THE CANADIAN ENTOMOLOGIST. 157 



after its first moult ; then leaves it, and feeds externally for about two 

 days, usually on the under side of the leaf, but occasionally also on the 

 upper side. There it spins beside a rib a thin sheet of white silk, beneath 

 which it spins a cocoonet, in which it again assumes the horse-shoe shape, 

 and passes in about a day to second moult. Emerging from its cocoonet, 

 it continues to feed externally for three days, when either on the plant or 

 near to it, it spins its ribbed cocoon, in which it passes the pupa state. I 

 have not observed accurately the length of this stage ; in August it is 

 about a week. The mature larva is about three lines long. 



I have frequently been puzzled to understand how the larva could spin 

 this singular cocoon, but I have now fortunately been enabled to watch it 

 at work under the microscope. The cocoon shows six longitudinal ribs or 

 ridges, with depressions like valleys between them. Each rib consists of 

 four threads, and is four times as thick as the depressions ; the threads of 

 the ribs are longitudinal and rigid, those of the valleys run obliquely 

 transverse, and each is permitted to droop or sag down, and they are spun 



first from right to left, then from left 

 to right, crossing each other at a some- 

 what acute angle, the one set being 

 kept always about four threads in 

 advance of the other, the finished por- 

 tion of the cocoon showing the two 

 threads crossing each other, while the 

 unfinished shows only two threads 

 without any thread crossing them, as 

 shown in fig. i at a finished, at If 

 unfinished, portion of the cocoon. 

 But properly speaking, this is no part of the cocoon, but only a reticulated 

 frame or net-work, within and attached to which the true cocoon is spun. 

 The whole net-work is a continuous thread, with no break ; each trans- 

 verse thread continues entirely across the cocoon, but the ribs are not 

 continuous threads the length of the cocoon ; each rib is made by a mul- 

 titudinous succession of movements forward and back again, each move- 

 ment only the length of the space between two transverse threads. 

 Whenever in the transverse movement of the head, the apex of the 

 spinneret touches a rib, it is moved forward and back again. Thus, the 

 larva (having laid the floor or foundation of its reticulated frame-work by 

 spinning its web somewhat densely over the portion of the leaf that is to 



