CYANIR1S SEMIARGUS. 285 



of writing, larva number one has eaten a good bit of the outer part of 

 the leaf, i.e., nearly half of the outer half of the outer leaflet ; the 

 leaflets are beginning to slightly separate, and the larva has bored a 

 hole through the inner half of this leaflet, and is half-way through it, 

 and so reaches the margin of the outer leaflet (missing the middle one) 

 of the other side, which it is eating. The general colour is a light 

 green, with some scattered hairs (no black points), a longitudinal stripe 

 of yellower-green, with paler borders just below the darker dorsal 

 line, and then a darker, slightly echeloned, stripe followed again 

 by a paler (yellower) band. Then the marginal flange with more 

 abundant hairs on a paler margin, and slightly darker above. One of 

 the larvae left its leaf on April 13th, and rested for the night on the 

 stem, but returned to its leaf the next morning ; it is now 6mm. long, 

 of a green just matching the clover-leaf, and is of a very transparent 

 glassy appearance, due essentially to the larval coloration. By April 

 20th, one larva had fed so far ahead of the other, that, on this day, both 

 moulted, one into the final, the other into the penultimate, instar. 

 The more backward one is only 4-4mm. long, and it moulted between 

 two leaflets, so fastened together as to form a little pocket or cocoon. 

 The skin cast by the larva remains distended, is, in fact, a little, 

 balloon-like, blown larva skin, 4mm. long and 2mm. across. The 

 larva escaped by an opening, extending from the head in front down 

 each side to the middle of the mesothorax on one side, to its posterior 

 border on the other; not quite down the side, however, tut more 

 dorsally, so that the dorsal flap is triangular, the prothoracic spiracle 

 being on the ventral portion, to which, also, the head is attached. 

 The larvae now seem to prefer the young tissues encased in the 

 sheathing stipules, burrowing into these, and even getting quite out of 

 sight ; they seem to like the growing stem better than the young 

 leaves. When nearly fullgrown, one of the larvae was noted to eat a 

 leaf still unfolded, but quite outside the sheaths of the bud. As it 

 gets older the larva eats less, is very lethargic, and acquires a yellow 

 tinge before pupation (Chapman). Frohawk says that the young larva 

 makes its exit from the egg, by eating a small hole in its side, just large 

 enough to allow of its escape ; the newly-hatched larvae ate through the 

 bases of the calyces of the flowers of Anihyllis culneraria, and commenced 

 feeding on the green seed-pods within ; they continue to feed on the seed- 

 pods during the second and third stages, but they do not refuse the yellow 

 petals of the flowers, and feed, indeed, on all parts of the bloom ; those 

 that feed on the petals are distinctly yellower in colour. Besides the 

 flowers of AntJujllis vulneraria, the larvae feed freely on white and pink 

 clover-blossoms, preferring pink, however, greedily devouring them 

 and eating all parts of the blossom. In the early part of August, the 

 larvae prepare for hybernation, and some kept by Frohawk concealed 

 themselves within the calyces of Anthyllis during the second week of 

 August, others hid under the leaf-like bracts, and yet others between 

 the calyces, whilst others were already at rest (hybernating) on clover- 

 flowers. Although these larvae w x ere kept in a warm temperature, fully 

 exposed to the early morning sunshine, they remained quite motionless 

 during the hot days of August and September. Daring the first week of 

 September, a few of these larvae, hybernating on flower-heads of clover 

 and Anthyllis, were placed out of doors, the pots containing the plants 



