142 Dr. T, A. Chapman^s notes on Pnpw, 



it. The anal prolegs have nine hooks disposed in a 

 curved line that is less than a semicircle, the hooks are 

 larg^er and stronger than the ventral ones. The true legs 

 appear to have four joints before the claw, which is 

 accompanied by a transparent, battledore palpus ; the 

 antenna is three-jointed, the second joint carrying a long 

 hair and a fleshy tubercle, the third carries two terminal 

 bodies. 



These larvfB were full-grown by the middle of May ; 

 they lived on the upper side of the bramble leaf, spinning 

 no silk except a very slight web. When about to moult, 

 and when travelling from leaf to leaf, they are very 

 sluo-gish in their movements, yet fond of moving occa- 

 sionally to a fresh leaf; when at rest the head is rather 

 hidden by being withdrawn into the second segment. 

 They are of a glassy transparency, and match the colour 

 of the bramble leaf very closely, so as to be difficult to 

 see. In moulting, the slit appears to be dorsal, and 

 the empty skin remains stretched out at length, with the 

 head bent under the first segment. The cast skin makes 

 a good preparation of the hairs. 



The cocoon is a very remarkable structure, and I 

 regret not having seen how it was made. It is attached 

 by one margin, naturally probably zo the bramble stem 

 near the root, aiid consists of a network of stiff' gum- 

 1ike strands of silk, one set of these being in continuous 

 lines, like the mortar between consecutive courses of 

 brickwork, the other uniting these alternately like the 

 vertical lines of mortar between the bricks in each course, 

 the bricks being represented by the openings, which are, 

 however, square rather than oblong ; the primary strands 

 are sometimes, as it were, in one set, from the four sets 

 in which they appear to be constructed being accurately 

 worked together ; in other cases the four sets are more 

 or less distinct, they would be separated by slicing the 

 cocoon longitudinally and transversely down on to its 

 attached margin. When each side is accurately con- 

 structed the courses arch upwards from either end of 

 the line of attachment meeting in the line of transverse 

 division suggested. There is a valvular longitudinal 

 opening at either end, through one of which the larval 

 skin is ejected, through the other the pupa emerges for 

 the exclusion of the moth. 



The pupa lies free in this cocoon with little spare 



