Colour vision

by Sir William de Wiveleslie Abney

CHAPTER II.

It will be seen, then, that in measuring colour or light several circumstances have to be taken into account. These are not simple, and require differentiating one from another before the results of colour measures can be finally laid down as correct, or as being held to be applicable to all cases.

We must naturally ask, what is colour? The answer I should like to pass over entirely. It can only be described as a sensation, just as we should describe touch as a sensation. It has, however, one advantage over most sensations, in that it is a sensation which can be submitted to empyric measurement. The question whether certain phenomena, such as the colours produced by simultaneous contrast, are subjective or real, does not require answering for the purpose that we have in view, but the results recorded may probably help to throw light on it. Colour is an impression caused by the stimulation in the eye of some apparatus, that lies near the outer wall of the retina, the effect of the stimulation being conveyed by the optic nerve to the brain. If this apparatus be complicated by being made up of distinct parts, each of which transmits its own kind of impression to the brain, it is not only quite possible, but more than probable, that when one part is absent or injured the particular impression for which it is responsible will be lacking, and that the sum of the impressions due to the remainder will be unlike that perceived when they are all working together.

In every investigation, whether it be in physical or in any other branch of science, it is better to work up from the simple to the more complicated; and acting on this plan, it is better to commence experimenting with simple rather than with complex colours, though they may apparently produce precisely the same sensations. I shall, with this in view, devote most of the remaining part of this chapter to some necessary experiments with simple colours. The simple colours are those of the spectrum, and are the result of motion in the ether, which pervades all space. The motion is in the form of undulations or waves, and each colour is due to a series of these waves, which have a definite length. Thus, 6562 ten-millionths of a millimetre produces to most of us a red colour in the spectrum (see Plate I.), occupying the position indicated by a black line known as the C line in the solar spectrum.

A table of wave-lengths of certain lines in the solar spectrum is given below:—

TABLE OF WAVE-LENGTHS IN TEN-MILLIONTHS OF A MILLIMETRE.

B, deep red	6866
Lithium, cherry red	6705
C, red	6562
D, orange	5892
E, green	5269
b, green	5183
F, bluish green	4861
Lithium, blue	4603
G, violet	4307
H, extreme violet	3968

The rays in the different parts of the spectrum being due to these simple vibratory motions, cannot be decomposed further. And it makes no matter whether we see them as different colours or not, they will always issue at the same angle from the same prism (if the prism be used to form the spectrum), when it is turned to the same angle to the incident light. Milestones are useful along a road to tell us where we are in reference to some central place, and these black lines in the spectrum serve the same end. But they have the advantage over the milestone, for whilst the last will tell us how far we are from, say, York or London, the former tell us our distance from a zero point. We thus have a scale of light of different wave-lengths laid down for us, which we can apply to the study of the sensations stimulated in the eye, and so have the means of instituting a comparison between the colour vision of different eyes. A mixed or composite colour is in a different category, however, to the simple colour, as you will see directly. It is one which may be formed by any number of rays of different wave-lengths falling on the eye. What these rays are we can only tell by analysing the light and referring them to the spectrum.

The instrument before you is one which I have used before in this theatre; but as the major part of my experiments have been carried out with it, in case those who are present may not be acquainted with it, it will be necessary to describe it very briefly. The general arrangement of the apparatus is given in the accompanying diagram, [Fig. 4].

Fig. 4.

R R are rays coming from the source of light, be it sun light or the electric light, and an image of the one or the other is formed by a lens L₁ on the slit S₁ of the collimator C. The parallel rays produced by the lens L₂ are partially refracted and partially reflected. The former pass through the prisms P₁, P₂, and are focussed to form a spectrum at D by a lens L₃. D is a movable screen in which is an aperture S₂, the width of which can be varied as desired. The rays are again collected by a lens L₄, and form a white image of the surface of the last prism on the screen E. If the light passing through S₂ is alone used, the image at E is formed of practically monochromatic light. Part of the rays falling on P₁ are, as just said, reflected, but as it and the refracted part are portions of the light passing through the slit S₁, they both must vary proportionally. If then we use the reflected portion as a comparison light to the spectrum colours, the relative intensities of the two, though they may vary intrinsically, will remain the same. The rays reflected from P₁ fall on G, a silver or glass mirror, and, by means of another lens L₅, also can be caused to form a white patch on the screen E, alongside the patch of colour. At M, or anywhere in the path of the beams, an electro-motor driving a sector with apertures which can be opened or closed whilst rotating, is placed, and the illumination of either beam can be altered at will. To obtain a large spectrum on the screen E, all that is necessary is to interpose a lens of fairly short focus in front of L₄, when a spectrum of great purity and brightness can be formed.

