436 MR. E. R. POULTON ON THE COLOUR-RELATION BETWEEN EXPOSED 
Four pupae were living which had been on the white surface : 
Of these, all 4 were (1), /3, 2 an opaque whitish-green, 1 yellowish, and 1 chiefly grey, bat with a 
greenish tinge ; the grey very marked in all of them. 
Here, again, the surfaces may have produced some slight effect, but the conclusions 
are not certain, because the pupae already formed when the experiment began became 
intermixed with those which had been exposed to these colours during Stage III. 
IV. On September 15 (see Series 1, I., C.) a girdled larva of P. rapce (Stage III.) 
was found on a black tarred fence and was removed (4.15 P.M.), and was placed on 
white paper. At 5.30 p.m. on the next day it had pupated an hour or two, and the 
pupa when it took the permanent tint became a (1), very dark, and like those which 
had been found upon the fence on the 15th. In this case the powerful effects of the 
black surface during Stages I. and II. and the first part of III. could not be altered 
by the exposure to a white surface during the remainder (24 hrs.) of the last stage. 
This experiment is very conclusive against the former theory of pupal sensitiveness. 
Y. October 12, a single larva of P. rapce in Stage I. was found crawling upon a 
chocolate-coloured paling at Seaview ; it was placed in a small chip-box with the other 
pupte found on this date (see Series 2, III.) and became a (4), about normal, with a 
yellowish tinge. In this case the pupal colour appears to have been entirely due to 
the influence of the light-yellowish tint of the chip-box in which both the terminal 
stages were passed. 
VI. October 5, six larvae of P. rapce were found on the green gate (described in 
Series 1, VI., E.), of which three were in Stages I. or II., and three in Stage III. 
(girdled). All six were placed in a cylinder (6 centimetres in internal diameter and 
1'05 decimetre in height) covered with a single layer of black tissue-paper, and with a 
roof and floor of the same material. All the six pupae which were obtained were alive 
when the comparison was made. 
Of the 3 pupse winch, bad passed Stage III. 
in the cylinder and were girdled, 1 was (3), typical, yellowish. 
2 were (4), 1 typical and yellowish-pink, and 1 was faintly 
greenish and very deficient in pigment. 
The 3 pupae which had passed part of 
stage in the cylinder were all .... (4), typical, 1 pinkish, 1 yellowish-pink, and 1 
yellowish. 
Here, again, some slight effect appears to have been produced by the transference, and 
the results harmonise well with those of all the other experiments of the kind. 
In concluding the account of experiments upon these pupae, it must be remarked 
that the effect of the coloured surroundings upon the dark pigment is, perhaps, the 
least important part of the changes produced, for there are other consequences which 
