110 BRITISH BUTTERFLIES. 



and there lay without web or attachment, whilst others lay hidden at the 

 bottom of the flower-pot also without fastening of any kind. The larvae 

 were very hardy, even in the quiescent period preceding pupation, when 

 shrivelled and almost ready to change, for one or two were accidentally 

 dropped on the floor in removing their food, but they cast their skins 

 as though nothing had happened, and became as perfect pupa3 as the 

 rest. By July 10th, three dozen larvae had assumed the larval state, but 

 not one imago appeared until the following spring. He adds that not one 

 pupa was attached or suspended in any way, either by the anal extrem- 

 ity or by a silken band round the middle, and suggests that, probably, 

 in nature, its habit is to lie near the ground among the thickest grass 

 and herbage. Buckler says that the larva enters the earth, but only 

 just beneath the surface, to undergo pupation ; the appearance of the 

 pupa would assimilate very well with pellets of earth. Prideaux 

 observes that, when fullfed, the larvae pupate upon the earth provided 

 (but never beneath it), others on the sides or corners of the cage, 

 spinning a few threads in no definite direction, but just enough to hold 

 the pupa in the spot selected, from which, however, it is easily dislodged. 

 Bossier says that the larva pupates on the ground beneath fallen leaves 

 and moss, and that the pupa hybernates. Head says that, near Scar- 

 borough, pupation takes place usually at the roots of bilberry, 

 but often also in a dead curled-up leaf. Greene records that a pupa 

 of this species was once found under moss on a log of wood in Bucks 

 (Zool., 1856, p. 5384). Chapman notes of some larvae reared at 

 Hyeres, in 1906, that "the most forward larva was fullfed and stopped 

 feeding on May 21st, and commenced to grow smaller and darker. 

 Being supplied with some loose earth, it went down into it almost at 

 once, and in the morning could not be seen ; the others, in turn, went 

 down, the last on May 30th ; all of them went beneath the soil to pupate, 

 and the pupa has no attachment to anything, except the very slightest to 

 the cast skin." He adds that "on either side of the anal area are about 

 fourteen pale brown cremastral hooks, with a double anchor-shaped 

 hooked extremity ; they are quite dissociated from true hairs, of 

 which several are mixed up with these cremastral ones. The pupa 

 has, therefore, a definite cremastral structure, though I have not found 

 it making any attachment to anything." Lambillion observes that it is 

 difficult to find the pupa, which is usually attached very low down to a 

 rootlet of the plant on which it has fed, or to some object placed on the 

 soil, sometimes lying freely on the earth, and that, owing to its form and 

 colour, it resembles very much a small piece of earth. The pupal state 

 lasts from July and August till the following May or June. Barrett 

 notes that, of a considerable number he reared in 1868 (see supra), all 

 the pupae passed the winter and emerged the next spring except one, 

 and this passed another winter and emerged the second spring. 

 Gauckler states (lllus, Zeits. fur Ent., 1868, p. 183) that the pupae of 

 this species can be forced in spring, but respond better to increased 

 temperature (14°R.-20°R.) applied shortly after pupation, than if first 

 exposed to severe frost and then submitted to the above temperature. 



Pupa. — The pupa is very deep brown, almost black, and has a 

 coating of brown hairs everywhere, except on the appendages. The 

 hairs are about 0"4mm. long dorsally, and about 0*2mm. ventrally. 

 In form it exaggerates the usual Lycacnid outline, with bulbous 

 abdomen and narrower front portion. It is nearly globular posteriorly 