If it be required to measure the width of the slits S₂ (which we shall see further on is often necessary), a small lens of short focal length placed behind L₄ and near the slit will cast a magnified image on E, and by means of a scale placed there, the widths of each slit, if there are more than one, can be read off on the scale by bringing them successively into the same colour.

Fig. 5.

Originally the comparison light was a candle, and it answered its purpose fairly well, and for obtaining absolute measures is convenient at the present time. [Fig. 5] will show its arrangement, but as both the candle and the electric light may vary independently of each other, it will be seen that for merely the comparison of the different spectrum colours, the previous arrangement is the better. In both cases the two beams—the direct and the comparison—may be made to cast shadows by placing a rod in their path, the shadow cast by one light is then illuminated by the other light. By moving the rod towards or from the screen the shadows can be brought side by side.

With this instrument it is easy to demonstrate that a mixed colour may be mistaken for a simple colour of the spectrum. In a glass cell with parallel sides is a solution of potassium bichromate, which, to myself and probably most of you, has a beautiful orange colour. The spectrum of white light is now on the screen, and if this orange liquid is placed in the path of the white light before it reaches the prisms, all the violet, blue, and most of the green is cut off, leaving some green-yellow, orange and red only on the screen. That these form the orange colour of the bichromate is readily shown by removing the auxiliary lens. The spectrum, which has its focus at D, is now recombined into a patch of light, which is at once seen to be the colour of the solution.

Fig. 6.

The colour of the bichromate is therefore a complex or mixed colour according to our definition, for it is made up of a large number of simple colours. What I desire to show, however, is that this complex colour can be mistaken by the eye for a simple colour. First, let us interpose the cell with the bichromate in the path of the reflected beam, and throw the patch of light formed by it on a white surface A ([Fig. 6]), alongside the patch of light B formed by the spectrum. Next let us pass a single aperture ([Fig. 7]), which can be opened and closed by a screw arrangement, through the spectrum. By careful movement we at length come to an orange ray, which is spread out by the apparatus to form a patch on B, that to the majority (and the word majority is used with intention) of people exactly matches the colour of the bichromate. Thus we have a proof that, as far as the eye is concerned, the simple and the complex colours are identical. This illustration of the want of power of the eye to analyse colour might be repeated as often as we like. We may pass coloured wools, for instance, through the length of the spectrum and show that they have the property of appearing bright in, and therefore of reflecting, some colours and of almost disappearing in others—a sure indication that these colours are mixed colours as they are made up of the rays which are reflected. Yet when viewed in white light they can in many cases be matched with simple colours in the way we matched the colour of the bichromate solution. This tells us that there is something which requires investigating as to the constitution of the perceiving apparatus, and points to the probability that it is less complicated than it would be were it able to differentiate, without the aid of the spectrum, between simple and complex colours. If the eye had a separate apparatus—and when I say apparatus I use the word for want of a better—for taking up the impression of every simple colour, it might well be assumed that a differentiation must take place.

There is one class of colours, it must be remembered, which can never be mistaken for simple colours. I refer to the purples—mixtures of red and blue—for there are no spectrum colours which unmixed can possibly match them. All other colours, as no doubt will soon be apparent, can be referred to some one spectrum colour, either in its pure state or else mixed with some variable quantity of white light. We are all familiar with the fact that there are three primary colours, and we are naturally led to consider these in the light of the experiments just made. As good a definition as any other of a primary colour is that it is a colour which cannot be formed by the mixture of any two or more colours. The original investigators in colour phenomena were the artists, and they found that neither red, nor yellow, nor blue could be formed by any mixture of pigments on their palette, but that all other colours could be made by a mixture of two or more of these three. Hence to these three were given the name of primary colours. When, however, the physicist began to work with the simple colours of the spectrum, it was speedily found that, at all events, the yellow was not a primary colour, as it could be formed by a mixture of green and red, whilst a green could not be formed by a mixture of any other two colours. This we can prove with our apparatus.

Fig. 7.

Three apertures, all of which can be opened or closed as required (see [Fig. 7]), are placed in the spectrum, one in the red, one in the green, and one in the violet. The last we shall not require at present, so it is entirely closed; but we vary the width of the other two. We find that with a little red added to a bright green, a yellow green is produced; with more red added we have yellow; with still more red, an orange. The relative brightness of the two colours mixed together can be shown by removing the lens which recombines the spectrum to form the patch of light. Each colour issues through its slit and forms its own patch on a white screen which, for the purpose, we make rather larger than usual. The two patches overlap in the middle ([Fig. 8]), and the pure colours are seen one on each side of the mixed colours.

Fig. 8.

Now, placing one slit in the yellow and another in the blue of the spectrum, we find that whatever width of slit we take, no green is produced, but that, in fact, a yellowish or a bluish white results, and that when the two slits are properly adjusted, a pure white is produced. Evidently since none of the intermediate spectrum colours between the blue and the yellow can be made by their mixture, certainly green cannot. Hence, with pure colours a green and not a yellow is one of the primaries.

