446 ^Ir. Poiilton's/?/r^//^r r.rprriments upon 



Green glass in front of reflecting backgrounds produced 

 far more effect in the direction of green pupfe than 

 green backgrounds (devoid of yellow and orange) in 

 strong light. With non-reflecting backgrounds it pro- 

 duced dark pupae. 



Pale blue f/lass, transmitting much besides blue, simi- 

 larly produced far greater effects than blue backgrounds 

 in strong light, when placed in front of reflecting sur- 

 faces, but produced the same dark pupae when the back- 

 grounds reflected but little of the light it transmitted. 



Deep blue glass, with one or two exceptions, always 

 produced dark pupae, and the exceptions were evidently 

 not related to the background, but were simple irregu- 

 larities such as often occur in darkness. 



So far as the experiments go, they point to the same 

 conclusions as those reached in the case of V. io. 



Conflicting colour experiments. — The results of Experi- 

 ment 32 (XCV.) entirely confirm those upon V. urtiece 

 and T^. io. They also show that the influence of black 

 was stronger than the strongest antagonistic influence 

 when both were working together upon the same 

 organism ; for, out of 7 P. brassiccs, 5 were in the 

 darkest class except one. 



D. Experiments upon the Cocoons of Lepidoptera. 



For several years I have, from time to time, made 

 experiments of this kind, whenever I met with a suitable 

 species. 



If the power occurs at all, we should chiefly expect to 

 meet with it in species building cocoons upon the surface 

 of the ground among leaves, sticks, &c., or freely ex- 

 posed on bark, and, among such species, especially in 

 those which pass the winter in the pupal state. Ex- 

 amples of these are to be found below ; but all the 

 experiments upon them have yielded negative results, 

 except those upon the genus Halias. 



The first three species have been experimented upon 

 during the last few years, but I cannot now fix the exact 

 date, 



Cerura vinnla. — Some careful experiments were con- 

 ducted upon this species. The cocoons bear the most 



