158 PEOF. W. H. JACKSON ON THE 



individuals do not develop quite at the same rate. My caterpillars taken on July 10, 

 pupated between July 13 and 17 ; the butterflies emerged between August 1 and 4, 

 i. e. they were 20 to 23 days in the pupal condition, or as a possible maximum 23 days, 

 and as a possible minimum 16 days. The second batch taken July 14 pupated between 

 July 18 and 22, and emerged between August 3 and 7, i. e. they were in the pupal state 

 17 to 21 days, or as a possible maximum 21 days, and as a possible minimum 13 days. 

 It may be added that in both sets of instances euaergence was quite irregular, or, in 

 other words, those which pupated first did not necessarily emerge first. It is stated by 

 implication in Buckler's ' Larvse of the British Butterflies and Moths,' edited by 

 Stainton, Ray Soc. 1886, i. p. 179, that the pupal state of this butterfly lasts 13 to 15 

 days. The great prolongation of time in some, at least, of my own specimens was due 

 beyond a doubt to the cold weather that set in just after they became pupae, and the fact 

 that emergence of the whole batch took place within the limits of one week was caused 

 by the sudden heat that began on July 29 and lasted through the first week of August. 



(1) The full-geown Caterpillar. 



Tlie larval ovaries are situated in the fifth abdominal somite, and close to the dorsal 

 middle line, as in caterpillars of other species. Their proximal or attached extremities 

 are approximated and they diverge from one another posteriorly. In colour they are 

 whitish, with a yellow tint over their proximal halves, a tint which deepens in intensity 

 during the quiescent period preceding pupation, and during the first few days of pupal 

 existence. They are cylindrical in shape, but when the neighbouring parts of the fat- 

 body are drawn away their outline becomes somewhat irregular owing to traction on the 

 delicate filaments which connect their edges to the adjacent lobules of fatty tissue. 

 Four opaque white lines, the future ovarioles, traverse the larval ovaries lengthwise and 

 converge to their hinder extremities, I'rom which spring the larval oviducts. The latter 

 are very delicate filaments, extremely hard to see. If the caterpillar is still feeding they 

 are generally free for their whole course or involved in the fat-body only here and there, 

 but the nearer the quiescent period approaches, the more involved in it do they become. 

 They run backwards more or loss parallel to one another in the latero-dorsal region; 

 they approach and pass between the tracliea3 arising from the seventh pair of stigmata, 

 and are tied, eacli by a delicate filament, to one of the main tracheal stems. They then 

 turn abruptly towards the ventral aspect of the body and are attached to the hinder edge 

 of a trachea. This trachea arises from the seventh stigma on each side, runs inwards 

 beneath the musculature of the seventh somite, and unites with its fellow in the ventral 

 median line. Each oviduct accompanies the transverse trachea of its own size as far as 

 the ventral edge of tlie ventro-lateral longitudinal muscle-bundles. It then takes a 

 curved course beneatli the ventral longitudinal muscle-bundles to tlie spot where it is 

 attached. This attachment is close to the posterior edge of the seventh somite, internal 

 to, i. e. nearer the middle line than the attacliments of the ventral muscles. A small 

 space separates the attachments of the two oviducts. These facts are illustrated 

 PL XVT. figs. 21', 30, 31, and are explained in the description of the figures. The 



