§ 4. Descriptive Analysis of Light-Sensations.—We must distinguish between neutral tints and colours proper. Neutral tints consist of black and white and intermediate greys. Starting with pure black, we can arrange the greys in a series, so as to pass by gradual transitions to pure white. Each grey may be interposed between two others which it resembles so closely as to be barely distinguishable from them. It differs from the one which precedes it in being a little lighter, and from the one which follows it in being a little darker. Thus, though the greys differ, the general form of transition between them is throughout identical.

* Cleland, Evolution, Expression, and Sensation, pp. 77,78. Psych.

The eye is capable of distinguishing about 700 shades of grey, from the deepest black to the most brilliant white. It should be noted that though black is not due to a positive physical stimulus, as other visual sensations are, it is yet a positive experience. The eye which sees darkness is not at all comparable with the back of the hand, which sees nothing. There is reason for believing that the grey field which is present to consciousness in the absence of light is due directly to a brainprocess, and does not involve excitation of retinal elements at all.

Differences of colourtone, apart from differences of saturation and intensity, are best studied in the order in which they occur in the spectrum. The spectrum is formed by passing ordinary white light through a prism, and so breaking it up into its component simple lights, and projecting these on a screen. The simple components of the white light are then arranged in a series in the order of their wavelengths. At one end are the longest wavelengths, giving the sensation of red, at the other the shortest, giving the sensation of violet, viz. a blue tinged with red. Between the red end and the violet end are interposed all the various colourtones,* with the exception of the purples. The purples can be formed by intermixing red and violet lights in varying proportions. In what follows we shall suppose the spectrum completed by the addition of these purple tints, so as to form a closed figure.

* Not of course all degrees of saturation and intensity.

"We have said that colours are best studied in the order in which they occur in the spectrum. But unfortunately the spectrum is unsuitable in other respects for the analytic comparison of colourtones. In comparing a series of colours merely with reference to their colourtones, their brightness and saturation ought to be kept as uniform as possible. But the colours of the spectrum differ greatly in brightness. Hence in what follows we shall suppose a series of colours arranged in the order of the spectrum, but uniform in brightness and saturation. Such a series may be made by taking bits of transparent coloured paper, and adjusting their degree of brightness and saturation by placing bits of grey or white paper underneath them.

The whole series of colourtones, beginning with red and returning to red, is continuously graduated, like the grey series of which we have just spoken. But there is an important difference. In the region of greatest wavelengths, the transitions are from red to yellow ; each member of the series is interposed between two others which it resembles so closely that the difference is barely perceptible, but it differs from the one in being redder, and from the other in being yellower. Thus the form of transition in the series is uniform throughout, and is quite analogous to that between black and white. But after passing yellow, there occurs what may be best described as a change of direction.

Fig. 2.—Circle illustrating serial order of colourtones.