430 MR. E. B. POULTON ON THE COLOUR-RELATION BETWEEN EXPOSED 
green passing through with hardly any diminution. The transmitted rays were 
especially bright from 51-59. 
The pale bluish-green opaque paper made use of in Experiment VI., C. The 
reflected light differed from that transmitted through the green tissue-paper in the 
much larger amount of blue rays which entered into its composition. The red, 
orange, and yellow were also reflected to a slight extent, but were chiefly absorbed. 
The chief reflected rays extend from 45 to 5 7 on the scale given below. 
The dark-blue opaque paper .—The chief reflected rays extended from 44 to 48, the 
rest of the spectrum being very dim. As the chief rays were not intense, the colour 
was a very dark blue. 
It is, therefore, seen that the colours employed were fairly pure, although, of course, 
they were all mixed with an immense proportion of white light. The spectroscopic 
results indicate that each of the colours should be made into separate ordinates—the 
green tissue-paper and the pale bluish-green being separated because of the different 
amounts of blue in their composition. The red brick walls (III., B.) are omitted 
because of the reasons given at the end of the description of the pupte found upon 
them, and because the spectroscopic character of their colour is unknown; and for the 
latter reason the greens in VI., D., and VI., E., are also omitted. 
In the graphic representation (fig. 6) given below each ordinate which starts from 
the part of the base line which represents the visible spectrum is made to diverge and 
include on the scale the limits of the rays which were shown by the spectrum to 
constitute, at any rate, the chief part of the colour transmitted or reflected by the 
colour indicated on the ordinate. It should be noted that the P. rapce line merely 
passes the ordinate for dark red, for there was no experiment to indicate the amount 
of pigment which is formed under the influence of this colour in the case of P. rapes. 
To the left of the red end of the spectrum four ordinates are added to represent black 
and white under three conditions of illumination, in order to compare their results 
with those produced by certain parts of the visible spectrum. 
The actual lengths were as follows :— 
P. rapce. 
P. brassiccB. 
millims. 
millims. 
White in very strong light 
7-36 
White in light. 
12-91 
White in nearly complete darkness 
22-5 
56-34 
Black. 
47-81 
61-25 
Red. 
m m 
61-09 
Orange . 
8-33 
26-0 
Yellow. 
12-5 
28-0 
Green . 
10-5 
33-25 
Pale bluish-green. 
16-66 
58-54 
Bine. 
18-75 
