304 Dr. T. A. Chapman on 



a comparatively recent ancestry with denexed wings, we 

 have here to go back to the Skippers, or possibly the 

 Castniag, for such ancestors, and one would regard it as 

 probable that any memory of a denexed position of rest 

 had been entirely lost. Possibly this is so, yet in the 

 process of expanding, or rather of drying the wings after 

 expansion is completed, there is a phase, that does not 

 seem to occur in the Heterocera, that strongly suggests that 

 some reminiscence of the denexed attitude still exists, but 

 has been pushed backwards from the point just following 

 drying, till it has become involved in the period of drying. 

 In any case, whether it admits of this interpretation or 

 not, this difference of procedure is obviously of interest, 

 and seems to require some explanation. In butterflies, the 

 wings are expanded, as indeed in most Heterocera, in 

 approximately the resting attitude, but, when expanded, 

 are definitely placed in the drying position, all nearly 

 parallel, hind-wings close against fore-wings, but the fore- 

 wings only touching at their tips and hind margins. Then 

 begins the special butterfly process : after a few minutes 

 the wings are separated, to such a degree that were they 

 stiff the upper surfaces of the opposite wings would be at 

 an angle of 90° at least with each other, but being limp 

 the wings hang in somewhat bell-shaped fashion, as ob- 

 served on the costal aspect. They are retained here for 

 some thirty seconds, and again closed; they are then 

 opened and closed, each phase occupying about half a 

 minute, but varying a little, for some six to ten times, 

 the wings gradually stiffening, not being opened so wide, 

 and so losing the bell-form, the wings being nearly straight. 

 In the last open positions the hind-wings sejDarate from the 

 fore-wings, so that there is an interval of a full millimetre 

 between each adjacent pair of wings; finally the wings 

 are closed, so that rather more than the tips touch, and 

 very gradually are more closely approximated, so that 

 the costae coincide for fully two-thirds of their length. 

 This is the assumption of the true resting attitude; no 

 further change of position occurs until the insect is dis- 

 turbed, or takes flight, etc. This description refers more 

 particularly to Pier is rapae, than to the other species 

 observed. 



This is a very different process from the typical Hetero- 

 cerous one of suddenly throwing the wings back, keeping 

 them there immovably till they are dry, and suddenly 