Further investigation on these lines has placed the violet of the spectrum as a primary rather than the blue, but this is still a matter of debate. Suffice it to say that a red and a green in the spectrum are really two of the primary colours, and most probably the violet the third. Experiment shows that there is no other primary colour in the strict sense of the word. We thus arrive at the fact that, except the primary colours themselves, every colour in nature may be made by a mixture of two or three of these primaries.

Just a word of explanation as to why, with pigments, the primary colours appear to be red, yellow, and blue, and not red, green, and blue. The colour of a pigment, it must be recollected, is a complex one. If we analyse a yellow—a yellow glass will be just as good an example as anything else—we find it is made up of green, yellow, orange, and red. A blue is made up of blue and green. If a yellow is placed behind a blue glass, and we look at a white surface through them, the only light that can get through the glass is the green. If the light, coming through each glass separately, falls on the same spot on a white surface, it will be either colourless or bluish white, or yellowish white, whichever colour preponderates. As the light reflected from mixed pigments is made up principally by the light coming through the different particles, first coming through one and then through another, and only partially by mixed lights, it will be gathered why the primary colour, when deduced from experiments with pigments, was yellow, and not green.

With the spectrum colours there is this fact to remember, that though all intermediate colours between the pairs of primaries can be formed by their mixture, yet in some cases the resulting colours are slightly diluted with white, and that they thus appear less saturated than the spectrum colours themselves. The reason for this we shall be able to account for when we consider the colour sensations themselves.

When making matches to simple or other colours by the method of mixtures, we have to be careful of the conditions under which we experiment. This can be shown by a very simple experiment. I will make a match on B with the white light, which is thrown on the surface A ([Fig. 6]), by mixing the red, green, and violet that pass through the three adjustable apertures or slits already described. The apertures are altered till the match appears to myself perfect. From an appeal made to those of the audience who are at least 25 feet away from the patches of light, as to the correctness of the match, I gather that the match is to them imperfect. The mixed colours appear to them to give a pinkish white. The reason of this defect in the match is due to the fact that, as the lecturer is viewing the two square patches of 2 in. side from a distance of 2 ft. 6 in., their images on his retina extend beyond the boundary of the yellow spot, whilst the audience receives the whole of the image on that portion of the retina which is completely covered by it. To the lecturer only part of the blue and green is absorbed by the yellow spot, and the part of the retina outside it on which the image falls receives and records the full intensity of these colours. To the audience the full amount of absorption takes place, with the result that the patch of mixed colours must appear too red when it is correct to the lecturer. In this case habit makes the eye take an average of the different intensities which must exist at the various parts of the image. We can, however, cause a perfect agreement between all parties if the experimenter views the surfaces in a mirror placed some 12 feet away and then makes the match, for he is viewing the patches from what is practically a distance of 24 feet. If after making the match without the aid of the mirror the lecturer’s eyes are directed a little to one side of the illuminated surfaces, a match will no longer exist; the mixed colour, which is to the audience pinkish, will now appear a bluish green to him. The reason for this alteration in hue is that the whole of the images falls outside the yellow spot.

It will now be quite apparent that we must discount any assertion in regard to colour matches, unless we are told the distance of the eye from the surface on which the match is made, together with the size of that surface. This yellow spot is often provokingly tiresome in the study of colour mixtures, and one might almost be justified in doubting whether any absolutely exact matches can ever be vouched for, owing to the important region of the retina which it occupies.

The fatigue of the retina to colour after it has been presented to the eye for any length of time is a difficulty, but in a less degree. That the retina does experience fatigue can be shown by a very simple experiment. The lecture theatre is now illuminated by the incandescent light, and if we throw an image of the bright carbon points of the electric arc light on the screen and steadily fix the eyes on the image of the white-hot crater for some (say) twenty seconds, and then we suddenly withdraw it, a dark image of the points will be seen on the partially lighted screen, and will appear to travel with the eyes as they move away from the fixed point. This phenomenon is due to the fact that the perceiving apparatus for white light gets fatigued on the parts of the retina on which the bright image of the white carbon points thrown on the screen fell, and that when the source of brightness was removed, the less intense illumination of the screen failed to stimulate the vision apparatus at those parts to the same extent that they were stimulated over the rest of the field. We can vary the experiment by placing a red glass in front of the electric light, and, following the same course as before, we shall see a greenish-blue image of the carbon points upon the screen. In this case the retinal apparatus which has not been stimulated by the red sensation will be capable of the maximum stimulation by the feeble white light, whilst that part which has suffered fatigue will not respond so freely to the red contained in the white light. If we abstract a certain amount of red from the spectrum, its recombination will give a white tinged with greenish blue, which is a counterpart of the colour we feel when the eyes have been fatigued by the red light.