LYCAENA ARION. 331 



Frohawk hints at the blossom and younger shoots of furze being tbe 

 foodplant, though why not thyme is difficult to see, and again 

 broaches the suggestion that the larva changes its food, although there is 

 no evidence whatever of the fact, and states (Ent., xxxix., p. 147) that 

 he thinks that "it is tendered by the ants (Lasius fiavtis), in the same 

 way as their own larvae are fed from mouth to mouth with food the 

 ants disgorge." There is no more evidence that this is so, than there 

 is for the supposition that they want feeding in the winter, or wish to 

 change their food when they simply leave the foodplant for hyber- 

 nation. Rothschild refers (op. cit., p. 172) to Frohawk : s mysterious 

 " food," and his hints that the larvae do not hybernate when they 

 leave the thyme, but require feeding, so that the creation of difficulties 

 in the life-history of this species has not struck us alone. 



[The whole of the foregoing section, including the footnote, is given 

 word for word as written by Tutt. As it is almost the only section 

 which he left ready for publication, I have not felt at liberty to change 

 it in any way, though I feel quite confident that, had he lived, it would 

 have been considerably modified. Much information is available now 

 which was not published when the above was written, and further, Mr. 

 Frohawk's view that the hybernating stage of the larva is its last has 

 been fully borne out by Dr. Chapman. It will perhaps be permissible 

 for me at this stage to express my own very decided view 7 that every- 

 thing points to the probability that the larva is, in its last instar, 

 carnivorous. It is now well-known that many Lycaenid larvae are so, 

 and that these belong to the Lycaenid, as opposed to the Plebeiid, 

 division of the "blues," while all known Ruralids have some tendency, 

 in confinement, towards cannibalism, which implies some degree of 

 carnivorous instinct. The Hon. N. C. Rothschild argued (Ent. Rec, 

 xxii., p. 4.0) that "it is absurd to assume that any portion of the larval 

 stage is really associated w T ith ants, as, were this the case, the numerous 

 nests that have been submitted to rigid and minute examination must 

 have yielded examples of the larva, and none have been found." "Mr. 

 Frowhawk himself," he adds, "is, we believe, convinced that his original 

 suspicions in this direction are really unfounded." Of course there is 

 essentially something "absurd" about all "assumptions," including 

 the assumption that if the larvae of L. avion had existed in ants' nests 

 they must have been found, but apart from this, the obvious answer to 

 Mr. Rothschild's contention is that both the full-fed larvae and the 

 pupae hare been found in ants' nests, and have been found nowhere else. 

 Mr. Frohawk's latest publication on the subject (Ent., xlvi., p. 321) 

 makes no mention of the fact that he has actually seen the larva of L. 

 avion eating both the pupa and the cocoon of an ant, and apparently 

 surrenders the position that it is carnivorous. His reasons for doing 

 so are however utterly insufficient, and indeed appear to me to tell 

 rather in favour of the hypothesis. It would seem probable that what 

 the larvae placed in Mr. Donisthorpe's observation nests w r ere searching 

 for was darkness, since, if they hybernate in the nests of Formica fiava, 

 itself a hybernating species, it is at least probable that they hybernate 

 also, and in that case experiments in observation nests are practically 

 useless, for either the larvae must fail to find darkness and retirement, 

 or they must cease to be under observation. Failing to find a suitable 

 place for hybernation they must feed or die, and if the carnivorous 

 instinct does not awake in them till after hvbernation, there is nothing 



