DESCBIPTION AND SEASONAL HISTORY. 23 
When the larva is fully grown, it ceases to feed and seeks out some 
place in the crown of the plant among the litter and trash or on the 
ground among similar material, where it spins a cocoon (fig. 6) . 
COCOONING AND PUPATING. 
The cocoon is composed of fine white threads and the construction 
by the apparently blind larva was in part observed by Mr. Ainslie, 
who describes its movements as follows: A larva was seen moving 
about in its snow-white, almost transparent, gauzy, unfinished cocoon. 
It proved to be spinning a closer mesh from within. Instead of 
spinning the silk from a gland that opened into its mouth, as was 
supposed, the fluid from \^hich the silk is made is taken into the mouth 
apparently from a gland in the caudal segment. The larva applied 
its mouth to an opening or gland close to the anus, appeared to move 
its jaws slightly, and then, with a quick 
movement of the body, was straightened 
out as much as possible in its confined space, 
and instantly the head was applied to the 
inner network of the cocoon. A slender 
glistening thread w^as seen leaving the 
mouth, being attached glutinously to each 
thread that it crossed. The larva worked 
rapidl}' and nervously, nearly always car- 
rying its new thread in a rather straight 
line. From 30 to 50 seconds were required 
to discharge a single mouthful supplying 
thread for one-third or one-half a revolution fig. 6.— The aifaifa wee\ii: cocoon. 
inside the cocoon. ^^^^^ enlarged. (Author's illus- 
When all the supply was exhausted, the 
head groped aimlessly about for a few seconds, then was applied to the 
caudal gland as before. The body would then straighten with a quick 
movement and almost instantly the thread would be again flowing as 
before. The new thread was guided skillfully across the meshes, rarefy 
if ever follomng the line of a thread already laid. A very slight jar 
would cause a sudden halt for perhaps half a minute, then the opera- 
tion would hesitatingly proceed. As the irregularly oval cocoon is 
too small in any diameter to allow the larva to straighten out, the 
larva moved about by thrusting its small head into a mesh, swinging 
the body into the desired position; the head would then be moved 
to another mesh and the operation repeated. The fluidity and amount 
of the silk must vary as spinning progresses, the silk becoming more 
viscous or less copious as the cocoon approaches completion. 
The pupal period, according to Mr. Parks's notes, during the middle 
of May lasts about 9 days, the larvae spinning their cocoons about 5 
days before pupating. (A pupa is shown in fig. 7.) At the end 
