4G 
species, different broods, Ac.; prothorax rather wider than remaining 
segments, scuteUum rather large, plain, bearing four long hairs at 
anterior margin, subdorsal and lateral hairs on this segment large and 
prominent; remainder of body even width; division of segments 
distinct; anal claspers widely spread, hairs on anal segment more 
prominent than on the intermediate ones. Prolegs with a circle of 
hooks. Abdominal subsegments apparently four in number, the 1st 
very extended, ? subdivided. Tubercles not thoroughly studied ; those 
on 8th abdominal are set four-square ; there are three lateral tubercles, 
supra-, post-, and subspiracular, forming an irregular triangle. 
Spiracles appear somewhat raised. 
The larvae of autumnata, like those of jilitjrammaria, soon settle 
down to feed, and are thereafter scarcely ever seen off their food-plant 
unless laid up for a moult; J mention this because it is in somewhat 
marked contrast to the habit of dilutata, which for some days after 
hatching are of a very restless and roving disposition, frequently being 
noticed right away from their food-plant, and escaping whenever they 
can find any possible point of egress. After feeding for two or three 
days the larva; of all the species become lighter and more yellowish - 
brown, and by the end of the first week are dirty greenish-yellow or 
yellowish-green, varying somewhat inter se, and have about doubled 
their length. Dilutata are, however, much more liable to hang back 
and grow irregularly than their allies ; during some cold weather in 
March-April, 1897, some of my dilutata hardly made any progress at 
all in a fortnight, while jili<iranimaria kept at the same temperature 
went forward at an almost normal pace. 
I have no very complete notes on the appearance of the larvae at 
the first moult, which is reached in about five to eight days, according 
to the the temperature,the colour is as last described, with the head, 
scutellum and anal plate brown; I found no valid differential characters 
between the species in this skin. 
In the second instar, the colour is greener, although rather a 
nondescript yellowish grey-green; tubercles minute, bristles very short 
and inconspicuous. At this stage it became easy to distinguish dilutata 
from its allies, as it was much yellower-green (though somewhat blue- 
green ventrally), a trifle stouter in proportion to the length, the brown 
head and anal plate lighter, and the prothoracic plate almost concolorous 
with the rest of the body, only very slightly tinged with brown 
posteriorly. I also here noticed a quite perceptible difference between 
Jilvjrammaria (two broods, Bolton) and autmnnata (three broods, 
Aberdeen and Kincardine) ; and shall be greatly interested to see how 
far it holds constant. The day after the first moult jiliijrainmaria is 
yellowish, with yellower lateral line, and even the subdorsal distinctly 
traceable; autumnata, on the other hand, seems invariably of a more 
nondescript olivaceous shade, and markingless. After feeding in this 
skin both species tend to become gradually more and more green in 
colour; but filii/rauimaria is the more decided green, with distinct 
yellow subdorsal and lateral lines (sometimes even quite prominent); 
autmnnata is of a peculiar olivaceous green, or grey-green or even (in 
some specimens of one of the broods) grey, with hardly any green 
* All my time indications must be taken with some caution, ns mv larva: were 
not kept fully exposed to the weather. 
