180 Dr. T. A. Chapman on the 



appreciably, in fact about double their size, besides going 

 through a whole instar and making a moult. 



I am afraid that in my previous observations, made in 

 the first place hurriedly in the field, and not afterwards 

 properly verified, I mistook the heads of the cast skins for 

 larval frass, and assumed some such amount of frass to be 

 what one would reasonably expect after their first meal of 

 the maternal tissues. In H. paradoxa one finds, however, 

 nothing but some threads, the cast larval skins (skins as 

 well as heads), and the pellets of urates, which from their 

 small number and comparatively large size are obviously 

 maternal and not larval (Ubris. The young larvae there- 

 fore go through a whole instar, more than double their 

 size, undergo a moult, and then perforate the pupa skin, 

 and make their way through the cocoon without ejecting 

 any frass or effete material. It seems to follow that the 

 maternal remains must consist almost entirely of com- 

 pletely assimilable materials. As the larva? on facing the 

 world have to begin life in many cases by making a long 

 journey to find their food-plant, a meal before starting is 

 an obvious advantage, but this does not explain the 

 curious details of the process, or why it should differ in the 

 several species. 



I should note that I have examined just now pupae of 

 paradoxa and canalensis, but have no adequate specimen 

 of penella by me. With regard to it, therefore, I merely 

 assume from my previous observations of it that it agrees 

 with canalensis. 



It results that at corresponding instars the larvae of the 

 several species are very much alike, it also follows that 

 the discrepancy between Mr. Fletcher's observations as 

 to the hibernating instar of H. penella and mine of 

 H. paradoxa does not exist, though it is my fault that 

 there appeared to be one. 



The distinction I drew between the first instar larvae of 

 II. penella and H. paradoxa does not exist as a structural 

 difference at all ; nevertheless there is a difference between 

 them in habits and instincts, which is probably of quite as 

 great importance. 



The only other matter I observed this year at La Granja 

 as regards IT. paradoxa was that it had two widely-separated 

 habitats, one at about 5200 ft. elevation, where its food- 

 plant was Adenocarpus hispanica, the other at about 

 7000 ft. on Gytisus purgans ; the opportunity was wanting 



