CHRYSOPHANUS DISPAR. 449 



and then are not joined together as one. They pass over the centre 

 of the metathorax of the larva obliquely down the side of the first 

 abdominal, and are fastened below the second abdominal segment. 

 In the pupa the girdle passes between the first and second ab- 

 dominal segments. A second larva, treated in the same way, 

 also chose the muslin cover to spin on, and was fixed up by 5 p.m. 

 on June 20th, becoming a pupa between 3 p.m. and 4 p.m. on June 

 21st. On June 30th, at 9.45 a.m., I saw that the first mentioned 

 pupa had yielded a female imago which had evidently just emerged. 

 She crawled a little way from the pupa-case along the muslin, and 

 then remained quite still. The wings, which had just commenced to 

 expand when 1 first saw her, had reached their full size in eight 

 minutes, though they were then still limp. The pupal stage thus 

 lasted nearly twelve days. The second pupa remained in that stage 

 for fourteen days, and the imago, also a female, emerged between 

 6 p.m. and 9 p.m. on July 5th. This appears rather a late hour for 

 a day-flying species to emerge (Sich). Two pupa3 examined : One is 

 loose, the other has formed a puparium from a portion of a leaf, now 

 somewhat shrivelled. It enfolds the ventral area, and a portion of the 

 dorsal, on one side only. A silk net, or loose pad, has been spun on 

 the leaf, and this forms the attachment for the anal armature, 

 composed of numerous, short mushroom-topped bristles, set at varied 

 angles. The anal end, which is comparatively smooth, has a broken 

 ring of these hairs surrounding a smooth depressed area; they occupy 

 the extreme end of the anal segment, which has a ventral aspect, and 

 are continued along the sides of the ventral portions of the 8th and 9th 

 abdominal segments, but are absent from the medioventral area of the 

 8th abdominal segment. The anal slit and sexual organs are very 

 obscure. I fancy, from the traces one can make out of the latter, that 

 the smaller loose pupa that I am now examining is a male. With 

 regard to the puparium, some silk has been used to draw the leaf to 

 the sides of the pupa and curl it upwards, making a saucer-like 

 hollow for the ventral area to occupy, no doubt the curling has been 

 somewhat exaggerated by subsequent drying. Some of the threads 

 used have been caught under the rays of the star-like processes (hair- 

 developments), on the lateral area, and are not, so far as I can 

 ascertain, continued across the dorsal area, but are doubled back and 

 reattached to the leaf. Others are continued and join up to form a 

 band or girth ; there is also a single thread which crosses separately to 

 the main girth. Neither of these supports lies in the dorsal groove or 

 waist, the main and upper one crossing at the middle of the 2nd 

 abdominal segment, and the single thread at the junction of the 3rd 

 and 4th abdominal segments, the waist, as usual, being at the 

 junction of the metathorax with the 1st abdominal. So far as this 

 particular pupa is concerned, the position seems to be that the cocoon- 

 making habit of its ancestors is represented by the few silk threads 

 used, and that a definite girdle has not yet been fully evolved. The 

 pupa spun up is, judging by its size, likely to be a female, the loose 

 one, as already mentioned, a male (Bacot). The larvae spin up with 

 the head downwards, and, in this position, with a silken band round the 

 middle of the body, change to pupae, those I had from July 24th-31st, 

 1892 (Nicholson). The notes of Sich and Nicholson suggest that a fairly 

 satisfactory girth is constructed [see also jjostea, p. 450] . Bartel notes that 



