148 Mr. E. B. Poulton. Colour-relation between the [Feb. 4, 



Salix cinerea (Nos. 11, 12, and 13) produced very interesting results, 

 varying from a good yellow variety to intermediate, nearly all being 

 upon the yellowish side of the latter. Thus, there is a distinct, though 

 slight, advance upon the effect of this food on the parents in the 

 direction of yellow. 



The results obtained from the three larvae of No. 12 are extremely 

 interesting, showing that individual variation may sometimes play an 

 important part in the colour produced, although the whole of the 

 results of all observations and experiments conducted up to the present 

 time certainly prove that such a factor is generally insignificant, and 

 rarely causes any effects that can be detected. By individual varia- 

 tion I mean the development of a different colour than that which 

 would be produced by the food-plant acting upon a larval tendency 

 which is uniform for nearly all the larva? from each batch of ova, the 

 latter tendency being probably explicable as the inherited results of 

 previous food-plants for many generations. In other words, I mean 

 breaches in the uniformity, however caused, of the larval tendency, 

 and a study of this and the previous paper upon the same subject, 

 will show that such irregularities are comparatively rare, and especially 

 so when the food-plant itself is known to possess a strong influence in 

 the direction of either extreme of coloration. 



Popidus nigra (No. 14). — The results of this food-plant, inter- 

 mediate as far as the evidence went, cannot be compared with any 

 other experience, for this is, I believe, the only instance of the larvae 

 having been known to eat this food-plant. From the green glabrous 

 undersides of the leaves I should have anticipated a tendency towards 

 yellow, which was only partially verified. 



Salix triandra (No. 15). — I was especially anxious to gain abundant 

 evidence as to the effect of this food-plant, because I believed that its 

 tendency was towards yellow. Mr. Boscher described numerous 

 instances of typical white larvae having been found on it. I have, 

 however, since ascertained that Mr. Boscher was mistaken in his 

 identification, and that the trees upon which he found the whitish 

 larvae were Salix alba, and such a result from the latter food-plant is 

 what I should have anticipated. At the same time, I should wish to 

 point out that the identification of the various species of Salix is 

 immensely difficult, and that I have only been saved from hopeless 

 confusion by the skilled assistance of Mr. G. C. Druce, who has most 

 kindly helped me, and, when necessary, has obtained other opinions, 

 throughout this investigation. During the past year (1885) I have 

 proved by observation in the field (as will be seen) that the effect of 

 S. triandra is to produce yellowish varieties, and the same thing is 

 proved by these experiments, considering the hereditary influences. 

 I have therefore verified the prediction upon which I ventured in my 

 last paper (already quoted, p. 306), although at the time when the 



