1890.1 103 



and had good reason for believing that they had moulted twice before 

 hibernation, which would make the number of moults four, and hibern- 

 ation to come in half-way. But of a small collection wintering at the 

 present moment, some correspond with the above, and all in their third 

 skin, whilst the rest appear to be a moult behind them. 



The larva has the usual shape, with the opaque skin, and the rich 

 and varied colours characteristic of so many of the genus. 



When very young, the back and sides are yellow-brown, the spiracular region 

 and under parts whitish-green ; the dorsal line (a true skin mark) is a narrow thread 

 rather darker than the ground-colour, and present only in the abdominal segments ; 

 sub-dorsal lines broad, and with ill-defined edges, dark yellow-brown ; head greyish- 

 brown, almost black at the sides, and armed with long pale hairs. By the time it is 

 ready to hibernate, and when, as I estimate, it is in its third skin, the colours have 

 become darker and warmer, and the pattern more complex. The dorsal area of the 

 thoracic segments is now much irrorated with white on the fore part of the segments ; 

 the corresponding change on the abdominal segments is less complete, and is limited 

 as yet to a pair of whitish spots at the divisions, followed in some cases for a short 

 distance by a pale irroration ; the white spiracular line also makes its appearance, 

 and is a broad band on the thoracic segments, but on the abdominal ones it is as yet 

 narrow, wavy, and interrupted in the middle of the segments ; the under parts are 

 mottled with brown. Sometimes, as I have already indicated, the changes are less 

 advanced — the white spiracular line is present, but there is little trace of the white 

 dorsal irrorations, and the under parts are without the darker mottling : the speci- 

 mens, too, are considerably smaller, and evidently a moult behind the others. After 

 hibernation, further changes take place, until with the last moult the full and very 

 beautiful pattern is completed. The head of the full-grown larva is ochreous- 

 brown, streaked at the sides with blackish-brown ; the back in both thoracic and 

 abdominal region is alternately white and orange, these colours occupying respec- 

 tively the fore and hinder parts of the segments ; the brilliantly white spiracular 

 band extends from 2 to 12 inclusive, on its upper border is a large orange-spot in the 

 middle of each abdominal segment (5 — 12) ; the narrow but conspicuous rich brown 

 dorsal line is interrupted on the thoracic segments, but continuous on the abdominal 

 ones ; the broad sub-dorsal lines are also rich brown, with their edges still ill-defined ; 

 just inside them, and almost in a straight line, lie the small white dorsal spots; the 

 under parts are dark brown. 



For this description of the mature larva I depended upon larvae 

 collected early in September, and kept under observation till they 

 went into pupa at the end of May. A full-grown specimen, taken at 

 large on May 22nd, 1888, was so different, that it was described as 

 a distinct variety, but, as it afterwards proved to be stung, it would, 

 perhaps, be more correct to consider it an abnormality ; yet none the 

 less interesting, inasmuch as the peculiarity mainly took the form of 

 an arrest of development, the pattern being little, if at all, in advance 

 of that of the larva at the hibernating age. 



