We have said that there are cases in which persons are only defective in colour perceptions, and not wanting in them altogether. The former are more common than the latter, and to the experimenter are by no means so interesting. They are only alluded to here to indicate that there are degrees in the defectiveness of eyes to colour. One point which must be remembered here is that all colour production for registration by the mixture of three colours is delusive, unless the eye of the operator is tested for its colour sense. CHAPTER XII. Formation of Colour Equations--Kœnig's Curves--Maxwell's Apparatus and Curves. The plan of obtaining colour equations will by this time have become fairly evident. And we may as well illustrate it by equations obtained with the apparatus we have been using in our previous experiments. Let us suppose we have an individual who is desirous of having his eye-sight for colour tested, and that we have the slide with the three slits _in situ_. It will be found that when we alter their width and form white light with them, matching in purity the white light of the reflected beam, that we shall have to reduce the intensity of the latter very considerably, by means of the rotating sectors. The aperture may sometimes be as small as 4°, and at other times perhaps somewhere between 4° and 5°. Now the variation in aperture between 4°, and say 4·7, is very considerable, but it is highly probable that the latter might be estimated as 4·6, since only degrees are marked on the sectors. It therefore becomes essential to use a less brilliant reflected beam for the comparison, and this is secured by using as a mirror a plain unsilvered glass. What before read 4 will perhaps read 60, and 4·7 will be 70-1/2, whilst 4·6 would be 69, a difference easily read. We can now commence operations. Let us then place the red slit at say (35) of the scale, the green at (28), and the violet at (17), and make white light of the same intensity by altering the apertures of the slits. Let us do the same with the slits at (34), (28), and (17), instead of at (35), (28), and (17); and again make white light, and similarly with the slits at (35), (28), and (18); and let the following be the results-- (1) 20(35) + 60(28) + 40(17) = 100 W (2) 10(34) + 55(28) + 40(17) = 100 W (3) 20(35) + 59(28) + 10(18) = 100 W Subtracting (1) from (2) we get-- 10(34) = 20(35) + 5(28) or (34) = 2(35) + 1/4(28) which means that the colour sensation at (34) is made up of two parts of the sensation of (35), together with 1/4 part of the sensation of (28). In the same way we find that the colour sensation of (18) is made up of the sensations of (17) and (28). (18) = 4(17) + 1/10(28).