

 it distally. At the sides of the second segment are the great 

 compound eyes, already differentiating in the late larval 

 period; and in the middle lie the great cerebral ganglia 

 (fig, 3). 



