APPENDIX. 487 



did not, however, attempt to eat. When the leaf was eaten away 

 to the stem on one side, the other side was treated in the same way, 

 leaving the apical part of the leaf untouched, i.e., all that part beyond 

 the point at which it had commenced feeding. When it had devoured 

 the young leaves at the top of the plant (which it ate to the last 

 morsel, and in some cases the shoot also), it tried the larger and 

 more mature leaves, and, after serving some of them in the manner 

 described, it descended the main stem and ascended the lateral 

 shoots, which it stripped of the young leaves as it had done at the 

 top of the plant. When moving from point to point on the plant 

 in search of fresh food, it had a peculiar motion, swaying gently 

 the head and front segments (from the 2nd abdominal), accom- 

 panying this action with a movement oi slow recoil and extension ; 

 had this movement been a quick instead of a very measured one, 

 it might have been described as a darting action. Whether the 

 larva, when thus occupied, was searching by sight or scent for its 

 next point of attack upon the foodplant, can only be conjectured. 

 It did not leave the plant until it was fullfed and ready for pupa- 

 tion. Whilst feeding it would sometimes assume a striking posture 

 in which, sustaining itself by the last three pairs of prolegs 

 only, it would stretch out and bend over its thoracic segments 

 until the head rested on the 1st pair of abdominal prolegs ; this 

 attitude brought into strong relief the two ocellated blue spots 

 which, stretched to their full extent on the dilated segment, resembled 

 two staring eyes, while the pendent segments and head below looked 

 like a long proboscis. More generally the attitude of repose was 

 that of quiet extension, with the thoracic segments contracted ; on 

 these occasions all that was visible of the ocellated spots was a 

 slight suggestion of the posterior edging of their black circumscrip- 

 tion, as the rest of their iorm was covered by the heavy enwrapping 

 skinfold of the mesothoracic segment (Dollman). After ceasing to 

 feed the larva rests for an hour or so, and then commences to pass 

 its mouth over the body, secreting a rather sticky-looking fluid as it 

 does so. It applies the fluid to all parts of its body, working 

 slowly and carefully. The thoracic segments do not escape this 

 treatment, and, to reach them with its mouth, the larva has naturally 

 to go through some difficult contortions. The fluid dries rapidly on 

 the skin, and, as far as I can see, the only object of this proceeding is 

 to toughen the skin ; certainly the skin is much tougher in the 

 morning, and by this time the change of colour is complete. It 

 takes the larva many hours to thoroughly cover every part of its 

 body with the secretion. An hour or so after sunrise, when the 

 heat is beginning to make itself felt, the fullfed, and now 

 darkened, larva becomes restless. If it be now taken from 

 the tree and placed on some loose mould mixed with rubbish 

 or dead leaves it will at once burrow into the heap, and 

 commence to spin a cocoon, but if left to itself on the tree it 

 becomes more and more restless as the power of the sun increases. 

 It makes its way down, or falls to the ground, and, being 

 thoroughly warmed, it starts off at a rapid pace, covering a con- 

 siderable distance before it decides to spin up. I have often found 

 them on the road at this period. The late larvae, influenced by cold, 

 do not go through the darkening process in the normal time ; the 



