1200 THE BUTTERFLIES OF NEW ENGLAND. 



at all times feeds to repletion, so that the skin of the hody is tense and 

 glistening when it has finished a meal. It never eats the leaf at the edges, 

 and generally or always leaves the veins untouched, feeding upon the under 

 surface only. 



When about to change to chrysalis the caterpillar seeks the under sur- 

 face of boards or garden fences protected by weeds, and may often be 

 found in considerable numbers in dry and sheltered places. 



Pupation. Fitch, in his Thirteenth report on the insects of New York, 

 gives such a careful account of the change to chrysalis in this insect, that 

 we copy it in full : — 



Having found such a situation a.-? it requires, the larva determines in what position 

 it will place itself to remain during its pupa state, for there is much diversity in our 

 white liutterfly in this respect, it being sometimes suspended ui)ou the under side of a 

 liorizontal or an inclined surface, and at other times against the side of a vertical sur- 

 face; and it is held iu a variety of positions, horizontal, oblique, or almost perpen- 

 dicularly upward or downward, usually ^vitli its back, but sometimes with its right or 

 its left side downward. 



Having selected the spot it will occupy and the position iu which it will suspend 

 itself, it requires a loop to be made around its body to hold it in this position. To 

 give its feet the secure foothold they will require wdiile this loop is constructing, it 

 first spreads a slight carpeting of silk tlireads upon the surface on which it is to stand, 

 forming also at its lower end a thick mass or little liillock of tliese tlireads. It then 

 fastens the hooks of its hind feet securely into this mass of threads, and clinging to 

 the carpeted surface with its middle legs, its body is so very soft and flexible that it is 

 able to bend and turn its head backward, touching either side or the top of the back 

 at a point one-third of the distance from the hind to the fore end. And with its head 

 thus tamed backward it fastens to the surface, at one side of its body, a thread of 

 silk which it spins from its mouth. It then carries Its head up over its back and 

 down upon the opposite side of the body, where the mouth fastens the other end of 

 the tliread, thus forming a loop around its body, holding it to the surface on which it 

 is standing. This single thread, however, is exceeding fine and possessed of but little 

 strength. The worm therefore carries its mouth back again by the same route, to the 

 opposite side, thus spinning a second thread, with its end fastened at the same point 

 ■where the first one was commenced. And it thus continues to move its mouth from 

 one side to the other, until it has formed a .skein of threads of sufficient size and 

 strength to securely sustain the weight of its body. Reaumur states the number of 

 threads in the loop to be about fifty. But In our American species there is no uni- 

 formity in this loop, it being in some instances not a fourth the size that it is in others. 

 If the larva gains some secluded corner wliere it will experience no molestation, it 

 does not troidile itself to spin but a few threads to form this loop. 



The most laborious part of tliis work, and that which occupies the iiriucipal part of 

 the time that this loop is being constructed, is the fastening of the ends of tiie threads 

 to the surface on each side of the worm. As the tension upon the threads is almost 

 directly upward, it has a strong tendency to tear them from their attachment, and the 

 worm shows particular care in fastening the cuds securely, applying its mouth to the 

 surface at numerous points, to glue the thread thereto, whereby a dense web comes to 

 be formed upon the surface around each end of the loop. The worm, moreover, 

 moves its body from side to side with each thread that is spun. When a thread is 

 about being fastened upon the right side, to give the mouth the requisite room for 

 attaching it to the surface, the body is crowded to the left, as far as the threads 

 already spun will admit, thus putting these threads on the stretch, whereby tlie worm 

 will ascertain if it is fastening them sufficiently secure. 



