( ciii ) 



pupa hardens until near the end of the pupal period, 

 Experiments tend to show that a winter pupa is in a state of 

 almost absolute torpor for a long period,* perhaps up to the 

 time when it has been subjected to cold sufficiently long 

 to bear the application of a forcing temperature, but 

 that, after that time, though the cold may continue, the 

 development begins to move very slowly forward, for it will 

 emerge in a shorter time if the forcing is deferred until long 

 after that critical period has been surmounted, than if forced 

 at once. In the case of summer pupa?, there appears, however, 

 to be but a short period, if any, of torpor ; my experiments with 

 S. tetralunaria have shown that the pupal period of this 

 summer phase can, by icing, be protracted to 22 weeks, and 

 that changes in the facies of the moths that came from pupa? 

 transferred to a warmer temperature began in six weeks, i. e. 

 long before the middle of the pupal period. 



" Direct " effect of temperature. 



But besides these changes, which — at least most of them — 

 cannot be ascribed to the direct effect of cold, there are others 

 which may be called by way of distinction direct, and these I 

 have found to be produced mainly (in the case of some species 

 I may say exclusively) in the latter part of the pupal period. 



I do not think sufficient regard has been given to what may 

 thus be called the " direct" effect of temperature during the 

 pupal period on the colouring of Lepidoptera. But there is 

 no manner of doubt that it has such an effect, and in many 

 cases this effect is so considerable that it may have protective 

 value. As already stated, many of the seasonal differences, 

 and those much the more striking, must find an entirely 

 different explanation, that is to say, in the hereditary pre" 

 disposition of the organism called into operation by the 

 appropriate stimulus (as the fur of the Arctic hare is whitened 

 by the advent of winter cold) ; for example, the differences in 

 A. levana and, to take a marvellous example, those in a species 

 of Precis, the one phase, P. octavia-natalensis, being on the 



* The torpor often lasts for years, as in the numerous cases where 

 winter pupa? lie for several years in that stage. 



