THE AESTIVATION AND HYBERNATION OF BUTTERFLY LARVAE. 9 



all commencing to feed at the same time in the spring, and producing 

 imagines together, just as Vandouer described 80 years ago in B. 

 euphrosj/ne. Edwards notes (Can. Ent., xxi., p. 42) that, of a batch of 

 young larvae of Colias meadii, hatched July 23rd, 1888, one larva alone 

 fed on, pupating August 19th, and producing an imago August 25th, 

 whilst all the other larvae of the brood fed up slowly till after the third 

 moult, commenced to hybernate on August 28th, and did not produce 

 imagines till the following year. 



Prideaux notes (Ent. Rec, xviii., p. 246) a peculiar case, in which 

 a larva of Loweia alciphron var. gordius, with twelve others, com- 

 menced hybernation in November, 1905. The larvae all recommenced 

 feeding in the following March, and by May 25th, 1906, all, except one, 

 began pupating, and the imagines appeared at the end of June. This 

 one larva, oddly enough, ceased feeding after its last moult, shrank 

 somewhat in size, attached itself to the side of the cage, and remained, 

 aestivating, in this situation, for about three months, when, early in 

 September, it again showed signs of activity, and fed on through the 

 month evidently with the intention of pupating, but it ceased again 

 in October, hybernated through a second winter, till the end of 

 February, when an attempt to force it in a warm room killed it. 



As already noted, the only European species of which we have actual 

 evidence of its ability to hybernate in two distinct stages is Pararge egeria. 

 There can be no doubtof this most unusual phenomenon. Merrifield notes 

 (Ent. Rec, viii., p. 168) that, in August, 1892, New Forest eggs of this 

 species hatched, producing pupae between October 28th and November 

 27th, which were kept out-of-doors, and emerged in April, 1893. Carpen- 

 ter observes (op. cit., p. 169) that eggs, laid at the end of August, 1895, 

 produced larvae, some of which fed up rapidly and pupated in November, 

 emerging in March and April, 1896, the others fed up slowly, hyber- 

 nated from October till mid-February, after which they fed up rapidly, 

 pupated as usual, and emerged a little later than that part of the brood 

 that had been pupae all the winter. Williams notes (op. cit., pp. 181-2) 

 that eggs obtained July, 1892, hatched August 7th, that one larva out- 

 stripped its relatives, pupated September 1st, and imago emerged Septem- 

 ber 19th ; the remaining larvae (about 30) pupated September 20th-27th, 

 emerged in a warm room between November 19th to December 31st ; 

 but 8 pupae placed out-of-doors November 20th, remained exposed all 

 the winter, and produced 8 imagines April lst-9th, 1893. Our 

 experience has been for the larvae to hybernate rather small from 

 November to February, and then feed up rapidly. Even as far south 

 as Grasse, in April, 1898, Chapman found fullfed larvae of this species, 

 evidently having hybernated in this stage. 



One cannot, however, pass over this phase of the subject with- 

 out noting, in some detail, Scudder's account of Brenthis bellona, 

 of which he observes (Butts. New England, p. 615) that ima- 

 gines appear in May-June, again in July-August, the larvae 

 resulting from this second-brood producing a partial third-brood 

 of imagines in September, a considerable proportion, however, 

 commencing to hybernate after the second or third moult, whilst 

 the eggs from the third brood produce larvae in about 8 days, 

 . . . . and these hybernate at once after leaving the egg. There 

 appears to be no evidence that this is a fact, the supposition that it is so 

 resting on the observation that the spring butterflies emerge over a long 



