332 MR. E. B. POULTON ON THE COLOUR-RELATION BETWEEN EXPOSED 
A very uniform group. The (2) and dark (3) resemble those described above, and 
have the same relation to the A. division. 
Group 3 was under the white paper, and consisted of 5 pupse, of which all were (3), dark. 
All exactly resembled those described above under the same degree. 
No less than 22 out of the 41 pupse were darker than any of Series III., and, 
omitting the four (2) of the latter, all the pupse in Divisions A. and B., except the 
two lightest (3), are darker than any in the preceding series. There were pupse among 
the dark (3) in Series III. with a greater amount of pigment than some of the dark 
(3) in the A. division, but the pigment in the latter was much darker in colour, 
although their pink ground-tint was also more distinct. It is, however, perhaps safer 
to exclude the five (3) of A. (IV.) from the above general statement as to the 
comparative darkness of these divisions and of Series III. 
The comparison instituted above seems to show that these larvse tend, as a whole, 
to produce darker pupse than those of Series III. Such a conclusion shows how 
important it is in these experiments to keep the larvse of the different companies 
carefully separate, for the error due to the different tendencies largely disappears in 
the experiments conducted within the limits of a single company. It appears that 
the white paper added when the preparatory period was far advanced nevertheless 
produced some slight effect, for none of the pupse suspended under it were darker 
than dark (3), although on the more shady side of the cylinder. 
C. Some of the larvse from the other smaller cylinder, covered with “ opal ” glass, 
were (August 16, evening) placed in a cylinder (2’64 decimetres in internal diameter, 
and 9 centimetres in height) which had a roof of white glazed paper and a floor of 
ordinary white paper (common unruled white foolscap was always used for the white 
paper floors). This was placed in a good east light, about 5 feet from a large window. 
None of the larvse in the former cylinder were suspended when the change was made, 
but many were prepared for suspension. The following results were obtained :— 
Of 3 pupas on the floor, but not fixed to the 
food-plant.2 were (3). 
1 was (4). 
,, 6 pupae in a small group suspended from 
the roof ..all were (3), 1 of them dark (3). 
,, 6 pupae in another small group, suspended 
from the opposite side of the roof . . 1 was (2). 
5 were (3), 4 of them light, 1 with unusual gold, 3 
with unusual pink. 
„ 1 pupa, suspended to the side close to the 
— last group, .. was (3), dark. 
16 
The results show clearly that the larvse of this series were greatly influenced by 
white surroundings, for the above list is a very sharp contrast with either of the 
