12 BRITISH BUTTERFLIES. 



in the evening to feed. The larva of Augiades sylvanus, as soon as 

 hatched, takes up a position in the middle of a young grass-blade, 

 fastens the edges across with white silk, spinning together, as it gets 

 bigger, the edges of a larger blade, making an opaque web, not much 

 larger than itself for a hiding-place, and, towards the end of October, 

 converts it into a long, close-fitting, tough, silken hybernaculum ; 

 after hybernation the larva spins larger and longer tubes, and hides 

 therein until full-fed. The larva of Urbicola comma, when it leaves 

 the egg, in late March or early April, spins the fine grass together 

 into a somewhat dense cluster an inch or two above the ground, 

 living in this shelter and feeding upon the grass surrounding it ; in 

 this it remains, almost always completely hidden, passing its whole 

 larval life thus concealed, crawling rapidly either backwards or for- 

 wards, putting its head only out of the tube to feed on the surround- 

 ing blades, feeding with great speed, and retreating rapidly when 

 disturbed ; it appears to be particularly nocturnal in its feeding-habits. 

 The young larvae of Cyclopides palaemon, from the very first, makes a 

 little tubular dwelling by drawing together the edges of a grass-blade 

 with silk, but, as the larva gets older, the tubes become less perfect, 

 and, finally, the larva is content to lie exposed on the underside of a 

 leaf. Fletcher notes (Can. Ent., xxi., p. 113) that the larva of C 

 mandan, the American form of 0. palaemon, has precisely similar 

 habits, and, at no time does it construct a tent by catching several 

 leaves together, as do the larva? of Urbicola mystic, U. cernes, and U. 

 manitoba (the last named a Nearctic form of U. comma) ; he says that, 

 in Canada, C. mandan prefers the broad-leaved species of grass, a 

 probable instinctive preference for protection, both from the body- 

 shape and coloration, the habit of the larva of lying extended down 

 the midrib with the body closely appressed, the lower part of the head 

 protruded, and the apex drawn back, making the larva well hidden 

 from observation. 



Of other Nearctic species, whose larvae make their hiding-places in 

 or of grass-blades, Scudder gives (Butts. New England, ii., pp. 1561 

 e.t seq.) many interesting observations. He observes that the cater- 

 pillar of Ancyloxvpha numitor, when first hatched, constructs a nest 

 for itself on either side of a blade of grass by fastening together the 

 opposite edges with from five to twelve strong bands of silk, the 

 threads of each band crossing one another, thus making the bands 

 broader at their origin than in the middle ; after the first moult the 

 tube in which the larva lives is perfected by the interstices being 

 closed with a thin irregular web of silk. The young larva of Ambly- 

 scirtes vialis draws about half-way together the opposite sides of a 

 blade of Poa pratensis near the top, with one strong strand of silk, 

 thus making a sort of tent ; later it fastens together the opposite edges 

 of a blade of grass by about a dozen strands of silk, tightening these 

 until a sort of closed tube is formed ; later in life, and especially 

 when about to moult and desiring greater concealment, it constructs 

 a nest by doubling a leaf back upon itself and sewing the sides 

 together, or by stitching together three or four contiguous blades into 

 a tube, lining the floor in either case with a dense coating of silk ; 

 when more than half-grown it not infrequently comes out of its nest, 

 and, after the iiocculent covering, which distinguishes the larva in its 

 lasi stages, has fairly developed, it will rest stretched out at full 





