Isaac Newton, by his famous prism experiment, showed in 1666 that sunlight consists of a mixture of rays, each bent to a different degree in traversing the prism and thus falling at a different place upon the far wall. He showed that each ray was elementary in the sense that it could not be changed into a ray differently bent. Each elementary ray had a different colour, and the colour of objects depended upon the copiousness with which the various coloured rays were reflected or transmitted from the object to the eye.
Newton's conclusions, though true, met with fierce opposition. To Goethe it was absurd to assert that the mere mixing of all the rainbow colours could appear white since white is without colour. And artists had long known that it was not necessary for them to have a set of seven rainbow paints; a judicious mixture on the palette of a few bright paints — perhaps only three — was sufficient for masterpieces of natural representation.
A person's perception of everything in the world outside him depends upon three factors:
(i) the physical stimulus, such as vibrations of light or sound,
(ii) the sense organs that respond to particular stimuli in special ways, and send corresponding messages along their nerves to the brain, and
(iii) the mind that creates perception out of brain activity.
Newton analysed correctly the diversity of light rays that constitute sunlight. But he did not consider the limitations of the eye in responding selectively to these diverse rays. This was done by the physician Thomas Young in 1801, at St George's Hospital, London. He saw that human perception of fine detail implies a 'fine grain' of photoreceptors in the retina, and thought it unlikely that each 'grain' would be selectively responsive to every wavelength of light. Nor was this necessary to explain colour discrimination. Young suggested that each grain consisted of a triad of resonators each thrown into vibration by light waves. The 'green receptor' was moved chiefly by waves from the middle of the spectrum (which looks green), though neighbouring spectral waves also acted upon it less vigorously. The 'red receptor' and the 'blue receptor' respond likewise to waves near either end of the spectrum. Thus light of any composition falling upon the eye will throw these three resonators, R, G, and B, into determinate amplitudes of vibration. Their sum, R + G + B, defines the brightness, and their ratio, R : B : B, defines the colour.
This view, which is essentially what is believed today, is seen not to question Newton's physics but, by taking into account the limited discrimination of the eye, to explain the painter's experience that mixing a few paints will give the whole range of colours. Young's explanation should lead to a simple and striking result. Every colour (including white) should excite the R, G, and B receptors in a characteristic set of ratios. Consequently, a mixture of red + green + blue lights, adjusted to produce this same set of ratios, should appear white, or whatever the initial colour was. In 1854 this was systematically tested by James Clerk Maxwell (1831–79), the great physicist, while a student at Trinity College, Cambridge. He showed that every colour can be matched by a suitable mixture of red + green + blue 'primaries', although sometimes it is necessary for the experimenter to mix one of his three primaries with the colour to be matched rather than with the other primaries. This trichromacy of vision was confirmed by Hermann von Helmholtz in Heidelberg and later measured with spectral lights and great accuracy by W. D. Wright at Imperial College London and independently by W. S. Stiles at the National Physical Laboratory, London.
1. Visual pigments
2. Colour-blindness
3. Adaptation
4. Psychology
1. Visual pigments
What does light do to the photoreceptors of the retina to make them send nerve messages to the brain? Willy Kühne (1837–1900), professor of physiology at Heidelberg, observed in 1877 that a dark-adapted retina removed from a dead frog in dim light and then observed in daylight was initially pink, but bleached to pale yellow upon exposure to light. This showed that the retina contains a photosensitive pigment, i.e. one that changes its chemical constitution on exposure, as does a photographic film. This pigment, 'visual purple' or 'rhodopsin', is present in the photoreceptors called 'rods' that serve deep twilight vision, which is without colour. Therefore the pigments serving colour vision must lie not in the twilight receptors but in the daylight receptors called 'cones'. And Young was correct in supposing that these are of three types.Researchers in Cambridge were the first to measure the visual pigments in the living human eye, applying the familiar observation that if at night a cat's eye is caught in the beam of a car's headlamps it shines back with reflected light (Rushton 1952). By knowing the incident light and measuring the reflected light, it is found what light has been absorbed in the eye. And if these measurements are made before and after the visual pigment has been bleached away with strong light, the change in absorption, resulting from the change in amount of visual pigment present, is learnt.
