46 BRITISH BUTTERFLIES. 



winter as a larva, yet it can, and sometimes does, hybernate in the pupal 

 stage (see antea, p. 9). 



Within this general statement concerning the winter-life as larva, 

 there is much variation. Some species winter directly from the egg 

 without feeding, others are nearly fullfed before wintering, and, in this 

 respect, there may be very considerable difference between closely 

 allied species, a difference, one suspects, that is sometimes more or 

 less distantly connected with a final result of single- or double- 

 broodedness, e.g., the larvae of Coenonympha mathewi hybernates directly 

 from the egg (teste Chapman), but the allied Coenonympha pamphilus 

 in the third stadium at considerable size, so far as it does so at all, 

 although it really never becomes absolutely torpid (teste Eussell, Knt. 

 Rec, viii., p. 107) ; the former is single-, the latter, in the south of 

 Europe probably entirely double-, or even partially triple-, brooded. 

 C. iphioides, however, eats before winter, and is single-brooded 

 (Chapman), so also does C. tiphon (Hudson). 



Scudder observes (Butts. New England, i., p. 169) that " the young 

 larvae of Cercyonis alope, like the larvae of the nearly allied E uropean species, 

 Minois phaedra (Satyrus dryas) and GEneis aello, and many other Satyrids, 

 hybernate without having eaten a morsel of vegetable food. He also 

 states (op. cit., p. 173) that the eggs of Cercyonis nephele are dropped 

 loosely on the ground, the eggs hatch in a month's time, and, as with 

 C. alope, the larvae go at once into hybernation. Of this latter species 

 Edwards says that the eggs hatch in August, that the young larvae 

 become lethargic almost directly after leaving the egg, descending to 

 the base of the grass on which the egg is laid, and so hybernate, 

 commencing to feed towards the end of February. 



That this difference in habit is not always connected with the number 

 of broods in allied species is certain, for, whilst the larva of Erebiaaethiops 

 hybernates when only 2mm. long, and possibly without undergoing any 

 moult, that of the somewhat distantly allied Melampias epiphron appears 

 to hybernate in its third stadium when 8mm. in length. One suspects 

 that these two species are not so closely allied as Coenonympha mathewi 

 and C. pamphilus, and Melampias epiphron exists at such high altitudes 

 compared with Erebia aethiops, that one may safely assume that a 

 considerable growth is necessary before hybernation, otherwise the 

 short spring in the localities at the altitudes at which this species lives, 

 would be insufficient for it to feed up and emerge at its normal time. 



It would appear that, among others, the species that live at high 

 elevations and high altitudes may do considerable feeding before 

 hybernation, or otherwise, hybernate twice as larvae. Thus, Edwards 

 observes that (Eneis ivallda feeds on for some considerable time before 

 hybernation ; (E. bore is said to hybernate twice as larva, first quite 

 small in the second stadium, secondly when quite fullfed, (E. jutta, 

 according to Holmgren, also hybernates twice in Europe, first in 

 its third stadium, and secondly, when fullfed ; but Fyles (Can. 

 hint., xx., pp. 131-3) states that, in Canada, it only hybernates 

 once, viz., when practically fullfed, scarcely feeding at all in the 

 spring. One suspects that, in all these species, restricted to extremely 

 high altitudes and latitudes, there is some considerable range of 

 variation, and it is just possible that there may be two hybernating 

 points in the larval constitution, at one, or both, of which, rest may 

 come. 



Taking our more common Pahearctic species which divide up 



