RTJRALIS BKTVLM. 293 



diverge slightly, and then more rapidly on the thorax, 'joining the 

 flange line at front corners of the prothorax. The skin is'very glassy 

 and transparent, and the skin -points are represented by fine com- 

 plicated puckerings. It is to be noted that, in these larva3, the first 

 stage has numerous black skin-points, wanting in second stage, and 

 represented (then and) after by a network of lines of pavement- 

 epithelium pattern. The prothoracic plate has at least one (coloured) 

 hair ; all the hairs are finely spiculated. Hairs on the slopes very 

 short, baton-like, and deflexed, not quite the same on all segments, 

 sometimes one, sometimes two, at front margin in front of spiracle, 

 one at posterior margin, and two on middle of slope ; the three dorsal 

 hairs are supplemented by two or three small ones on dorsal flange, 

 behind the others. The lateral flange with ten or eleven hairs. On 

 the thorax are three hairs on the dorsal flange of metathorax, two on 

 niesothorax, with one below and behind, and two at front margin of 

 segment. The widely separated dorsal flanges (?) on prothorax, 

 carry three or four hairs ; these are yellow, oblique (downwards and 

 backwards) on the slope, and the flanges are also yellow ; these two 

 dorsal yellow lines are only about 0-12mm. apart, but are not what 

 we ordinarily call dorsal lines, being the lines of i and ii, not lines 

 much interior to them. There are eighteen hairs on each side of the 

 lateral flange of the conjoined 7th to 10th abdominals, and sixteen on one 

 side of the prothoracic margin. The hooks of the prolegs are very trans- 

 parent and faintly tinted, apparently five on each pad. Seen laterally, 

 the highest point of the larva is formed by the tops of the 3rd thoracic 

 and the 1st abdominal segments. Thence it slopes acutely to the front 

 margin, gradually behind to the 6th abdominal, where it has sunk 

 about one fourth, thence it slopes more quickly to the end ; from the 

 metathorax to the 6th abdominal the segments are rounded dorsally. 

 The prothorax and the 7th to 10th abdominal segments on the other 

 hand, slope smoothly. Now, when fullgrown, in this second instar, 

 the larva is in form and colour almost identical with the fullgrown 

 larva, the chief difference being that the primary hairs are still more or 

 less recognisable ; but, in the extreme narrowness of the dorsal area 

 and the length (or height) of the slopes, the identity is close. The 

 first oblique lines become longitudinal on the 2nd, and wanting on the 

 1st, thoracic segment. The second one is straight on all three thoracic 

 segments, and almost wanting on the first. There are some short 

 yellow lines inside the dorsal lines, two on the prothorax where the space 

 is greatest. Third instar (April 15th, moulted on the 14th) : 8mm. 

 long ; much the same appearance as before moult ; dark green, with 

 yellow lines, the yellow is not very clear yellow, but is in sufficient 

 amount to give the larva a general effect of light green. The two dorsal 

 lines (and ridges) are very close together along the abdominal segments, 

 so close as to make rather a single (but divided) ridge than a flat, 

 with raised margins; the "slope" is very long, especially on the 

 1st and 2nd abdominal segments. The dorsal ridge on abdomen 

 is about 0-15mm. wide, at front of prothorax about l-0mni. The 

 slope at anterior abdominal segment measures l-4mm. or l # 5mm. 

 The dorsum rises above the level of the lateral flange behind, up to the 

 middle of the 1st abdominal segment, whence it slopes down again to 

 the flange in front, at the same time that the dorsal ridges and lines 

 separate, and along the prothorax are l-Omm. apart ; the front of 



