344 



NATURE 



[August i i, 1892 



effects, they are not altogether invisible : their colour is bluish- 

 grey. The only optical~\.\i2iX. is, strictly /-^^^zVa/— difference 

 between the several ether-waves in the visible or invisible 

 spectrum is frequency of undulation, or, otherwise expressed, a 

 difference in wave-length. The chromatic — that is the colour- 

 producing— effects of the ether-waves depend on their power 

 of exciting sensations of colour, which vary with their frequency 

 of undulation. 



Although the retina is extremely sensitive to differences in the 

 frequency of ether-waves, it is not equally so for all parts of 

 the spectrum. In the red and blue portions, the frequency 

 varies considerably without producing marked difference of 

 colour effect, but in the region of yellow and green, compara- 

 tively slight variations in frequency produce appreciable differ- 

 ences of colour sensation. One striking difference between the 

 effect of ether-waves on the eye and sound waves on the ear is 

 the absence of anything corresponding to the octave of tone 

 sensation. The ether-waves in the ultra-violet, which have 

 twice the frequency of those of the red end of the spectrum, 

 give rise to no sense of redness, but merely that of a bluish- 

 grey. Even within the octave there are no harmonies or dis- 

 cords of colour sense corresponding to those of tone sensation. 



Colours are commonly defined by three qualities or constants, 

 — hue, purity, and brightness. Their hue depends upon the 

 chromatic effect of frequency of undulation or wave length. 

 Their purity or saturation depends on freedom from admixture 

 with sensations produced by other colours or by white light. 

 Their brightness or luminosity depends on the degree to which 

 the sensory mechanism is stimulated. The loudness of sound 

 depends on the amount of excitement produced in the auditory 

 mechanism by the amplitude of sound waves ; but a sound with 

 small amplitude of undulation may seem loud when the nerve 

 apparatus is unduly sensitive. The brightest colour of the 

 spectrum is orange-yellow, but it does not follow that the 

 amplitude or energy of the ether-waves is greater than in the 

 region of dull red. There is no physical evidence of greater 

 amplitude in the orange-yellow, and its greater luminosity is 

 no doubt purely subjective, and arises from the greater commo- 

 tion induced in the sensory mechanism. 



The theory of colour sense long ago proposed by Sir Isaac 

 Newton^ is now commonly treated with what seems to me very 

 undeserved neglect. Newton supposed that the rays of light 

 induce vibrations in the retina which are transmitted by its 

 nerve to the sensorium, and there induce different colour sensa- 

 tions according to the length of the incoming vibrations — the 

 longest producing sensations of red and yellow, the shortest 

 blue and violet, those of medium length a sense of green, and a 

 mixture of them all giving a sense of whiteness. At the be- 

 ginning of thi-i century Thomas Young proposed a theory which 

 seems to have been intended as a modification of that suggested 

 by Newton rather than as a substitute for it. Young supposed 

 that the ether-waves induce vibrations in the retina "whoe 

 frequency must depend on the constitution of its substance ; but 

 as it is almost impossible to conceive that each sensitive point 

 of the retina contains an infinite number of particles, each 

 capable of vibrating in unison with every possible undulation, 

 it becomes necessary to suppose the number limited to three 

 primary colours, red, yellow, and blue, and that each sensitive 

 filament of the nerve may consist of three portions, one for 

 each principal colour. " - Soon afterwards he substituted green 

 for yellow, and violet for blue, so that he came to regard red, 

 green, and violet as the three fundamental colour sensations, by 

 mixture of which in varying proportions, all other colours, in- 

 cluding white, are produced. Voung believed that his sug- 

 gestion "simplified the theory of colours, and might therefore 

 be adopted with advantage until found inconsistent with any of 

 the phenomena." 



Young's trichromic theory of colour sense was adopted by 

 Clerk-Maxwell and Von Helmholz, and underwent important 

 amplification. Helmholz suggested that the three sets of 

 fibres supposed by Young to exist in the optic nerve are con- 

 nected with three sets of terminals in the retina ; that each ter- 

 minal contains a different visual substance capable of being 

 decomposed by light ; that when the substance in the red nerve 

 terminal undergoes chemical change its nerve fibre is stimulated, 

 and the excitement travels to a cell in the brain by whose 

 specific activity the sensation of red arises. In like manner, 



' See quotations from Newton made by Young in Referenc; 2. 

 -' Thomas Young, "On the Theory of Light and Colours," Phil. Trans. 

 Lond., 1802, p. 12. 



