upon in this way was holed in a growing, sappy Hmb of the tree, 

 and in the following spring was found dead and mouldy. I then 

 tried holing them in the stump of a stout limb, the upper portion of 

 which was sawn off some years ago, and which, although a part of 

 the growing tree, is to all intents and purposes dead and sapless, 

 and from each of these the moth has duly appeared in the following 

 summer. My most recent experiment was with a larva brought to 

 me last autumn. It had been found wandering across a pavement, 

 and was slightly damaged about the anal segment. Instead of 

 holing it in the tree, I placed it on the limb where there were several 

 holes that had been previously tenanted by other larv^. After 

 crawling about for some little time it found one of these holes, at 

 once commenced to squeeze itself in, and ultimately disappeared 

 •wathin the hole, evidently having found what appeared to be a suit- 

 able place for pupation.* 



Then there are the pupae skins found last summer, and referred to 

 at the commencement of these notes. The trees in which the larvae 

 had fed grow in a flower border, about a couple of feet inside a low 

 park paling that divides the said flower border from an asphalt path. 

 The bottoms of the staves of which the paling is composed, pre- 

 sumably having become rotten by their contact with the earth, a 

 stout oak skirting has been placed along the bottom of the paling 

 on its outer face, i. e. the side next the footpath and the furthest from 

 the trees. Owing to the irregularity of the fence a space is left 

 between it and the skirting, varying in width from about half an inch 

 to perhaps a couple of inches, and some six inches deep. This space 

 has in course of time become filled with rubbish of one sort and 

 another that has been blown into it by the wind and fallen from the 

 overhanging branches of the trees, etc. All the pupae skins found 

 were protruding from this rubbish between the paling and the skirt- 

 ing, or from rubbish similarly collected behind " spurs " that had 

 been put in on the other side of the paling to support the posts, and 

 in many cases I was able to trace a hole through the staves of the 

 fence, just above the level of the flower-bed, by which the larva had 

 made its way to the rubbish-filled space in which the cocoon was 

 also found. 



In considering the evidence that we have before us we should, I 

 think, not lose sight of the affinities of the species that we are deaUng 

 with. The older authors appeared to consider Cossus in some way 

 allied to the Bombyces. Recent investigation shows, however, that 

 it has nothing in common with that group, but is closely related to 

 a group of the Tortrices, or one very closely allied to them. The 

 point is very clearly worked out by Dr. Chapman, who tells us in a 

 paper on " Some Neglected Points in the Pup^e of Heterocerous Lepi- 

 doptera" (" Proc. Ent. Soc," 1893, p. 112), "In Cossus I can find 

 no character at any stage to distinguish it from Tortrices. The pupa 



* Since writing the above I have had another opportunity for a like experi- 

 ment with exactly similar result. 



