§ 12. Physiological Theories of Light-Sensation.—"Very little indeed is known by direct observation and experiment about the physiological processes either in the retina or in nervous matter corresponding to light-sensation. The theories on the subject are hypothetical constructions based on physical and psychological data. The two which are best known are those connected with the names of Helmholtz and of Hering respectively. Neither of these is satisfactory; but that of Hering is based on a more complete survey of facts; and if it is not right, it may safely be said to be on right lines. It has recently been greatly modified and improved by Prof. Gr. E. Miiller, but his views are at once too complex and too recent for us to deal with them here. We shall therefore refer to Hering's theory mainly in its original form.

The theory of Helmholtz is primarily based on the facts of colour combination regarded from a physical point of view. The aim is to account in the simplest way for the production of the same colour by many different combinations of physical light. Helmholtz believed that this could be done by assuming three, and only three, ultimate physiological processes. Each of these processes takes place in the first instance in the retina and is conveyed by its own special nerves to the brain, where it produces a corresponding specific nervous excitation. The processes severally correspond to the sensations of red, green, and blue. Their combination in equal proportions yields the sensation of white or grey. Every kind and combination of light excites all three processes. Hence no colour under ordinary conditions of stimulation is ever quite saturated. It always contains a certain intermixture of white. By combining in various proportions the red and the green processes, the green and the blue, the red and the blue, all the colours of the spectrum, together with the purple, may be obtained.

This theory seems a highly satisfactory account of the results of combining lights of different wavelengths, so long as we do not test it by psychological analysis of the resulting sensations. But when we do this, a difficulty occurs in the case of white and yellow. By mixing green light with blue light, we obtain a bluegreen. This, says Helmholtz, is due to a compounding of the physiological processes corresponding to blue and green respectively. His account of the matter is borne out by a scrutiny of the sensation itself. A bluegreen partakes of the nature both of blue and green: it resembles both of them at once. It resembles each in varying degrees according as blue or green preponderates. But by mixing red and green lights we produce, not reddish green but yellow. The yellow does not partake of the nature both of red and green, as bluegreen partakes of the nature both of green and blue. No analytic scrutiny of sensation can discover such a colour as a reddish, green. The same is true of white. White, according to Helmholtz, is a compound of all three ultimate physiological processes. But, as a matter of fact, the sensation of white does not partake at once of the three colourtones, red, green, and blue.

Objections of this kind will probably have different weight with different persons. But it so happens that they are confirmed by some very important facts connected with colour-blindness. If white arises through a combination of the three elementary processes, all the colour sensations ought to be possible when the sensation of white is possible. But, as we have seen, there are wellestablished cases of total colour-blindness. Here all three elementary processes are absent, and yet the sensation of white remains unimpaired. On the theory of Helmholtz we must say that the three elementary processes are really present, but that they are on all occasions excited in equal proportions by all kinds of light. This is a rather improbable assumption, but the improbability becomes increased to the verge of impossibility, when we consider that the same hypothesis must be applied to a number of other cases in which coloursensibility is absent, and sensibility to white and black is preserved. All lights of whatever wavelength, produce only neutral sensations, when they act on the retina for a very short time. All the colours of the spectrum pass into grey when the illumination is sufficiently diminished. They pass almost completely into white when the illumination is sufficiently intensified. The extreme outer margin of the retina is sensitive to white, but totally colourblind. Under all these varying conditions we must, according to Helmholtz, suppose that the three elementary colourprocesses are present, and that the only reason why the corresponding colours are not perceived is that the processes are always excited in equal proportions.

An equally serious objection arises from cases of partial colour-blindness. It is evident that, if Helmholtz is right, the absence of one or more of the elementary colourprocesses must involve the absence of the sensation of white, which is due to their combination in equal proportions. "A person who is greenblind ought, upon this supposition, to see in white only its red and blue constituents, and hence white ought to look to him as purple looks to us. As long as his defect made him incapable of explaining to us what he felt, this might perfectly well, for aught we knew, have been the case. But we know now that a person who is greenblind in one eye only sees white with his defective eye exactly the same as he sees it with his normal eye."* A similar argument applies also to yellow. The partially colourblind usually retain the sensations of yellow and blue, although they are without the sensations of red or green or both. There is a marginal zone of the retina at which the sensibility to red and green ceases, and that to yellow and blue is retained. So, with great increase in the intensity of illumination, red and green are still discernible in the spectrum, though yellow and blue disappear. Such facts as these are incompatible with the supposition that yellow is due to a combination of the red process and the green process.

* C. L. Franklin, "On Theories of Light-Sensation," Mind, N.S., vol. ii. (1893), p. 479.

If the theory of Helmholtz is unsatisfactory in its account of colourcombination, its failure to explain other facts of light-sensation is still more conspicuous. It accounts for contrast effects between adjoining colours as errors of judgment. A fuller investigation of these phenomena has shown that such an hypothesis is quite untenable. The colourproduced by contrast appears and behaves in all respects like the colour produced by direct stimulation. Negative images are explained by Helmholtz as due to fatigue. By long continuance, one or more of the ultimate colourprocesses become exhausted, so that the others are predominantly aroused either by stimulation from without, or from the retina's own light. One objection to this view is that, on the principles of Helmholtz, fatigue of all three processes must be constantly taking place, as all three are excited by every kind of light. Now the fatigue which is to explain negative images must take place in the course of a few seconds. Hence we should expect a very conspicuous effect of fatigue from the ordinary use of the eyes in daylight. Hardly any capacity for light-sensations of any sort ought to be left at the end of an hour, especially after exposure to predominantly white light, which must exhaust all three processes equally.