314 DAVENPORT ACADEMY OF NATURAL SCIENCES. 



DEVELOPMENT OF THE FEMALE PUPA. 



Tlie larvje destined to become females undergo only very slight 

 clianges of form. Up to the stage represented in fig. 25, they are not 

 distinguishable from those of the males, but while those become narrow 

 and long, these become much broader in proportion to the length, usu- 

 ally about two-thirds as broad as long, and with the posterior (abdomin- 

 al) portion very much broader than the anterior, this change being 

 caused no doubt by the development of the ©varies. They also differ 

 from the males in remaining quite flat with only a slight dorsal carina. 

 These characters soon render them easily distinguishable from the male 

 scales. 



Just before the appearance of the males, the female pupa (if such it 

 may be called) is seen to be supplied with a new pair of sete coiled up 

 spirally on each side of the head (fig. 43) just as in the larva above des- 

 cribed, thus indicating a moult, although, as in the previous noulting 

 time, there is no appearance of a true shedding of the skin. Up to this 

 time the females have remained of a pale yellow color ; they now become 

 adorned, just in time to attract the attention of the males, with deep red 

 markings, rendering them much more handsome than at any other 

 period of their life, before or after. 



The female to all appearances remains a larva to the end of her life, 

 but I speak of this as the female pupa, because it corresponds with the 

 same stage in the development of the male. 



THE FEMALE. 



The female (fig. 32) is scale like, depressed, approximately oval in 

 form, widest behind the middle, with the anterior tapering more rapidly 

 than the posterior portion ; about one and three-fourths times as long as 

 broad, (UOO />« to 1800 /^ long by 800 p- to 1100 /^ broad). A distinct but 

 rounded and slightly elevated carina extends from the head to the anus. 

 The margin is entire except slight emarginations at the eyes and 

 opposite each of the four spiracles, and a deep fissure extending to the 

 anus. The margin is furnished with a large number of spines, (about 

 120) set at nearly regular intervals, and each secreting an elongated 

 waxen cylinder or filament, the whole, when unbroken, forming a whit- 

 ish fringe, not usually very prominent. In each spiracular emargina- 

 tion is a larger movable spine, supported by two smaller spines, which 

 present a different appearance from the others and do not appear to be 

 secretive in their nature, though I am not perfectly certain that they 

 are not. 



The dorsal integument is dense and tough and the divisions of the 

 segments are much obscured. The head is deeply set into the thorax, 

 and the head and thorax together are about equal to one half of the en- 

 tire length. All of the segments of both thorax and abdomen are longer 

 at the sides than in the middle, and in the seventh segment this differ- 

 ence is very great, so that it appears formed of two large lobes meeting 

 on the median line, surrounding the eighth segment and forming the so- 

 called " anal fissure." The eigMh segment is entirely surrounded by 



