574 BRITISH LEPIDOPTERA. 



with a smooth outline, and tapering continuously from base to apex. 

 They have, however, a joint about three-fifths of the length from the 

 base to the apex, looking harder, browner, and more solid, and they 

 seem to have a central tube. They appear to originate from deep 

 tissues, and the skin looks like a thick coating of glass, through which 

 they come. In this skin, however, the dorsal set of spines is now 

 double, i.e. , the double row, of which the alternate members are wanting 

 in the first larval stage, is now complete, and they remain so even 

 in the adult larva, though they are now merely prominences, and not 

 spines. In the second and third skins, there are, especially in the 

 thoracic region, round the bases of these spines, very minute spinules, 

 apparently of a structure very similar to the last joint of the spinules in 

 the first stage. [Besides this excellent description, Chapman gives 

 first-class figures of the newly-hatched larva, spines, etc., in -the Trans. 

 Ent. Soc. London, 1894, PI. vii., figs. 14-20.] 



Larva just previous to pupation. — When coiled up in its cocoon, 

 the larva is a curious object. The dorsum is green, with no darker 

 markings whatever ; there is a distinct dorsal depression, whilst the 

 sides appear now as subdorsal depressions, owing to the bulging out 

 of the ventral area. The lateral margins and anal area orange-yellow. 

 The thoracic segments not grooved dorsally or laterally. The skin, 

 under a two-thirds lens, now appears finely warted and much wrinkled 

 transversely. The eight abdominal spiracles are very tiny, pinkish 

 in colour, with a pale yellowish-white rim, the pair on the 6th 

 abdominal segment much larger than the others. The conspicuous 

 external structures surrounding them, and so prominent in the 

 adult, larvae, have entirely disappeared, leaving the spiracles them- 

 selves distinctly visible. The prothorax is only visible ventrally, and 

 the head is withdrawn into it, showing the almost transparent face in 

 its centre ; whilst the prothoracic spiracles show conspicuously on each 

 side. The venter is very transparent, its surface covered with an 

 adhesive substance, and the internal structures visible within. 



Cocoon. — The cocoon consists of a thin outer pellicle of flossy silk, 

 which is interwoven with the surrounding moss (in the examples 

 under observation). Within this is a closely woven, oval structure, 

 reddish-brown in colour, of a somewhat papery texture, but tough for 

 its thickness. It is provided with an easily separated lid. The cocoon 

 is covered inside with a delicate silken lining, which may be separated 

 from the part surrounding it, and the larva is so coiled up in it as to 

 occupy most of the available space. Fletcher observes that the cocoon 

 becomes flattened on each side when attached to others on a leaf, and 

 when spun upon an oak-leaf, fits so closely to it, as to take the impres- 

 sion of the smallest veins. The larva lies unchanged in the cocoon 

 all the winter, becoming a pupa in May or June. The pupal stage 

 lasts only about fifteen days, and the pupa escapes from the cocoon 

 before the emergence of the imago. Borkhausen says that he found 

 the cocoons of the ab. Umax under the fallen leaves in a beech wood. 



Pupa. — The pupa is described by Fletcher as being short, thick and 

 stumpy ; broadest about the middle of the abdomen ; thorax large, 

 rounded, extended in front above the head ; wing-cases long, well- 

 marked, with the neuration plainly traceable ; leg-cases almost free, 

 those of the third pair extending beyond the wing-cases to the anal 

 extremity ; creamy-white in colour, with head and thorax tinged with 



