NOTES ON THE GENUS CERURA (dICRANURa). 141 



last year's specimens. One, howevei', formed about a foot from the 

 ground, facing east, was perfectly fresh, but alas ! its contents had 

 disappeared. It was a beautifully made cocoon, of a greenish-white 

 hue, admirably harmonising with the piece of birch bark to which it 

 adhered. It had already been detected by a tit or woodpecker, and 

 ruthlessly torn open from end to end. This must have occurred but 

 a day or two previously, as the characteristic larval skin, with its two 

 tails, was lying exposed on the bark, and would certainly have been 

 blown away with the first gust. This was extremely annoying, but, of 

 course, inspired one with hope and determination to succeed. I found 

 no more, however, though I minutely examined that tree up and down, 

 smooth bark and rough, and all the birches and stumps around for 

 yards. So far the only addition to my collection was a varied sample 

 of empty cocoons, which I always make a point of preserving. I 

 may here make a few general remarks as to their position. They 

 were all on trees of considerable growth, on dry soil, and with 

 fairly open surroundings, and I took special note of the fact that 

 they occurred on all sides of the tree, with a possible preference 

 in the case of leaning trees for that side where they would be 

 most protected from the rain. The majority of trees possessed low 

 boughs, on which the larvte had probably fed, and could no doubt have 

 been procured in the autumn. But this was not always the case, the 

 lowest boughs on some trees being ten or a dozen feet above one's 

 head, and thus quite out of the ordinary range of inspection. Some 

 cocoons were spun in the crevices of the rough blackish bark, where 

 they were difficult to see, but quite as many, I think, occurred on the 

 smooth part, especially Avhere the silver was slightly toned with grey- 

 green lichen. The smooth groove immediately below a small branch 

 was sometimes chosen, and, in such a position, I am of opinion that it 

 is easier to detect than that of C . birida on a poplar trunk. 



My next expedition, on March 27th, resulted in the find of a new 

 locality on a heath near Horsford, some two miles further west, and 

 here 1 took my first full cocoon. It was situated in a channel of the 

 smooth greenish-white bark of a big solitary birch-tree, on the south- 

 west side and about four feet from the ground. The cocoon was of 

 large size and firm texture, and, though assimilating to its surroundings, 

 I saw it at once, and wondered how it could have been overlooked by 

 the birds. Possibly a hawk, hovering over a neighbourmg pinewcod, 

 may have had something to do with this, and indirectly aided its 

 pf'eservaticn. Certain it is that this tree had been ignored, for its 

 white papery bark was devoid of the scratches of birds' claws which 

 characterised so many others. It Avas half an hour before I had 

 secured my prize, as I decided, rather than open the cocoon, to probe 

 the fresh and bleeding bark with a sharp knife to its depth in a circle of 

 three inches, which I successfully accomplished. This was encouragmg, 

 but further search only revealed two old cocoons on smaller trees by 

 the roadside. Several other expeditions were made amongst the birch 

 scrub of the heath, but without avail, until April 8rd, when I once 

 again scented ('. bicnspis, finding three more old cocoons on small trees 

 growing in the heather, within a hundred yards' radius of the former 

 place. Returning to where I had left my machine, another hundred 

 yards off, I made my final search on a tree, old and wrinkled with age, 

 but of no great size. Here I had to creep on all fours under the 



