226 [October, 



it to other window sills when the rain hecnme heavy or the wind too 

 violent from the northerly quarter. 



On September 8th it moulted for the second time, appearing 

 with horns of a light nut-brown tint, as in the previous stage. From 

 that date it grew very slowly indeed, eating by night, and resting by 

 day on the upper-side of a leaf upon which it spun a web mat. 

 During this period, as far as we could ascertain, it did not eat by day, 

 nor did it devour the leaf on which it rested. It was in the habit of 

 crawling off its perch at twilight and mounting the sallow to some- 

 where near the top, where it took a short, vii^orous meal off a young 

 leaf, descending afterwards to its webbed resting place. Things went 

 on like this through September and October without any change, 

 except perhaps that it ate less and less after October 15th. By this 

 date sallow began to be very seared and dry ; but we continued re- 

 plenishing the pot at intervals of a week, so long as we could find a 

 single bit of sallow with a leaf or two yet clinging to the stem. 



It may be remembered that towards the end of October last year 

 we had several rather sharp spells of cold, with hail showers and frost 

 at night. On one occasion we left the pot on the northern window 

 sill exposed to the full force of a heavy shower. The larva was 

 drenched, but it made no attempt to seek shelter. It was about now 

 that we began preparing the pot for hibernation. All the earth was 

 removed, and withered saxifrage, the nearest suitable material to 

 hand put in its place. This was arranged round the sides of the pot 

 like a nest, leaving the centre hollow. In this centre the bottle with 

 sallow in it was placed, and about a dozen leafless twigs immediately 

 round it, all ready for the tiny creature to inspect. 



On October 31st it deserted its web mat, and for nearly twenty- 

 four hours crawled restlessly over the dry twigs, settling down at last 

 upon a clean spot where the bark had been torn off. This was at the 

 top of the pot, and completely exposed. Here, with its horns bent 

 forward so that they touched the twig, it commenced its long winter 

 sleep. 



It had now undergone a rather interesting change, having lost its 

 bright green colour, and turned to a muddy, yellowish-brown 

 tint. This change, while not being very decided, was sufficient to 

 make us fear that the larva was going to damp off. However, it re- 

 mained motionless from November 1st up to March 17th of this year 

 without showing any signs of deterioration, except, perhaps, in 

 January, when it looked rather shrivelled — the weather then being 

 unusually mild and trying to its constitution. 



