126 
Proceedings of the Royal Society 
distinct. This is particularly the case with delicate tints, as they are 
not easily distinguished if not contrasted with white. 
Other forms of apparatus with lenses may he used for combining 
colours ; an ordinary telescope answers fairly well, hut an opera- 
glass will not do. The colours to be combined are held close to 
the front of the object-glass end of the tube, if they are transparent, 
such as glass plates ; if the colours are opaque, they are placed in a 
good light, and the tube directed towards them. On looking 
through the eye-piece we get the result of combining the colours 
in the proportion in which they are exposed to the end of the 
tube. The object-glass of the telescope is not necessary for these 
experiments. The lenses of the eye can also be made to combine 
the colours without any assistance, but the results are not very 
satisfactory. If we fix the coloured transparent media with their 
edges close to each other on a piece of glass, and hold them very 
near to the eye, confining the entering light to a small angle, we 
shall then see the colours change through the different combinations, 
as they are moved across the eye. 
With regard to the selection of the colours for these experiments, 
a few words are necessary. We may of course select any colours we 
please, and the instruments will combine them for us, and show us 
the resultant colour or sensation. We shall find that there is an 
infinite variety of colours which, when combined in proper propor- 
tions, will give us white. If, however, we wish to arrange the 
apparatus, so that it will not only give us white, but will also 
produce all possible colours, with their shades and tints, then the 
only three colours which, so far as I am aware, will produce these 
results, are red, green, and violet. With these three colours we can 
make white and all colours, such as yellows, blues, and purples 
of their different shades and tints, but by no combination of colours 
with which I am acquainted can we make red, green, or violet. 
In selecting the colours for the instruments, I took for red a ruby- 
red glass, dark enough to give only a band in the red when 
examined with the spectroscope ; for green, a glass which passed a 
narrow band of light in the green of the spectrum, and is about 
what is known as emerald green; and for violet a glass which 
passed only the light of the violet end of the spectrum. It is 
impossible to describe these colours more definitely by means of the 
