294 



Mr. Poulton's notes in 1886 



how the outline of the larva on the branch is approxi- 

 mated to that of (a) by means of the processes. 

 Although the cleft is largely filled up in (c), a con- 

 siderable furrow remains, but this is not apparent 



because of the light colour of the fleshy processes, which 

 prevent the attention from being directed to the shadow 

 which would otherwise indicate the position of the 

 groove. The processes therefore attain the object of 

 softening the contact between the larva and its food- 

 plant in a two-fold manner, by partially filling up the 

 cleft, and by neutralising the shadow in the groove 

 which remains. I have also noticed the processes in 

 the larva of A. betularia, and I believe that they are of 

 very general occurrence in Geometrce. The appearance 

 of such structures in the one small part of the larva 

 where su ch a cleft exists and their absence elsewhere, 

 together with their obvious function in this and the 

 other cases, seem to render Prof. Meldola's explanation 

 a matter of certainty. 



£. An extreme instance of the specialisation of a larva 

 to its normal food-plant. — We occasionally meet with 

 larvae which are specialised in relation to the minute 

 details of their peculiar food-plants, or of their com- 

 monest food-plants, if the larvae occur upon more than 

 one. Such specialisation in the details of protective 

 resemblance would seem to imply an extremely ancient 

 association of the larvae with their food-plants, and it 

 may occasionally aid us in deciding upon the ancestral 

 food-plant of a larva out of the many species of plants 

 which may be eaten. Such an ancestral association 

 must have existed between the larva of Deilephila 

 Hippophaes and its food-plant, Hippophaes rhamnoides, 



