112 



BKITISH BUTTEKFLIES. 



front, the gradation from one to the other being somewhat abrupt. 

 On the lateral view, nevertheless, the anal apex is nearly a right angle ; 

 viewed from behind, and a little above, it also has a trace of angularity. 

 The dorsal line is so steep posteriorly, that the anal extremity is hardly 

 visible from a directly dorsal view. The face is very ventral and the 

 suture across the base of the maxillae and first legs is about 2 -5mm. from 

 the front. The maxillae (there is no trace of labium) extend down for 

 about 3'3mm., whence, for another 3mm., the antennae, side by side, 

 continue to the end of the wings. The first legs are very broad and 

 short (about 2mm. long), the second legs are about 2mm. long, 

 narrow, each end tapering to a point ; the upper apex is shut out from 

 the eyes by first leg by about the same distance as the lower is from 

 the ends of the maxillae (about - 6mm.). The labrum and mandibles 

 are three small portions, of about equal size (some 0-22mm. long), the 

 labrum triangular, with blunt, rounded apex, the mandibles with 

 apices opposed in a straight line of about l*8mm. in length. The 

 wings show no definite " Poulton's line." In the specimen examined, 

 the 7th and 8th abdominal spiracles are both abortive, and the 8th, 

 9th, and 10th abdominal segments are so fused together as to be one 

 piece, no sutures between them being discoverable. Examining the 

 anal area of this specimen, a $ , there is, ventrally, a very short bit 

 of suture, apparently that between the 8th and 9th abdominal seg- 

 ments ; behind this is a small, smooth, or faintly-wrinkled, area 

 (05mm. in diameter) ; at the front of this is a small longitudinal 

 mark, and, behind it, the anal scar, with a good deal of darker chitin ; 

 on either side are about fourteen cremastral hooks scattered irregu- 

 larly. These hooks are pale brown (chitinous), about 0-lmm. or 

 O'ISmm. long, and have a double, or anchor-shaped hook at the 

 extremity. They have this special peculiarity that they arise from the 

 skin -points (from which elsewhere no sort of hair arises), these skin- 

 points being united by ribs as in other situations, and quite disso- 

 ciated from true hairs, of which several are mixed up with these 

 cremastral hairs. C. rubi has, therefore, a definite cremastral struc- 

 ture, though I have not found it taking any attachment to anything. 

 In the living pupa a good deal of the surface-sculpturing can be made 

 out, and some of the coloration, but it is not till the empty case has 

 been rendered transparent (as by " Canada balsam ") that it is quite 

 easy to see it to be of two tints, a paler brown chitinous ground- 

 colour and a darker one in spots, somewhat scattered on the 

 wings and appendages, but more massed elsewhere, especially 

 dorsally, when the spots become fused into large patches. The 

 hairs are universally distributed, except over the front of the 

 head, the appendages, and tha medioventral portions of the 5th, 

 6th, and 7th abdominal segments. They are long (about 0-35mm.), 

 black (i.e., deep brown), of fairly uniform diameter from end 

 to end, acuminate at point, and armed with minute spicules, which 

 stand out almost at right angles to the stem of the hair. They are 

 wanting on the basal third or half of the hairs. Round the spiracles 

 especially are many lenticles, nearly always distinguishable from a 

 hair-base that has lost its hair, by its smaller lumen, sometimes 

 apparently closed by dark chitin, but the larger ones closed by a finely- 

 dotted membrane. The third structure, which appears to occur on 

 nearly all Lycaenid pupae, is the series of skin-points, not altogether un- 