The same measurements may be made in human eyes, though here there is a very black surface behind the retina instead of the cat's shining tapetum lucidium. Since the measuring light sent in may not be made very strong (or it will bleach away the pigment that is being measured), the great sensitivity of a photomultiplier tube is needed to measure the faint light that emerges from the eye. Using this technique, it has been possible to measure the spectral sensitivity and kinetics of bleaching and regeneration in the living human eye, first of rhodopsin, then of the red and the green cone pigments. There was never sufficient blue light reflected to measure the blue cone pigment.
This work has been confirmed and extended by Marks, Dobelle, and MacNichol at Johns Hopkins University, Baltimore. They used fresh retinas from monkeys' and human eyes removed at operation, and with superb technique measured the visual pigments in single cones. They found Young's three types of cone and specified the visual pigment in each, confirming the measurements made in Cambridge on living colour-blind subjects who possessed only one of the two cone pigments measurable by the Cambridge technique.
2. Colour-blindness
Almost all so-called colour defectives have some appreciation of colour, and generally resent being called colour blind. The common defective cannot distinguish well between red and green. This is a hereditary defect, something wrong with a gene carried on one (or rarely both) of the sex chromosomes in the female or on the single active sex chromosome in the male. If in the male the gene is missing or abnormal, colour vision will be defective, and 8 per cent of all males exhibit some defect. But in the female it needs both chromosomes to suffer the loss before her defect will show, and of course the probability of the double event is much smaller than that of the single one. In fact only 0.4 per cent of females show some abnormality. Even so, women who are abnormal in only one chromosome, though showing perfect colour vision themselves, have a 50 per cent chance of transmitting their weakness to their children; and half their sons will be 'colour-blind', since the (normal) father holds his normal gene on the one sex chromosome that will make his child a daughter, and he has none for his sons.Dichromacy. In the extreme conditions of the red–green defect, the subject cannot tell red from green and can match every colour of the rainbow exactly with a suitable mixture of only two coloured lights, for example red and blue. Such subjects are called dichromats, to distinguish them from ordinary people (trichromats) who need also a green primary if every colour is to be matched.The cone pigments in the red–green spectral range of dichromats have been measured by a reflection technique. It has been shown that instead of the red- and green-sensitive cone pigments of normal vision, these subjects have only the red or the green. They lack one dimension of colour vision because they lack one kind of cone pigment.Anomalous trichromacy. The majority of colour defectives are not true dichromats, but anomalous trichromats: like normal subjects, they require three variables in a colour-matching experiment, but the matches they make are different from those of the normal. Thus, in matching a monochromatic yellow with a mixture of red and green, some ('protoanomalous') observers require more red than normal in the match, whereas other ('deuteranomalous') observers need more green. Usually, though not necessarily, the abnormality of matching is associated with a reduced capacity for discriminating colours. It is thought that anomalous trichromacy arises when one of the cone pigments is displaced from its normal position in the spectrum.
3. Adaptation
Newton's physics of colour was inadequate because it did not take into account the selective physiological action of Young's three cone types. We are now on the way to understanding their selectivity and the sort of nerve signals they generate. But though light waves and nerve signals are factors that lead to colour vision, there is still the miracle of how some nerve signals generate a sensation in the mind. This sensation certainly does not depend exclusively upon the R : G : B excitation ratios of the three cone types. We all know how adaptation to any strong-coloured light leaves the eye, as it were, fatigued to the colour so that some extra light of this colour must be added to any presentation if its appearance is to be the same as it was before adaptation. This adaptation is often called 'successive contrast' to distinguish it from the rather similar 'simultaneous contrast', where two different-coloured objects seen close together have their differences enhanced through their proximity. Some of these effects can be objectively measured by recording from nerves in the visual pathways of animals.4. Psychology
Colours are so gay that those with total colour loss cannot but be pitied, and it must be wondered what it is that makes red produce the wonderful red sensation most people perceive. What has been said here explains only what cannot be discriminated, and nothing has been said about how sensations arise from what is seen. Let it be concluded that Newton ended his first paper with these strong words: 'But to determine ... by what modes or actions light produceth in our minds the phantasms of colours is not so easie. And I shall not mingle conjectures with certainties.'(Published 1987)
— William Rushton
- Bibliography
- Boynton, R. M. (1979). Human Colour Vision.
- Brindley, G. S. (1970). Physiology of the Retina and the Visual Pathway (2nd edn.).
- Rushton, W. A. H. (1952). 'Apparatus for analysing the light reflected from the eye of a cat'. Journal of Physiology, 117.
- — — (1975). 'Visual pigments and colour blindness'. Scientific American, 232.
