Temperature Experiments on two Tropical Butterfiies. 449 



former to be the older, chiefly on the grounds that the 

 latter is widely different in appearance from any now 

 existing member of the group, and that in certain speci- 

 mens tlie remains of a former bar are more or less visible. 

 I do not know that either of these objections is insuperable, 

 but if so, I can only assume that the germ-plasm has now 

 become so fixed by inheritance that no form of shock can 

 throw it backwards to an earlier type, but only disturbs it 

 to such an extent as to cause it to produce the easiest 

 variation of which it is now capable. 



On the other hand, all experimenters on the earlier 

 stages of European Lcpidoptera, Weismann, Merrifield, 

 Standfuss, and others, lay great stress on the fact that 

 shock tends to throw the insect towards the ancestral 

 type, and I certainly know of no detailed experiments to 

 the contrary. It is scarcely reasonable that the same 

 agent would throw one insect back to the type, and another 

 to the form towards which it is tending. There is one fact 

 recently brought to my notice by Mr. Doncaster of great 

 importance. He tells me that he has received from 

 Coimbatore, in the Madras Presidency, a brood of bred 

 chrysippus in which were a considerable number of dorip- 

 pus, the parent being the type. This is the first instance 

 of such an occurrence, and it is the more interesting as 

 doripp^ts has never to my knowledge ever been taken in 

 Madras. I regard these dorippus as a throw-back to the 

 ancestral form. We shall see that we shall be confronted 

 with precisely the same difficulty wlien dealing with 

 Hypolimnas misippus. It is, however, clear that neither 

 form is a sudden mutation, but has been formed gradually 

 the one from the other. 



Hypolimnas misippus. 



Few remarks are needed regarding its life-history, but 

 I give the following from my note-book to show the rapid 

 growth of the larvae, and the remarkably short time during 

 which Natural Selection can have any influence on the 

 mature butterfly. It will be noticed that the female lays 

 the whole of her eggs in about ten days. 

 " 17.10.09. Captured diocippus $ in cabinet condition. 

 20.10.09. Noticed a considerable number of eggs laid. 

 24.10.09. Several larvae are hatched and evidently a 

 day or two old, if not older. Transferred 

 ^ to another cage. 



