CELASTKINA ARGIOLUS. 453 



the legs. The 1st legs have three lenticles and a rib-point on one leg, 

 three lenticles on the other, apparently at the tibio-tarsal joint ; the 

 2nd legs have a similar arrangement, except that the leg with the rib- 

 point is on the opposite side ; one of the 1st legs, and the 2nd on the 

 opposite side have one odd lenticle lower down (on a tarsal articula- 

 tion). On the 2nd, 3rd, 4th, 5th and 6th abdominal segments, there 

 is a curious puckering, in the centre of the dark spot (upper of two) 

 situated about halfway between the dorsum and the spiracle, on each 

 side. This, probably, corresponds with the " upholstered " hollow on 

 the side of each segment halfway up the "slope" of the larva. There is 

 no indication, at all definite, of scars of prolegs. The spiracles are, as usual, 

 accompanied by a little cloud of lenticles (pl.xxvii.,fig. 2), that on the 4th 

 abdominal has, for example, 25. The spiracles show a very narrow central 

 slit, with a wide margin, with radiating structural lines ; that of the 8th 

 abdominal segment only obsolete, the 7th normal. The 9th and 10th 

 abdominal segments are not distinguishable, and the 8th is narrowed 

 to evanescence, and fused with the 9th in the ventral line ( 2 pupa). 

 The circle, representing the 9th and 10th, has a transverse suture, 

 that one might take to be between dorsal and ventral plates of the 9th 

 abdominal, with the 10th abdominal buried between them, but that, 

 separate from it, at each end, is a short bit of suture that is, perhaps, 

 more probably part of that between the 9th and 10th abdominals ; the 

 dorsal part of this circle carries 30 odd hairs. The cremaster is repre- 

 sented by a number of hooks dorsal to the transverse median suture, 

 continued by a few hooks round the end of this, inside the scrap of 

 suture, to a further portion along the front margin of the circle. If 

 we assume all these hooks to be on the 10th abdominal segment, then 

 the 9th, like the 8th, is evanescent ventrally, both being represented 

 by a common projection, and the dorsal portion of the suture between the 

 9th-10th is smoothed out and non-existent. The cremastral hooks are 

 about 50 in number on the dorsal portion (pi. xxvii., fig. 1), 25 on each 

 side of the ventral ; each hook has an anchor-shaped end, set on, however, 

 in a plane oblique to that of the shaft ; they are about 0*05 mm. long. 

 Examining a mounted pupa, for the curious hair noted on the meta- 

 thorax of the pupa of Lampides boeticus {antea p. 355), I find it on both 

 mesothorax and metathorax, not far from the middle of each piece ; it 

 is not, however, after all, a hair; it looks like a short bit of twisted 

 ribbon, and its length is about that of a hair ; it is, however, on the 

 inside of the pupal- shell, not the outside, and must represent the lining 

 of some pore, drawn out on the emergence of the butterfly, just as 

 tracheal-lmings are. Except to note it as a remarkable structure 

 previously unknown to me, I have as yet no further light to throw on 

 it (Chapman). The pupa is about 8mm. long, 4-75mm. wide, of a 

 dumpy figure, thickest at the middle of the abdomen, with the head 

 and thorax rounded, and the latter very slightly keeled ; a depression 

 occurs between the thorax and abdomen where the cincture passes, and 

 this holds it secure ; thence the abdomen swells out full, and arched 

 towards the bluntly rounded anal end ; the wing- covers are long in 

 proportion, but not at all projecting. In colour it is pale brownish- 

 ochreous, with a blackish- brown thin dorsal line marking the thoracic 

 keel, and, on the abdomen, a series of rather blotchy arrow-head dorsal 

 dashes, and a subdorsal series of longer dark brown blotches, that 

 nearest the thorax being the more conspicuous, owing to the next 

 segment being without one. The thorax is marked with oblique rows 