NO. r 189, VOL. 46] 



when the visual substances in the green and violet terminals are 

 decomposed, nerve impulses travel through different fibres to 

 different cells in the vision centre, by whose specific activities 

 the sensations of green and violet arise. With Helmholz there 

 was no question as to difference in quality of sensation depend- 

 ing on difference in frequency of nerve vibration arriving in the 

 sensorium ; no such hypothesis was entertained by him either 

 for tone or for colour sensation. With sight, as with hearing, 

 he supposed that the function of frequency of undulation vir- 

 tually stops at the nerve terminals in the eye and ear, and that 

 the frequency of undulation of the physical agent has no correla- 

 tive in the quality of motion passing from the receiving terminal 

 to the sensory cell. He believes that the different frequencies 

 of ether-waves sjmply excite chemical changes in different nerve 

 terminals. He expressly states^ that the molecular commotion 

 in the nerve fibres for red, green, and violet is identical in kind, 

 and that its different effects depend on the specific activities of 

 the different cells to which it passes in the sensorium. It is 

 evident that Helmholz entirely dismissed the Newtonian theory 

 of the production of different qualities of colour sense, and sub- 

 stituted for it the doctrine of his own great teacher, Johannes 

 Midler. 



The theory of Young and Helmholz offers an explanation of 

 so many facts, and has at the same time provoked so much 

 criticism, that I must enter more fully into some of its details. 

 On this theory, the sense of white or grey is supposed to result 

 from a simultaneous and duly balanced stimulation of the red, 

 green, and violet terminals. The red terminals are supposed 

 to be excited chiefly by the longer waves in the region of the 

 red, orange, and yellow, but also by the shorter undulations 

 extending as far as Fraunhofer's line F at the beginning of the 

 blue. In like manner, the green terminals are excited chiefly 

 by the waves of medium length, and to a less extent by the 

 waves extending to C in the red, and by the shorter waves 

 extending to G in the violet. The violet terminals are stimu- 

 lated most powerfully by the shorter undulations between F and 

 G, but also by the longer ones reaching as far as D in the 

 yellow ; therefore, optically homogeneous light from any part 

 of the spectrum, except its extreme ends, does not usually give 

 rise to a pure colour sensation ; all three primary sensations are 

 present, and consequently the colour inclines towards white — 

 the more, the stronger the light. 



The experimental facts in support of Young's theory are 

 familiar to all who have studied physic-. Compound colour 

 sensations may be produced by causing light of different wave 

 lengths to fall simultaneously or in rapid succession on the same 

 part of the retina. The commonest experimental' device is to 

 rapidly whirl discs with sectors of different colours, and observe 

 the results of the mixed sensations; or to cause the images of 

 coloured wafers or papers to fall simultaneously on the retina 

 by Lambert's method ; or to transmit light through glass of 

 different colours, and cause the different rays to fall on the 

 same surface ; or to mix pure homogeneous light from different 

 parts of the spectrum. For obvious reasons, the last method 

 yields the most trustworthy results. We cannot, by any mixture 

 of homogeneous light from different parts of the spectrum, 

 obtain a pure red or green sensation, and according to Helm- 

 holz, the same holds true of violet. On the other hand, a 

 mixture of homogeneous rays from the red and green parts 

 produces orange or yellow, according to the proportions em- 

 ployed. A mixture of rays from the green and violet gives 

 rise to intermediate tints of blue, and a mixture of red and 

 violet light produces purple. Therefore, Young regarded red, 

 green, and violet as primary sensations, and orange, yellow, 

 and blue — ^just as much as purple — he regarded as secondary or 

 compound sensations. Helmholz discovered that to obtain a 

 sense of white or grey it is not necessary to mingle rays from the 

 red, green, and violet portions of the spectrum. He found that 

 he could obtain a white sensation by mixing only tivo optically 

 homogeneous rays from several parts of the right and left halves 

 of the spectrum. The pairs of spectral colours which he found 

 complementary to each other are, red and greenish-blue, orange 

 and cyan-blue, yellow and ultramarine-blue, greenish yellow and 

 violet ; the complement for pure green being found not in any 

 homogeneous light, but in purple — a mixture of red and violet. 

 The complementary colours may be arranged in a circle, with 

 the complementariss in each pair placed opposite one another. 

 Of course, the circle cannot be completed by the colours of the 



' Von Helmholz, Haudbiich der physiologischen Optik, 2nd edition, 

 1892, p. 3SC. 



