In Hering's theory, a strenuous attempt is made to escape the difficulties which beset that of Helmholtz. Following the clue given by psychological analysis of light-sensations, he assumes six ultimate processes, corresponding to the sensations of white, black, red, green, yellow, and blue. These he arranges in antithetic pairs; white and black go together, and similarly red and green, blue and yellow. To each pair there corresponds a separate retinal substance, and a distinct modification of central nervous matter. The redgreen substance is susceptible of two antagonistic processes, chemical in their nature. Red light excites the one, and green the other. "When red and green light are combined in equal portions, neither process is produced because of their mutual incompatibility. Hence there is no such colour as a reddish green. But both the red and green act on the blackwhite substance, so as to produce the sensation of white. When red or green sensations occur, their intensity is mainly due to this excitement of the blackwhite substance. Hence, when red and green light act simultaneously in equal proportions, though the two stimulations neutralise each other as far as their colour effect is concerned, they continue to act on the blackwhite substance, and produce the sensation of white. The relations of the blue and yellow processes are analogous. The black and white processes are supposed by Hering to be antagonistic in much the same way; but as a matter of fact there must be an essential difference here, as black and white combine to form intermediate greys, so that the two processes are not incompatible any more than the blue and green, or the yellow and red processes are.*

* " Really, black and white do cancel each other in the retina: there is no grey process there. But the cortical cells with which the optic nerve is connected are always in a state of commotion (owing to changes of temperature, etc.), whether there is a stimulus before the eye or not; and this commotion gives us the 'intrinsic'or'subjective' sight-sensation, the sensation of grey." Professor Titchener, Primer of Psychology, p. 42. (This proposed explanation is due to Gr. E. Miiller, who works it out with great care.)

Hering is no doubt right in assuming that the processes corresponding to black, white, red, green, blue, and yellow, are separate and distinct in their nature, so that none of them can be resolved into combinations of the others. But he probably pushes this point too far in assuming that his pairs of antithetic processes always take place in separate substances. There is no doubt that there is a special apparatus connected exclusively with the whiteblack process. Total colour-blindness and allied facts seem to clench this conclusion. But it does not follow that the whiteblack processes cannot also take place in those retinal elements which subserve the colour processes. When blue and yellow lights act simultaneously so as to give rise to the sensation of white, Hering has good reason for denying that the blue process and the yellow process take place together. As such, they neutralize each other; but when he refers the conjoint effects of the two lights merely to their action on the whiteblack substance, his position is not without difficulty. Since the yellow and blue processes neutralise each other, the combined intensity of these processes must be deducted from the resulting sensation of white. This involves the assumption that the intensity of the yellow and blue processes, as such, is very small indeed, and that when they actually occur, the brightness of the corresponding sensations is mainly due to the intermixture of white. This is hard to believe. "Hopeless confusion is introduced into all our conceptions of colour when we are asked to believe that the entire brightness of every sensation of light is nothing but the brightness due to the white sensation which is mixed with it. . . . Can they be thinking beings who have allowed themselves to follow Hering into the intellectual vagary of supposing that a perfectly saturated red, for instance—that is a red wholly free from white admixture—no matter what the amount of chemical activity which called it forth, would have no brightness whatever, that there would be nothing in sensation corresponding to differences in amount of this photochemical process?" * Hering's view cannot be set aside in this sweeping way. But we can scarcely accept it without fuller evidence than has yet been adduced in its favour. If Hering is right in holding that brightness is almost entirely due to the effect which it produces on his whiteblack substance, colours, which are indistinguishable in intensity and tone when the lights which produce them fall on the yellow spot, ought also to be indistinguishable when the lights which produce them fall on the totally colourblind margin of the retina, and are seen as greys. As a matter of fact, this appears to be approximately true. On the other hand, two colours apparently identical in brightness and tone, one produced by a mixture of spectral lights, and the other by homogeneous light, may differ very greatly when they are seen as greys in twilight vision. But this can hardly form a cogent argument against Hering, because the conditions of twilight vision differ essentially from those of ordinary daylight vision. (See § 6, ad Jin. Total Colour-Blindness)

• Franklin, op. cit., p. 480.

The red and green of the spectrum in combination produce yellow. Hering accounts for this by the composite nature of the spectral red. It contains an admixture of yellow: and when the red and green lights neutralise each other, the yellow alone is left. He may be right, but this is one of the points on which it is difficult to see that his theory is quite satisfactory. It would in some respects be preferable to suppose that the yellow process* can be produced by positive cooperation of red and green lights.

* Viz., a distinct retinal process corresponding to yellow, not a combination of the processes separately produced by red and green lights respectively.

Hering's theory may easily be made to account for the phenomena of contrast and negative images. They are due to a disturbance of the chemical equilibrium of the retinal substances. For instance, we may suppose that the white process has for its product the accumulation of material for the black process, and vice versa. Hence the white process will positively tend to give rise to the black process, both in the portion of the retina affected and in the adjoining part. Whatever may be the special details of the process, it seems clear that both contrast effects and afterimages are due to a positive tendency on the part of each of the ultimate retinal processes to produce its complementary process. In conclusion, we may say that the theory of light-sensation, although beset with difficulties, is in a hopeful condition. New facts are being constantly discovered, and more exact quantitative measurements being made. New theories are being propounded, and old theories modified in accordance with these fresh discoveries, and, on the whole, satisfactory progress is being made. It is, above all, interesting to note that those attempts are most successful which follow most accurately psychological data and psychological analysis. In this respect, the comparison of the predominantly psychological method of Hering with the predominantly physical method of Helmholtz is instructive.*

* The general plan of exposition and much of the detail in this chapter is taken from Ebbinghaus. Grundziige der Psychologie, Erster Halbband, Leipzig, 1897, p. 320 ff. The student who can read German should consult this work. In English the treatment of Light-Sensation in Foster's TextBook of Physiology, fifth edition, book iii., chap, iii., pp. 12221247, is excellent. See also Mrs. C. L. Franklin's article "On Theories of Light-Sensation," Mind, N.S. vol. ii. (1893), pp. 473489. The latest developments are mainly to be found in German and especially in the pages of the Zeitschrift fur Psychologie und Physiologie der Sinnesorgane, where the contributions of Konig, von Kries, and G. E. Miiller are most important.