State Agricultural Society. 
551 
ing tlie crown of the head and top of the neck inward, face to face 
with the body, whilst the mouth, the breast and anterior legs face 
outward. It here fastens the other end of the thread, applying its 
mouth to the surface at several different points to render the attach¬ 
ment secure, a minute or more being occupied in turning from one 
side to the other and fastening the thread. It then turns back again, 
spinning a second thread in contact with the first, and fastening its 
end at the same point where the first thread was commenced. 
It does not spin this thread from the forward extremity of its head. 
Its mouth is turned downward toward its breast, so far that the 
threads which have been spun pass across the throat and shoulders of 
the worm when it is turning over, the neck as well as the middle of 
the body being inclosed in the girt, whereby both the head and the 
posterior half of the worm are upon the hind side of the girt, and 
the remainder of the body, doubled into a half circle, is upon the 
fore side. 
In this manner the worm carries its mouth from one side of the 
body to the other, each time adding another thread to the girt, until 
it has formed a skein of threads of sufficient strength to hold its body 
securely suspended. A much stronger girt is probably formed to 
hold the pupa through the winter than to sustain it only a week or 
two in the summer. Reaumur supposes there are about fifty threads 
in the girt, he having observed thirty-eight spun, and about a dozen 
had been spun before he began the count. In one instance which I 
observed, the worm placed but twenty-one threads in its girt, and was 
occupied twenty minutes irf spinning and fastening this number. In 
another instance thirty threads were spun, the worm being engaged 
thirty-five minutes in the work. And both these girts were for win¬ 
ter pupae. These facts render it evident that the number of threads 
in the girts is extremely variable. 
When the girt is finished the worm requires to withdraw its head 
and neck from it. This is quite a nice operation ; for if it should 
draw its head directly forward out of the loop, it would incur a risk 
of separating and drawing apart the threads in the bundle, and would 
probably break asunder some of the threads thus separated. To avoid 
any such casualty the worm first retracts its head into its neck as far 
as it can, and then bends it inward, crowding it against and under 
the side of the body, farther and farther, until the girt slips off from 
it. This curious expedient will be better understood by the reader 
from a familiar illustration. If he holds the fore finger of one hand 
straight and hooks the fore finger of the other hand around it, to free 
