7 6 Proceedings of the Koyal Society of Edinburgh. [Sess. 
GREEN 
YELLOW 
Newton found that almost all colours could be produced by mixing red, 
green, and blue light. This result is not explicitly stated in his Opticks, 
although it is implied in his diagram. Two centuries later, Helmholtz and 
Maxwell developed Newton's treatment almost simultaneously ; and in 
a paper read to the Royal Society of Edinburgh sixty-five years ago. 
Maxwell referred to it as the geographical method of exhibiting the 
relation of colour. According to this method, red, blue, and green are 
taken as the corners of an equilateral triangle, and any colour compounded 
of these three is represented by a point found by conceiving masses 
proportional to the several components of the colour placed at their 
respective corners, and taking the centre of gravity of the three masses. 
I have taken as my standard colours Wratten & Wainwright’s standard 
tricolour gelatine films, now supplied by Kodak Limited, Wratten division. 
These are spectrally very pure indeed, 
and a long way in advance of the coloured 
papers and pigments hitherto used in 
col our- vision work. If we add red and 
green in equal proportions we get yellow, 
which is represented by the point midway 
between red and green (fig. 1). Similarly, 
red and blue in equal proportions give 
magenta, and blue and green in equal 
proportions peacock - blue. The three 
medians meet in white. It is obvious 
from the diagram that peacock-blue, yellow, and magenta are respectively 
the complementaries of red, blue, and green, and that, for example, twice 
as much peacock-blue as red is required to make white. The colours of 
the spectrum lie roughly along the red- green and green-blue sides of the 
triangle, and the violet end lies below blue outside the triangle. Violet 
cannot be produced by mixture of the three primaries. 
Suppose, now, we have two identical patches of colour on a screen, and 
gradually alter the tint of one until the difference in colour becomes visible. 
Then we can represent the gradual alteration of tint by a displacement on 
the colour diagram. If we take the first colour as a fixed point, and 
the displacement in different directions, the point representing the second 
colour traces out a closed curve surrounding it and enclosing an area all 
the colours within which appear alike. The number of such areas or patches 
it is possible to draw inside the triangle is the number of different tints 
which the observer is able to perceive under the conditions of the 
PEACOCK 
BLUE 
BLUE 
MAGENTA 
Fig. 1. 
RED 
experiment in question. 
