330 BRITISH BUTTERFLIES. 



appears to have judged the moult by the size of the larva ; however, 

 he notes that, after the third moult, the larvae persistently refuse to 

 remain on the blossoms or any other part of the plant, and appear to 

 have a tendency to hide in the ground. Nothing more, however, was 

 seen of these lame (Ent., xxxii., 104-106). In 1902, Frohawk obtained 

 further larvae, and, on August 8th, states that he had larvae in four 

 different stages, some only just hatched, one over the third moult, 

 others fixed for third moult, a large number after the first and second 

 moults, and adds that after each moult, they ate part, and in some 

 cases nearly all, the cast skins. He further noted that on August 11th, 

 many having passed their third moult, when they cease feeding on 

 thyme (one supposes simply because they are ready to hybernate), he 

 experimented on them and ants, thinking they might feed on the 

 larvae or pupae of the latter (though why he should think this, because 

 the larvae left off, as do so many " blue" larvae at this time for hyberna- 

 tion, is not stated) ; it is, therefore, the more remarkable, that one 

 larva, supplied with an ant's cocoon with one end removed, at once 

 began eating it, feeding on the jelly-like substance of the pupa 

 as well as the cocoon, which it ate in the same manner as it would 

 a leaf, by biting the edge; it fed for several minutes, and he states 

 that he thought he had found the right food for their subsequent 

 stages, but this proved not to be the case ; one wonders, indeed, what 

 other food they could want during hybernation. By accident, Frohawk 

 then found that the larvae were myrmecophilous, for putting a larva 

 in another box with four ants of Formica fiava, they ran to it, waving 

 their antennae over and upon it, apparently smelt and licked it, and 

 seemed particularly attracted to the dorsum of the tenth segment, and he 

 then describes, as happening in L. avion, the symbiotic connection long 

 before discovered by Edwards, de Niceville, etc., with larvae of 

 Celastrina argiolus, and many other species. He notes that the larvae 

 appear to be perfectly at home with the ants, neither molesting the 

 other; and there appears to be no attempt at cannibalism at this time, 

 although, on one occasion, July 29th, 1902, two small larvae were 

 observed rolling together under the thyme-blossom, the smaller one 

 of which had seized the larger with its jaws, which were buried in its side 

 apparently sucking it, and this cannibalistic habit he considers was the 

 possible cause of so many mysterious disappearances. Frohawk, in 

 some inexplicable manner, considered that L. avion larvae wanted 

 feeding during the winter, and he mysteriously states that he 

 discovered, after trying over three dozen different plants, a certain food 

 upon which he induced them to feed for many weeks, during which 

 time they slowly grew from J to Jin. in length. One isn't surprised 

 to find that this interference with the hybernating larvae, and the 

 unnecessary ultra-care, killed all the larvae off, although they are said 

 to have doubled in length under the treatment ! At last, Rayward 

 and Frohawk, in June, 1906, discovered the full-grown larva in one 

 of its well-known haunts in Cornwall, hiding in an ant's nest. Four 

 were turned out of the same small portion of an ant's nest, only just 

 beneath the surface among the roots of the little plants of grass 

 growing with the wild thyme, the soil surrounding them being loose 

 and friable, worked up by the ants ; in company with the larvae of 

 L. avion were ants and their larvae and pupae, but a great deal of time 

 spent in searching other nests was attended by no further success. 



