258 Dr. T. A. Chapman on a lepidopterous 
1st abdominal segment in this matter declares itself to 
be truly a 4th thoracic segment, presenting similar 
chitinous plates to those on the 3rd thoracic, though 
rather smaller, and with more intermediate membrane, 
and not at all like the simple colourless plate of the 2nd 
and following abdominal segments. 
The oval hoop that carries the jaws, and is so obvious 
in the cast pupa-skin, is not very evident before emer- 
gence ; it appears to have several oblique bars, but its 
actual form is difficult to fully grasp, and without figures 
hopeless to describe to any purpose. 
It may be well to mention that purpurella only spends 
some five or six weeks above ground as imago, ovum, 
and larva; the rest of the year is spent buried in the 
soil, chiefly as a resting larva, and for a few weeks as a 
pupa. To insure its safety, it buries itself deeply, and 
spins a cocoon of remarkable strength for so small an 
insect. The depth is probably usually } in. to 2 in., but 
8 in. to 10 in. are reported by persons who have reared 
them, but probably supplied them with light open soil ; 
this is proportionally as though, say Acherontia atropos, 
were to go down 15 or 20 ft. It is to enable it to escape 
from this hard cocoon and the superincumbent earth 
that the active jaws come into use. 
I ought not to omit to state the comparative size of 
the mandibles, which is great, so as to be comparable 
with, say, those of Lucanus cervus ¢, rather than with 
those of any lepidopterous larva ; the length of the pupa 
is about 4 mm., its greatest width about 1°3, the width 
of the head about 0°9, whilst the length of each man- 
dible is about 0°7 mm. 
When in the cocoon the empty larva-skin is tucked 
away under the extremities of the leg and wing-cases, 
and between them and the abdominal segments. In the 
curved attitude of the pupa in the cocoon there is cer- 
tainly more room for it here than elsewhere, and so it 
probably gets here rather readily by the movements of 
the pupa. 
Micropteryx purpurella emerges about 6 to 7 a.m., and 
on the mornings of March 10th, 11th, and 12th, and on 
several occasions afterwards, I had the pleasure of 
observing some specimens do so. I had sorted out the 
cocoons from the sand in which they had been made, 
and had them lying loose in a jar on the surface of some 
