334 BRITISH BUTTERFLIES. 



some ran over it. He says that directly an ant's foot touched 

 the gland or the skin in its near neighbourhood, it throbbed 

 more violently, swelled up, and ejected a globule of clear white 

 liquid, which was instantly licked up by an ant; in a few seconds 

 a foot again touched the gland and another bead of liquid oozed 

 out, which was at once licked up by an ant. He observes that 

 the larva took no notice of the ants running over it and around it 

 whilst it kept feeding, the gland being apparently only so extremely 

 sensitive to the touch of an ant's foot ; touching with a sable-hair 

 brush had no result, the larva only winced and contracted, nor could 

 any artificial imitation produce the secretion, but directly an ant's 

 foot, or the claws of the foot touched it, a bead would appear, and at 

 once be imbibed by the ants. Although the larva was kept in a box 

 with numerous ants, both workers and winged £ s* together with 

 their pupa?, the ants all acted precisely similarly, not one attempted to 

 bite the larva, but as soon as they touched it they slowly closed their 

 jaws, and waved their antenna? over and upon it (Entow., xxxvi., pp. 

 59-60). 



In the spring after hybernation, and probably also during the 

 winter, it sometimes, at least, lives in association with ants, and on 

 Jane 3rd, 1906, four larvae were found in orie nest of an ant, Formica 

 flava, by Mr. F. W. Frohawk and myself, in north Cornwall, one being 

 quite fully grown, and the other well advanced in the last instar. 

 All were near but beneath the surface at the crown of the ant hill, and 

 w r ere living among a large colony of ant larvae and pupae, with which 

 they were apparently on perfectly friendly terms. 



In colour they very closely resembled the ant larvae, being of the 

 same pale semi-transparent white ; there was traceable, however, a 

 slight shade of ochreous, which was not seen in the ant larvae, and 

 this became rather more pronounced after the larvae had been removed 

 from the nest for a few hours. 



The ants' nest in w r hich they were found was situated on the 

 sheltered and sunny side of a low gorse bush, the top of the nest 

 crown being some three or four inches above the surrounding surface 

 of the ground, and the entire surface of the hillock was overgrown 

 with thyme and short grass. 



Although a very large number of nests were opened by Mr. 

 Frohawk and myself, only in this one instance were avion larvae found, 

 and as the locality in which our search was conducted is the head- 

 quarters of the species, which is usually plentiful there, it appears 

 probable that the large ant hills are not the only, or indeed the chief, 

 home of the larvae. The whole hillside on which avion occurs in this 

 locality is so thickly infested with ants, however, that one can scarcely 

 turn over a stone or stick without disclosing a large or small colony, 

 and if the avion larva were wholly dependent upon the ants for its 

 food after its last moult in the autumn, it would rarely have to crawl 

 more than a few inches from the flowers of the thyme plant upon 

 which it had until then fed, before coming within reach of the new 

 source of supply. 



Assuming that the larva is so dependent upon the ants, and lives 

 for a period of its growth amongst their larva? and pupae, it is never- 



* ? Winged ? s in August.— (J. W.T.) 



