42 
the threads, cannot, from their conformation, slip back again, and so 
are held fast. The movement is repeated, and the hooks on the oppo¬ 
site side of the dorsum advance in like manner, till the crook on the 
headpiece of the pupa is pressed so hard against the roof of the cocoon 
that it snaps off, leaving a sharp, straight point on the pupal head. 
Perhaps the jerk of breaking causes the point to pierce the cocoon. In 
any case, the advance of the abdomen thrusts the point through, and 
the serrated edges of the head armature, aided by the lateral move¬ 
ments of the pupa, soon cut a clean transverse slit in the roof of the 
cocoon. The abdomen continues its forward movement until more 
than half the pupa is thrust out of the cocoon. By this time the 
struggles of the moth rend the pupal skin, which splits transversely 
across the thorax. On dehiscence the headpiece, clypeus, and mouth 
parts come off in one piece, but are held to the rest of the pupa-case 
principally by the antennas, which separate out partly, but are attached 
at their base and apex. The limbs are then withdrawn from their 
sheaths, and the moth, finally escaping from the pupal shell, seeks a 
convenient point, to which it clings, while the wings rapidly expand, 
and the living gem assumes all the glories appertaining to the species. 
The Mine. — In sketching the above rough life-history I have 
partly described the normal mine. The cuticle, severed from the leaf 
by the larval jaws, remains loose while the larva is beneath, but as it 
advances it ejects the digested fluid from the anus with such force 
that it spreads over or floods the space immediately behind. This 
causes the leaf cuticle to again adhere to the parenchyma, and the 
wave-marks of each ejectment may often be traced in an old mine. In 
perfectly normal mines no central excremental line appears. In some 
cases such a line does exist in a portion of the mine. It may arise 
from an ill-grown leaf, or from a defect in the larva’s jaws. In one 
very marked case I found the point of the right mandible was slightly 
deflected, and morsels of the more solid portion of the leaf were cut 
off and thrown into the mine. The larva appeared to have no choice 
but to swallow them with the juice, and as long as they were very 
small the pieces passed through without, apparently, any harm being 
done. But one morning this larva seemed to be in trouble. It lay 
still in the mine, and I saw quite a large black mass, microscopically 
speaking, lying in the cloaca, which was much distended. However, 
the next day it succeeded in passing the obstruction, and again began 
to feed. Two days after it was still feeding, but the next day it was 
almost dead, and it never recovered. The larva has the power of 
passing from one leaf to another by extending the mine down the 
petiole, along the twig under the skin of the bark, and up another 
petiole into a leaf. I must admit I have only observed this in captive 
examples, and then only in three instances. However, we should not, 
I think, be surprised to find that a method habitually adopted by one 
species should be occasionally resorted to by another closely allied 
species. I think Pln/llocnixtis salu/na normally passes from one leaf to 
another, a single leaf not sufficing for its full nourishment; but on 
this species I have only made very few observations. 1 first noticed 
this occur in Phi/llocmstis sufl'usdla, by observing that two larva) from 
mines in small leaves which 1 had marked were missing. As they 
are, however, often difficult to detect, I supposed I had overlooked 
them. These mines, nevertheless, did not progress, and yet there was 
