230 
dually in size in the direction of its mouth, where the head of the 
larva emerges from the case; it is therefore easy enough for the 
caterpillar to extend its nest in this direction whilst remaining within 
it, just as a bricklayer builds the shaft of a manufactory chimney 
from within, but the caterpillar must protrude a considerable portion 
of its body in order to affix the twigs on the outside ; and in this part 
of the business we also perceive an interesting piece of instinct, the 
loose points of the twigs being always directed backwards, so that in 
walking they oppose no resistance to the progress of the caterpillar, 
which they would do were they affixed either in the opposite direc- 
tion or without any order. 
The pupa of the male agrees for the most part with that of Oike- 
ticus Saundersti. The 2nd, 3rd, 4th and 5th segments are armed on 
the back, along the hinder margin, with a transverse series of fine 
recurved hooklets. The 6th segment is destitute of any hooks or 
hooklets, and each of the 7th and 8th segments has a row of stronger 
sharp teeth directed backwards at its base. The correspondence of 
the abdominal segments of the pupa with those of the larva, is 
proved by the appearance in the former of traces of the ventral 
prolegs of the latter. 
The pupa of the female is very robust and, of course, destitute of 
wing-cases, but with slight traces of leg-cases, and also with two 
elongated cases extending from each side of the head, of which it is 
not easy to understand the object ; the thoracic segments are rather 
strongly ridged, and the abdominal segments have lateral spiracles, 
and are also provided on the underside with four pairs of deep im- 
pressions, indicating the position of the ventral pro-legs of the larva 
(a peculiarity which I have not elsewhere noticed in Lepidopterous 
pupe); the body is terminated by two distinct hooks on the under- 
side. In the opened case, represented in fig. 17, I found an empty 
female pupa-case with the head lying towards the narrow apical 
opening ; near the anal extremity of its body the shell of the pupa- 
case had been burst through in several places, and behind it, lying 
among the soft woolly lining of the case, were four pupa-cases of a 
parasitic muscideous fly, of the subfamily Tachinides, which were 
also empty, the flies having made their escape ; behind these, towards 
the base of the case, lay the skin of the larva, shed on assuming the 
chrysalis state. 
The perfect male insect differs from O. Saundersii in its antenne, 
which are bipectinated to the tip; they consist of forty-eight joints, 
and the two thick basal ones as well as the terminal joint are simple. 
The fore legs are slender, the tibia being destitute of the elongated 
spur observed in O. Saundersii. The two hind legs are short and 
thick, with very short robust tarsal joints and strong but simple 
ungues. The fore wings of the male are much shorter than in O. 
Saundersii, and they are traversed to the margins by only eleven 
branches of the veins. From a careful comparison of this wing with 
that of O. Saundersii (Pl. XXXV. fig. 2), it will be evident that the 
wanting branch is the lower discoidal (y), or in other words, that 
the two branches, 3, and y, of the fork in the middle of the apical 
margin of the fore wings of O. Saundersii are coalesced into the 3rd 
