129 



iiig- frequently much more contracted or even almost globular. 

 (For. fig. of larva, see PI. \1I, fig. lo, loa, lob). 



The puparium is formed by the hardening of the larval cuti- 

 cle, which becomes black, brown, or red in colour. In some 

 species its surface has a dense and regular coriaceous, granular, 

 or rugulose sculpture, most conspicuous in the case of P. 

 cincrasccns. In some puparia there is no dif^culty in distinguisJi- 

 ing the anterior stigmata of the larva, though they are extremely 

 mnuite, but in others I fail to see them after the closest exam- 

 ination. Further back, however, on the dorsum there can lie 

 seen on each side a fine and minute process, apparently part 

 of the puparium, in reality not so, but connected with the deli- 

 cate cuticle of the enclosed pupa and merely perforating I he 

 puparium. When the fly emerges it frequently happens that 

 one or other of these processes is dragged back through the 

 hole in the puparium, but remains attached to the empty pellicle 

 of the pupa. (PI. VII. fig. 6). 



The dark posterior stigmatic area is in many species deeply 

 depressed, in others hardly at all, and bears one or more small 

 tubercles on either side, in fact it bears much similarity to the 

 same part in the larva. When the fly bursts from the puparium, 

 the latter appears to be always ruptured along the same lines, 

 in all the species we have examined, viz: along the front and 

 hind margin of the dorsal segment, which bears the anterior 

 processes, and along the hind margin of the ventral segment 

 corresponding with this. 



In the case of one species of Pipunculus, however, altliougU 

 the flv itself (P. ciitcrascciis) is not very remarkable, the larval 

 habits are unlike those of any of the other species, which infest 

 arboreal leaf-hoppers, since this larva does not fall to the ground, 

 and pupate beneath the soil. l)ut forms its puparium in the open 

 on the surface of living leaves. This puparium is very different 

 from the subterranean ones, for instead of the minute anterior 

 processes, are a pair of relatively enormous blunt ones, on the 

 tip of each of which is a fine spine-like process. It is not quite 

 clear to me whether the whole blunt process represents the min- 

 ute one of subterranean species, and really 'belongs to the pupa, 

 and not to the puparium; or wdiether the small apical spine-like 

 piece alone represents these. Posteriorly the stigmatic area is 

 large and deep, and instead of one or two minute tubercles on 

 each side, there are three, the upper one being very large and 

 conspicuous, the lowest one well separated from the upper two. 



