62 



of dying etiolated tissue or a dark mark of dead tissue often with 

 a light border. 



2. Or such a scratch may cut the leaf, the hole seen from 

 above showing black shadow with highly lighted border caught 

 by the edge' of the hole. 



3. A round hole pierced in a leaf or a nibbled irregular lesion 

 will present the same view from above. 



But these appearances are accurately those of the growing 

 puss larva seen from above. Its mimetic progress is somewhat 

 thus : — 



a. Newly-hatched — a black streak = a scratch on leaf. 



b. , Second stage — black with a little russet = a budscale 

 fallen from a higher level and resting on a leaf. 



c. Third and fourth stages — ground plan of a central irregular 

 chocolate coloured area with a (functionally valueless) lighter 

 border = a hole in leaf showing shadow through from below. 



These stages, at least b and c, would be better mimetically if 

 the light border had a real functional value, but the exigencies of 

 the final stage have probably eliminated an originally broader 

 pattern. May we suppose, speaking teleologically, that nature 

 having perfected the pattern up to the fourth stage, suddenly went 

 one better, altered her plan, gave up the light line in its earlier 

 function, and used it, narrowed, to pick out a pattern accurately 

 mimetic of leaf form as well as of generalised leaf colouring? 



The correlation of pattern to environment can be justly 

 appreciated only by rearing the larvae in the open in situ on the 

 food plant, preferably on Populus balsamijera. All the stages 

 but the last are lived on the upper side of the leaf — unlike the 

 Poplar and Eyed Hawk Moths whose larvae are coloured with 

 equal fitness for their sub-foliar habitat — the usual position in full 

 view on mid-leaf being maintained by day, the larva feeding 

 chiefly at night and returning to the same position of rest day 

 after day — but in the final ecdysis the larva is pendulous, and 

 hangs back downward from petiole or leaf-edge, frequently feed- 

 ing continuously day and night, and escaping detection unless a 

 tree be overstocked so as to increase the risk of discovery. 



The Puss in its early stages is admirably hidden on the small- 

 leaved species of Salix. Is it possible that it was once smaller 

 and confined to the smaller-leaved Sallows? What advantage 

 does it derive from being larger than the Kittens? Is it to bring 

 the growing larva rapidly past the stage when it can be mastered 

 by Ants? I have lost whole crops of minute larvae from the 

 attacks of ants against which protective colouring probably avails 

 little as they seem to hunt chiefly by touch. 



The shiny triangular patch from the dorsal hump to the head 

 of the Puss would be advantageous chiefly to larvae living on 

 polished foliage — this area reflects grey diffused light very sugges- 

 tive of shiny leafage, and in the inverted position a reflecting 

 surface may be of use for concealment, as is the polished cuticle 

 of Pheosia tremula which reflects the colour of its surroundings^ 



