138 BEITISH BUTTERFLIES. 



the eggshell, sometimes all except the base. The egg stage as deter- 

 mined by Sich (in Lausanne and England) lasted eight days, but 

 Hellins notes it as thirteen days, so that there may be some difference 

 in this respect. The young larvae will feed on various grasses in 

 confinement, Poa annua, etc. Hellins observes that, as soon as some 

 larva? he had in confinement had devoured the eggshell, they took up 

 a position of rest in the middle of a blade of cock's- foot grass, fastening 

 its edges across with five or six distinct little ropes of white silk. 

 They soon, however, spun together the edges of the grass blades, and 

 made an opaque web, not much bigger than themselves, for hiding- 

 places, and, at the end of October, spun long, close-fitting, narrow, 

 tough, silken hibernacula (Hellins). The hybernating tube is very 

 like, if not identical with, the feeding-tube, differing in having a strong 

 grass stem as one of its elements, which is rarely the case when the 

 larva is younger, although one hybernating individual now has only a 

 very weak one ; it also differs in that the leaves are now nearly all 

 brown and dead, although there are still green blades on the plant. 

 The larva gets somewhat shorter, thicker, and more sluggish after a 

 few weeks and is of a more transparent vivid green (having 

 apparently no material in the alimentary canal) than when feeding 

 (Chapman. October 12th, 1905). In early spring the larvae recommence 

 feeding. In confinement, if disturbed, the larva coils in a close ring, 

 •the head against the anal prolegs, the anal flap extending over both 

 as a shield. Its movements are extremely sluggish, and it crawls exceed- 

 ingly slowly from one point to another (Tutt). The larvae are fullfed 

 in May, each one spinning then a silken lining in a cylinder previously 

 formed by uniting the edges of a leaf of grass or Luzula, and in this it 

 changes to a pupa. When fullfed the larvae move somewhat slowly and 

 appear to be much more lethargic than those of U. comma (Hellins). 



Larva. — First instar (newly-hatched): 3"5mm. long, yellow-grey; 

 ihead large, black, polished and smooth, with a few short scattered 

 hairs. Prothorax with a black plate and another small square one between 

 it and spiracle (which is larger) ; this extra plate has a convex boss 

 (lenticle?) and one hair ; on the plate are five hairs on each side ; in front 

 of the spiracle are two hairs. On the 2nd and 3rd thoracic segments are 

 .two convex brown plates (lenticles) in front of hi. The abdominal subseg- 

 mentation is into seven, in size 2*5:1:1:1:1:1:1. The 2nd and 3rd 

 thoracics have five hairs almost exactly in a transverse line, and 

 .apparently representing tubercles i, ii, iii, iv, vii. The abdominal 

 hairs are i on 1st subsegment, ii on 1th, and iii on a slight elevation 

 above spiracle and just below where subsegmental incisions end. Of 

 iv and v, the anterior (v) is below spiracle and slightly in front, the 

 posterior (iv) well behind spiracle and at an elevation half-way between 

 spiracle and the front hair; one lower than this is not clearly 

 determined; on the 8th abdominal the spiracle is very large and 

 well above (and behind) iii; hairs representing iv, v and vii (?) are 

 distinct. On the 10th abdominal are four very long hairs O'Smm., 

 and four shorter Olmm., as compared with 0*05mm., the average 

 length of the other hairs of the ordinary tubercles. The prolegs have 

 a complete circle of about 35 equal booklets. There is an anal comb 

 of seven or nine tines. The claspers want the outer posterior fourth 

 (Chapman, July 9th, 1905). First instar (two days old) : Head black, 

 pitted ; thoracic plate black. The tubercles simple, each bearing one 

 clubbed seta. No secondary seta 1 or hairs. The meso- and meta- 



