■} 



NISONIADES TAGES. 283 



The way in which the prothorax, on dehiscence, remains attached to 

 the mesothorax is remarkable. It is a piece about l*6mm. across, and 

 0'6mm. from back to front, square at dorsum, but sharpened to a point, 

 like the prow of a boat at the outer end. When dehisced, the inter- 

 segmental membrane is quite a substantial plate, folded down the 

 middle ; the two portions, flat against each other before dehiscence, open 

 out, each portion about 04mm. wide, but the folds do not quite 

 straighten, so the pro- and mesothorax separate to something less than 

 the possible 0-8mm., but are held firmly together. The spiracle-cover 

 on the mesothorax is a large black knob, about 0-4mm. in diameter, 

 covered with very closely-set crowded hairs, very short and thick; giving 

 rather a velvety appearance. The dorsum of the mesothorax, as well 

 as the prothorax, and, in fact, the rest of the pupa (appendages 

 being only excepted), carries a fair number of fine transparent hairs 

 about 0'12mm.-O18mm. in length. Across the posterior portion of 

 the mesothorax is a moderate number of transverse ridges or wrinkles 

 (little more than microscopic), and the hairs are, to some extent, along 

 these. On the mesothorax, these fine ridges are a marked feature ; they 

 are narrow, with parallel sides, and, in many cases, curve forwards in 

 regular waves, so as to leave little hollows behind them, in which the 

 hairs arise ; in some cases, there is a fainter ridge close behind the other, 

 which curves backAvards at the hairs, leaving the hairs in a little oval 

 space between them. The hindwing is narrow, and, passing the 1st 

 and 2nd abdominal segments, ends just before the spiracle on the 3rd 

 abdominal. The 1st abdominal segment has hairs and ridges like the 

 metathorax ; except along the posterior border of the segment, the 

 ridges are ill-developed, though, at many of the hairs, the two 

 ridges bending round, one in front and one behind, are specially distinct, 

 and have the appearance of two eyelids with the hair as the eye. An 

 attempt to count the hairs on this segment makes them nearly 40 on one 

 side. They are most numerous in the middle and at the posterior border 

 of the segment, and are decidedly fewer dorsally than laterally. The 2nd 

 abdominal segment is similarly ridged, the ridges being more conspic- 

 uous along the anterior border of the segment, and the hairs more nearly 

 in transverse rows. The spiracle has a fine series of arched ridges 

 passing round it from the wing-margins above and below, with the 

 usual appearance, as if the wing had pushed the spiracle backwards, 

 and, in so moving, it had wrinkled up the surface in front of it. The 

 3rd abdominal segment is much the same, except that the ridges are 

 less distinct ; the hairs are few dorsally, and look as if there were two 

 rows towards the front margin, another row medially, rather irregularly 

 disposed, with two posterior rows, one marginal, the other a little 

 forward of the margin, but a closer scrutiny shows hairs that would 

 mean twelve rows if each hair was in a row, so that the five rows noted 

 are only those where the hairs seem to fall most in regular line. On 

 the 4th abdominal segment the spiracle is clear of the wing, and without 

 puckerings. Here, also, the hairs are an anterior, a middle, and a 

 posterior set, each approximately in two rows, but with other hairs 

 elsewhere, representing nearly obsolete rows; some of the posterior 

 hairs are on the intersegmental membrane. There are no distinct 

 ridges, but there are many minute raised circles (or ovals), almost like 

 hair-bases, but without hairs. These are more numerous and more in 

 transverse lines towards the front of the segment (Chapman). 



