434 BBITISH BUTTERFLIES. 



not pupated by the 14th (Mathew) ; fullfed larva? were found on 

 Rhamnus in September, 1904, near Berlin (Dadd) ; whilst on August 

 25th, 1906, larvae were taken on flower-buds of iv} r , at Mucking 

 (Burrows); larvae found September 11th, 1907, at St. Cloud, pupated 

 September 17th, one imago emerged October 15th, the rest of the pupae 

 overwintered (Oldaker). Hiittner's statement that larvae were found on 

 birch at Carlsbad needs confirmation. [For further notes on "Habits 

 of Larva," see antea, p. 408.] 



Ontogeny of larva. — Chapman writes {in litt.) : My notes show 

 that C. argiolus has only three moults or four' instars,' and I believe 

 this conclusion is correct. My doubts arise from the circum- 

 stance that some of my earlier notes reflect the belief that there 

 should be four moults, and so give rise to some confusion, without 

 affording any evidence that it is so. Again, Edwards, in his account 

 of C. pseud argiolus (Butts. Nth. Amer., vol. ii., and quoted antea p. 409) 

 gives four moults ; but reading his account of the several instars, one 

 sees that, careful observer as Edwards is, he gives no definite account 

 or dates for four moults, and seems rather to have assumed five 

 instars, and selected probable larvae to illustrate them. On the other 

 hand, it is quite possible that C. argiolus (and C. pseud argiolus) some- 

 times has three, and sometimes four, moults, possibly in spring and 

 summer broods. My preserved specimens, not, however, too numerous, 

 fall readily into four instars, without leaving any gap for an over- 

 looked fifth one, and this is so whether measurements be taken of 

 head, clypeus, jaws, true legs, or the hooks of the prolegs ; the 

 former enlarging at each moult by 50 per cent., the latter by 100 

 per cent, (linear)." Buckler also gives four moults, that is, five 

 instars, but he describes only four, which leaves it possible that he, 

 too, was not clear upon the point. The following are his short des- 

 criptions of the appearance in each instar : — 



First instar : — Plump, hairy (something like that of an Anthrocerid larva in 

 shape), body greenish-white, head dark, moderately large. 



Second instar : — Stouter in figure, pale ochreous-green in colour, clothed with 

 long, whitish, soft, silky hairs ; when about a fortnight old, 4-75mm. long, of the 

 usual Lycsenid shape, skin smooth and glistening. 



Third instar : — 8mm. long, stout in proportion, showing a pale streak on the 

 ridges of the back, thin double-slanting lines on the sides, and a margin of 

 yellowish-white along the subspiracular region. 



Final instar : — 9*5mm. long, rather more when crawling. Variable in colour, 

 with small, purplish-black, shiny head placed ventrally, and retractile in prothorax. 

 The ventral surface flattened, the mesothorax swollen, the segments with deeply- 

 cut segmental incisions ; a slightly-developed double dorsal ridge, and a faintly- 

 marked mediodorsal line. The mesothorax very broad and also the longest segment, 

 is only slightly convex above, the others are arched on the back, the mesothoracic, 

 metathoracic, and 1st abdominal segments the highest, whence they slope a very 

 little to the Gth abdominal. These segments carry the double dorsal ridge of 

 humps, between which lies a slightly sunken dorsal space broad and hollow on 

 the meso- and metathorax, whence it flattens and narrows gradually to the 6th 

 abdominal. The humped appearance of the ridges is due to the deeply-cleft segmental 

 incisions, those of the lst-3rd abdominals being bent forward laterally. The 

 segments from the 7th abdominal to anus simply convex, sloping towards anal 

 end, and the sides, although sloping outwards, become almost concave near the 

 projecting, rounded, subspiracular ridge, which continues round the anal segment, 

 overlapping all the short prolegs. The whole of the segments pitted with a vast 

 number of exceedingly minute depressions, from every one of which arises a tiny 

 hair. The spiracles excessively minute, whitish, with a well-developed marginal 

 ridge. On either side of the median line in each segment is an oblique dash 

 passing across the segment downwards from front to back, and rather paler than 

 the ground-colour, whilst just above the spiracles is another straighter line. The 



