60 
THE entomologist’s RECORD. 
finally, however, spun, but died without changing. The first moth 
emerged March 2nd ( $ ), the second March 4th ( $ ), and the others ( 
and $ ) a few days later. The treatment by which this fairly successful 
result was attained was quite artificial. The larvae were throughout 
kept indoors in glass jars and fed on Poa annua^ which was changed 
every second day until the larvae became sluggish in December and 
January, when it lasted about a week ; but throughout, the larvae were 
regularly disturbed, and not allowed to rest for hybernation or other- 
wise. When they reached the fourth skin each larva had a separate 
glass, small, less than three inches high and two in diameter, covered 
by a glass plate ; this kept the food fresh, whilst undue moisture was 
prevented by half an inch of clean dry sawdust at the bottom, changed 
with the food, and a sheet of blotting-paper under the glass cover, this 
also dried at each change of food. The larvae usually hid themselves 
in the sawdust during the day, they often made therein a smooth 
cocoon-like cavity, but without using any silk. I find that I rarely fail 
to rear anything to which I pay sufficient attention and individualise 
each larva in this manner. Dampness, stale food, and crowding are 
the great enemies of success in rearing larvae in captivity, and they all 
result from trying to do more than the time and attention available 
justify. The full-grown larva has at first blush a very Agrotid appear- 
ance. It varies from a nearly uniform nankeen-yellow with markings 
only indicated, to a handsome larva with distinct black stripes. There 
is a pale dorsal line, quite narrow ; thence to the spiracles (which are 
black) is divided into three longitudinal stripes, a dark dorsal one, 
a dark (but less dark) lower one, and a pale intermediate one. In all 
these the ground colour is the same nankeen-yellow, and the darker 
areas depend on the greater or less darkness (and abundance ?) of fine 
black mottlings, generally in fine wavy streaks running more or less 
longitudinally. The dorsal dark band is darker on each of itfe margins, 
and each margin is darker towards the incisions ; the pale band is some- 
what darker centrally, whilst the lower band again is darker marginally 
and towards the incisions, but only to a degree that requires looking 
for, instead of being obvious as in the dorsal dark band. The spira- 
cular region is paler, almost amounting to a pale spiracular stripe, and 
the lower surface is nankeen-yellow, with a slight fuscous tendency. 
The second segment is of a more uniform tint, and the pale dorsal line 
extends distinctly through it and on to the head, which is rather brown 
than yellow, mottled in a honeycombed pattern, there is also some black 
marking about labrum and jaws. In some lights the whole larva has a 
pearly lustre, or perhaps a bloom like that on a plum, would best express 
the effect. Each segment is divided into four (or five) subsegments by 
transverse incisions. The ordinary tubercles are marked by very fine, 
nearly transparent, brown bristles, of which there are also some on the 
head. The segments taper slightly towards the head, more rapidly (in 
two or three .segments) towards the anal extremity. The cocoon (made 
in sawdust, only just under the surface, so that naturally I imagine it 
gets under a dead leaf or piece of wood and does not bury itself in the 
soil) is made with sufficient silk to give it a firm and coherent structure. 
The chrysalis has much the colour, size, and general outline of that of 
Acronycta aceris, but of course very different except as to broad effect. 
The smallest pupa was 20 mm. long by 7 mm. wide, in front of 5th 
