72 THK entomologist's RECORD. 



its upper visible segment close behind and below the glazed eye, and 

 between here and the maxillffi, on the upper edge of the prothoracic 

 segment ; from this point the next segment (the tibia) descends ob- 

 liquely backwards, its lowest point being situated near the thick dark 

 medioventral line before alluded to, at the junction of the meso= and 

 metathoracic segments ; whilst the last segment of the leg (the tarsus) 

 terminates upon the 2nd abdominal segment, lying close against the 

 broad dark medioventral line. The upper visible segment (the femur) 

 of leg 2 commences at about the centre of the mesothoracic segment, 

 against the tibia of leg 1. The junction of this segment of leg 2 with 

 the tibia is not definitely visible through the pupacase, but the tibia 

 terminates on the anterior edge of abdominal segment 1. From thence 

 the tarsus extends, packed closely against the broad medioventral line, 

 to the centre of abdominal segment 3, where it terminates. The 

 detail of leg 3 could not be made out, as it was packed beneath and 

 behind the wing-costa ; but, as the pupa advanced towards the emer- 

 gence of the insect, its lowest joint could be faintly discerned through 

 the wings on the posterior portion of abdominal segment 2, from 

 whence its last member doubtless lay beneath the wing-costa, but its 

 termination could only be guessed at. 



The wing-casing extended to the extreme posterior edge of 

 abdominal segment 4, where it was carried slightly round the full 

 edge of the segment almost to its deeply depressed division from the 

 next segment. From here it ascended upwai-ds and forwards in a 

 graceful curve to an angle on the lateral centre of abdominal 

 segment 8, the point of the angle impinging upon, and covering, 

 the lower portion of the spiracular blotch there. It then was carried 

 forward in a hollow curve to the shoulder on the thoracic region. 

 The ridges covering the wing-nervures were well-defined, and were 

 firmly, though delicately, traced upon the surface, and the marginal 

 edging to the wingsheath was sharply and neatly incised. The lines 

 upon the head and thorax marking the structural divisions of the 

 casing were defined in thin brown lines, slightly, but sharply, depressed 

 at their edges. The habit of the pupa became much quieter after its 

 first week of existence, and it would seldom move or show signs of 

 life when the covering of flannel was lifted to inspect it. It very 

 gradually deepened in colour to an appearance of distinct sienna 

 colour, especially on the dorsal surface, and by October SOth seemed 

 to have reached its final stage in this direction. The ventral surface 

 and the wing-sheathing remained fairly light and clear in colour, the 

 latter possessing very little of the freckled appearance which was 

 noticeable on the other portions of its surface, and which had now 

 turned dark, almost to blackness, on the anterior part of the dorsal 

 area of the segments. 



On November oth the pupa had changed scmiewhat in appearance, 

 and had lost its transparent and shiny look. The whole of its surface 

 was opaque in eftect, and, while being warmer and more ochre- 

 coloured than putty, it suggested the dull, dead surface of it. The 

 ventral portion of the thorax containing the legs was lighter and 

 rather greyer in colour than the rest of the pupal skin, hut showed no 

 tendency towards the colour markings of the imago yet. The darkening 

 of the pupa from this time onwards was very gradual, and the first 

 portions to turn in colour were the head and eyes. These parts 



