THE STRUCTURE AND HABITS OF ANIMALS 35 



who found it was to be attributed to the direct 

 action of cold and heat. He found that if the 

 pupae of Levana, which would ordinarily have 

 produced Prorsa, were put in a refrigerator 

 for one or two months the butterflies emerged 

 partly as Levana and partly as another form, inter- 

 mediate in many respects between Levana and 

 Prorsa. Thus the direct action of cold causes 

 pupae, which in the normal state of things would 

 have produced Prorsa, to produce Levana. 



The reverse process, however, is not successful, 

 for heat does not change Levana into Prorsa. The 

 explanation of this is, perhaps, not out of place, 

 although it has no direct bearing on the subject in 

 hand. Levana is the parent form, and in the 

 glacial epoch there was only one single brooded 

 variety and one generation in the year. The 

 lengthening of the summer gave rise to a second 

 generation, and the gradual action of climate caused 

 the Prorsa form to arise. Now, Levana being the 

 parent form, the effect of cold in causing pupae 

 that would normally have become Prorsa to become 

 Levana is simply a case of reversion. The reverse 

 action could clearly not occur. It is a long time since 

 the glacial epoch was over, and therefore probably a 

 long time since the two forms Levana and Prorsa 

 were established ; hence the reconversion by cold 

 is only partial. 



A similar example is afforded by another butter- 

 fly, Pieris napi } the green-veined white. This is 

 also double brooded, and occurs as two varieties, 

 each producing the other. The winter form is 



