1(50 Mr. E. B. Poulton. Colour- relation between the [Feb. 4, 



greatly astonished at the colour of the single larva found upon it. 

 Thinking over such a result, I remembered that the one bright yellow 

 variety which I had found upon S. viminalis (on August 11, 1884; 

 see p. 301 of the above-mentioned paper) was upon a variety of the 

 latter plant with very small leaves, while the white larvae were found 

 upon the large-leaved variety which is the common one at Oxford. 

 Shortly afterwards, through the kindness of Mr. Boscher and Mr. 

 William White, I had the opportunity of looking at some twigs of 

 the trees upon which the eighteen spotted yellow larvae were found 

 (see p. 304 of the former paper, &c). The trees were the small-leaved 

 variety, and Mr. Boscher states that all the yellow larvae were found 

 upon such food-plants. Then again I remembered that in the case 

 of the very bright yellow variety found upon crab (var. acerba) on 

 August 14th, 1884 (see former paper p. 301) , the latter tree had very 

 small leaves. Finally, Salix incana at Brigue possessed leaves which 

 were very white and downy underneath, but they were very small, 

 and the larva found upon the tree was yellow. 



All this convergent evidence suggested the following explana- 

 tion. The larvae are only affected by that part of the environment 

 which is so close to them as to be almost or quite in contact ; 

 the tint of mature life is (as far as it is caused by the colour of 

 the food-plant) a resultant of the conflictirjg tints which have 

 formed part of the immediate environment of the larva through- 

 out its life, and the ultimate predominance of one larval tint is due 

 to the relative proportion of the whole larval life during which that 

 tint predominated in the environment. This conclusion is also 

 supported by the breeding experiments and the experiments upon 

 captured larvae. Such being the case, the ultimate whiteness of a 

 mature larva is largely due to the considerable proportion of its 

 earlier life which is spent upon the white under sides of the leaves. 

 (The young larvae invariably take up this position.) During this 

 period white is the only colour in its immediate environment, except 

 when it is actually engaged in eating, and so may perhaps be affected 

 to some extent by the colour of the upper sides. But when the larva 

 reaches a certain size and weight, it must in nearly all cases quit this 

 position and retire to the stem, because the leaf is not strong enough 

 to bear it without being dragged into an unnatural position, cr 

 because it is too small to form a background for the larval body. 

 Therefore the time at which it retires must chiefly depend upon the 

 size and strength of the leaves. Having once quitted the small 

 leaves the larva does not again rest upon them, because they can be 

 entirely eaten from the stem, whereas the large leaves cannot be 

 reached without venturing upon them, and therefore in the latter case, 

 the chances are in favour of the larva being left upon a partially 

 eaten large leaf during many of the periods of rest, even at an 



