436 BEITISH BUTTEKFLIES. 



with each other by very fine lines, forming a network with cells of 3, 4, 

 or 5 sides. The pattern is, in fact, by no means very different from 

 that on the egg of this species. Yet it is difficult to believe that these 

 cells are not the same as those in the groups in which the skin-points 

 occur, not at the angle but, in tbe centre of the cell. If this is so, 

 the points themselves are not homologous with ordinary skin-points. In 

 any case, either the lines or points do not correspond with those 

 ordinarily observed. The resemblance to the pupal sculpture (in 

 pattern) is certainly close, and also to tbe network with .rosettes at the 

 intersections on the pupae, but, in the latter case, the scale is perhaps so 

 much larger as to make a difficulty ; nevertheless, there are many 

 indications that the ribs forming the pupal network are only a 

 selection of lines from a much finer reticulation, the other portions of 

 which are usually obsolete. The points are about 0-01mm. apart. 

 The long hairs of tubercle i are most conspicuous on tbe living larva, 

 forming a tall flowing dorsal crest, each hair about 03mm. long, 

 curving backwards ; they arise from the 2nd and 3rd thoracic, and the 

 lst-8th abdominal, segments ; on the mesothorax are apparently 

 two, one in front of the other, both referable to this series ; that on the 

 6th abdominal is perhaps the largest, nearly 0-4mm. The bases are 

 large, tapered, about as high as broad (about O03mm. or O04mm.). 

 The hairs are clothed with numerous points, very slightly raised, but 

 with decurrent bases ; ii carries a much smaller hair, half the length and 

 thickness of that on i, placed a little outside and behindit; it is rather 

 larger on the front segments, it exists on the 2nd and 3rd thoracic and on 

 the 1st- 6th abdominal segments ; on the 7th its place is occupied by a 

 lenticle (the upper of the two subdorsal ones). Not apparently on the 

 mesothorax, but on the metathorax and on the lst-7th abdominal 

 segments there is a minute hair at the front margin of the segment, 

 almost directly in front of ii. The supraspiracular hairs, near place 

 of iii, one or other of which may represent the seta of that tubercle, 

 are long and club-shaped ; one only is present on the metathorax and 

 the 7th abdominal, but both are present on the lst-6th abdominal 

 segments. As to position, the larger is in front and higher, the 

 smaller is lower and directly above spiracle. [A line round the 

 middle of the segment, or, rather, such a line pushed a little forwards 

 dorsally, would pass through tubercle i, the two lenticles (not yet 

 noticed), the posterior supraspiracular, the spiracle, and the middle of 

 the three lateral hairs.] They are very similar in structure, very 

 colourless and delicate, and difficult to see well either in the living, or 

 prepared, larva; they are smallest where they arise from a little 

 rounded base, gradually enlarge, and finish with a rounded end ; they 

 are faintly spicular like the other hairs, and have a considerable 

 curvature ; the longest, but least clubbed, is that on metathorax, about 

 O09mm. ; the shorter average about O'Oyium., the larger, 0*06min., 

 or O07mm. [On larva of Polyommatus icarus they are little more 

 than 0-01mm.] The lateral hairs (iv and v (?) on the subspiracu- 

 lar-flange) are four on the 2nd and 3rd thoracic segments, three on 

 the lst-8th abdominal segments, possibly also on 9th, but these are not 

 readily distinguished from those on the 10th ; the middle one (already 

 alluded to) is the longest, 0*1 6mm. or 017mm. long, the posterior, 

 the shortest, about O07mm. long and is not produced to a point like 

 the others, but has a rounded end and is very slightly clubbed, it is 



