﻿NOTES ON HYBOCAMPA MILHAUSERI. 93 



also, but stuck fast just at a tertiary branch, the small remaining 

 portion of which precisely represented the backward process of 

 the larval spine. I never met with another rolled leaf that hap- 

 pened in exact number, size, and position, to represent all the 

 processes of the larva as this one did, but any rolled and aban- 

 doned leaf bore a very close resemblance to the larva. 



The cocoon is the most Centra-like phase of this insect ; it is 

 made on the bark of the tree ; though rather rounder and not 

 quite so flat, it is in favourable specimens very like a Cerura in a 

 similar locality. It differs, however, in several particulars. In 

 its construction the larva first makes a silken net over itself at the 

 site of the cocoon ; this is so open and flimsy looking, and so 

 close down on the larva, which is rolled together under it, that it 

 is surprising how the larva can move about inside to continue the 

 work, and how in doing so it does not tear all to pieces. Cerura, 

 on the other hand (I have watched bifida and bicuspis), selects a 

 place for its cocoon, and begins at what will be the tail of the 

 cocoon by making what looks like its head, a little structure like 

 the toe of a slipper ; it pushes this end forward, working beneath 

 it by loosening its attachments in front, and pushing its head 

 under, spinning a continuation of it ; in this way it gets under and 

 detaches from the surface any bits of lichen and loose fragments of 

 bark, which thus adhere in precisely their natural positions to the 

 surface of the cocoon, and so much help its concealment ; and 

 having in this way pushed the front of the cocoon forward to its 

 place, it closes the hinder end, and begins the excavating process by 

 which it obtains the materials to form the outer part of the cocoon 

 of a sort of artificial bark. H. milhauseri gets to the surface of the 

 cocoon some colouring matters or lichenous material from the 

 surface of the bark, and puts a little detached material round the 

 margin of the cocoon to make it fit a little ; but the face of the 

 cocoon, instead of being like that of Cerura, an artificial bark, is 

 almost pure silk of a dense gummy substance, such as I have seen 

 vinula make under difficulties when he could get no wood to work 

 with. Milhauseri only slightly smooths the bark beneath the 

 cocoon, and covers it with gummy silk. Cerura lies in a neatly 

 excavated cavity in the bark with practically no silken lining. 

 Partly by its original close structure, and partly by contraction 

 as it matures, this cocoon fits the pupa more closely than any 

 other cocoon I have met with, so that, combined with its dense 

 horny texture, it is extremely difficult to open it without injuring 

 the contained pupa. 



This close fitting of the cocoon to the pupa is related to 

 perhaps the most curious of all the facts I have observed in 

 this species, that is, its method of emergence ; which is, I think, 

 perhaps the most remarkable of any of the many remarkable 

 provisions in different Lepidoptera for the escape of the moth 

 from its cocoon. 



