254 THK ENTOMOLOGIST. 



Each jaw carries three teeth, not mere notches as in Lepi- 

 doptera generally, but each tooth is a long sharp spine, capable 

 of piercing, but certainly not of biting ; each jaw is probably 

 capable of meeting the other, so that the teeth may interlock, 

 but in the specimens examined one jaw is entirely in front of 

 the other. 



The eye-spots are six, five of them in a semicircle, the 

 other separate. 



The head, which looks sunk into the white fleshy tissue of the 

 under side of the larva, is really very moveable, and has a definite 

 neck (?), so that the mouth-parts, which are at front of head 

 and point more or less forwards, can be directed directly back- 

 wards, between the true legs, exposing the front or dorsum 

 of the head, which is rather longer than broad, nearly colour- 

 less, and has some hairs, and the usual suture marking off the 

 clypeus. 



I should like, by again mentioning, to emphasize what seems 

 to me as remarkable a feature as any it possesses — vis;, the de- 

 velopment, de novo, of a " Micro " proleg, by the obsolescense of 

 the real crotchets, though not of the base that carries them, and 

 by the appearance of an entirely new set of crotchets round the 

 base of the proleg proper. 



Not so remarkable as a structural modification, but more so, 

 perhaps, as connected with most unusual habits, is the modifica- 

 tion of the jaws, as piercing and tearing and no longer biting 

 organs, and, if I observe correctly, the alteration of the tropin 

 into a suctorial tube, from which the jaws are just able to pro- 

 trude ; remarkably similar, functionally, to the tube surrounding 

 the jaws in PJiyllocnistis, though the details of structure and 

 habits are so widely different. 



The jaws would most efl'ectively take a hold of the skin of an 

 ant larva, piercing its skin at the same time in six places ; they 

 would then draw the piece so seized within the closed cavity 

 formed between labrum, labium, and (laterally) maxillae, so that 

 the juices of the larva could be easily sucked out. 



There is also a larva of intermediate size, which differs from 

 the larger one in nothing except perhaps that the spiracles are 

 more readily seen than in the full-grown one. 



The pupa I have before me is very large, 28 mm. long, 14 

 mm. broad, and 10 mm. deep, whilst it is depressed in front, in 

 a way apparently due to pressure ; were the rounded contour of 

 the dorsum and sides continued its depth would be 12 mm. It 

 is typically Lycasnid in form, being very round at either end, 

 broadest at fourth and fifth abdominal segments, narrower 

 thoracically ; head beneath ; no moveable segments ; no trace of 

 cremastral hooks or of any silken girth ; first leg equally against 

 head and antennae. The maxillae are well developed; they appear 

 to contain no maxilla (the specimen being close on emergence), 



