larva of Lycaena arion does during its last instar. 295 



all the undigested material and effete matters to accumulate 

 in the rectum during the whole period of growth, to be 

 ejected when the period for pupation approaches. In 

 the case of some ichneumons it is, if I recollect aright, 

 voided by the imago itself. 



This hypothesis is in itself a very remarkable one as 

 applicable to the larva of a butterfly, but it seems difficult 

 in any other way to account for the mass of hairs of 

 scabrinodis larvae which represent, obviously, a number of 

 individuals, that must have taken a considerable time in 

 consumption, very much beyond that, that butterfly larvae 

 usually pass between each act of defecation. The mass 

 also occupies its position in a way very unlike material 

 passed along the canal in the ordinary regular manner. 



If we adopt this hypothesis, then the division of this 

 mass into two portions raises further questions. 



The lower and therefore earlier portion gives no indica- 

 tion of what food it represents, the other later portion 

 represents many larvae of M. scabrinodis, all apparently in 

 their last instar. 



Does the first portion represent some different diet ? it 

 certainly does not represent full-grown larvae of the ant. 

 Does it result from the earlier food being ova or young 

 larvae of the ant that were more thoroughly digestible, 

 and so left no recognisable detritus ? Was the earlier diet 

 a vegetable one, as some of Mr. Frohawk's observations 

 suggest ? Or is some other explanation available ? as to 

 which one might speculate, but not very profitably, on 

 several. 



As this larva affords me a skin at a period when it is 

 not full grown but still not very far from it, it may be 

 worth while to compare it with the little larva that dis- 

 appears in the autumn, as the material for doing this 

 which I used in assisting Mr. Wheeler's history of the 

 species, though perfectly satisfactory and conclusive to 

 myself and probably to most other people, did not, after 

 all, provide any photographs otherwise than rather frag- 

 mentary ones, so that one or two from this specimen are 

 probably useful. 



These photographs also confirm a point already alluded 

 to more than once, viz. that this specimen was not full 

 grown. Comparing with the photographs in Tutt's " Brit. 

 Lep." of Mr. Rayward's larva, it will be seen that the hair 

 bases are still much closer together than they "become in 



