Colour vision

by Sir William de Wiveleslie Abney

CHAPTER IX.

Not only, however, may we lose a sense of colour, but we may also lose all sense of light by reducing the energy of the different rays. We have seen that colour goes unequally from the different parts of the spectrum. We may therefore prognosticate that the light itself may disappear more rapidly from some parts than from others. You will scarcely, however, I think, be prepared for the enormous difference which exists in the stages of disappearance of the grey of the reduced red and of that of the reduced green.

Fig. 27.

But how are we to measure this extinction of light at the different parts of the spectrum? This is a problem which I have attacked during the last few years by a variety of methods; but as is the case with almost every scientific problem, when the mode of attack is reduced to its simplest form, it yields the more readily to solution. If we have a box, like that figured in [Fig. 27], and combine it with our colour patch apparatus, the problem is solved. B B is a closed box 3 feet long and about 1 foot high and wide, having two similar apertures 1½ inch in diameter in the positions shown. The aperture at the side is covered on the inside by a piece of glass a, ground on both sides, and a tube T is inserted, in which diaphragms, D, of various apertures can be inserted at pleasure. The most convenient form of diaphragm is that supplied with photographic lenses—an iris diaphragm. E is a tube fitted at the end of the box through which the screen S is viewed. S is black except in the centre, where a white disc is fastened to it. A mirror, M, placed as shown, reflects the light scattered by the ground glass on to the screen S. The rotating sectors are placed where shown, and are in such a position that they can be readily adjusted by the observer. The patch of any desired colour of the spectrum is thrown on a, and an appropriate size of diaphragm used, so that when the sectors are not less than 5° to 10° open, the light totally disappears. We can now make observations throughout the whole spectrum, and knowing the value of the different apertures of the diaphragm and the angular opening of the rotating sectors, we can at once find the amount of reduction of the particular part of the spectrum that is being required in order to just extinguish all traces of light from the white disc at the end of the box. From these measures we can readily construct a curve or curves which will graphically show the reduction given to the different parts of the spectrum. [Fig. 28] gives the curve of extinction for ordinary normal colour vision. The spectrum was of such a brilliance that the intensity of the square patch of light formed on a of the orange light (D) was exactly that of an amyl-acetate lamp, placed at one foot distance from the receiving screen. Knowing this, the actual luminosity of all the other rays of the spectrum can be derived from the curve of luminosity (see [Fig. 20]). Extinguishing the various parts of the spectrum by this plan, it is found that the red rays cease to stimulate the retina sufficiently to give any appearance of light long before the green rays are extinguished. It is only the rays in the extreme violet of the spectrum, and which consequently possess very feeble luminosity, that make any approach towards requiring the same amount of reduction as the red rays.

Fig. 28.

There is the fact to remember in making these measures in the extreme red and the extreme violet, that the luminosities of the colours are so small that the illumination of the prism itself, by the white light falling on it, has to be taken into account, since it forms an appreciable portion of the patch of feeble colour. By placing a proper shade of blue or red glass in the front of the collimator slit this white light disappears or becomes negligible, and when the absorption of the coloured glass is known from measurement, we can get a very accurate measure of the extinction of these parts. Some people may propound the idea that the rotating sectors may in such kind of measurements give a false result. Now such a criticism is quite fair, and it is absolutely necessary that it should be answered. Well, to test the accuracy or the reverse of the assumption that such measures are correct, the following small piece of simple apparatus was devised. A and B ([Fig. 29]) are two mirrors placed at angles of 45° to the angle of incidence of the beam. The path the beam takes can be readily ascertained from the figure. This piece of apparatus was placed in position in front of the spectrum, and the reflected beams used to form the patches of colour. For convenience only a small pencil of light was allowed to issue from the prism, a diaphragm of some ½-inch in diameter being placed in front of it. This allows a spot of any desired colour to fall on the screen, the ground glass being removed. The slit through which the spectrum colours pass is moved along the spectrum, and a position is arrived at where the last glimmer of light disappears.

Fig. 29.

The mirrors A and B may both be of plain glass blackened with smoke on one side, or one may be plain glass and one silvered, or they both may be silvered. This, with the power possessed of altering the aperture of the slit of collimator, puts us in possession of ample means of making our measures. We may also use the ground-glass arrangement and use different diaphragms, which puts a further power of variation in our hands. I may at once state that the resulting measurements fell on the curves, obtained by measurements made with the rotating sectors, a sufficient proof that the sectors may be used with confidence. There is still another method which avoids a resort to the sectors. A tapering wedge of black glass can be moved in front of the colour slit, and a different thickness of glass will be required to cause the extinction of each colour. Recently I have modified the extinction box, more particularly for the purpose of using it where the spectrum is to be formed of a feeble light, such as that of an incandescent lamp or a candle. If a really black wedge could be obtained, this would seem to be the best method, but no glass is really black. We have, therefore, to make a preliminary study of the wedge to ascertain accurately the absorption co-efficients for the different rays, a piece of work which requires a good deal of patience, but which, when done, is always at command.

In [Fig. 28] two branches of the curves are given at the blue end of the spectrum; one is shown as the extinction for the centre of the eye, and the other of the whole eye. Of course the former observations were made by looking direct at the spot. This may appear a very easy matter, but it is not really so simple as it sounds. It is curious how little control there is over the absolute direction of the eyes when the light has almost disappeared. The axes of the eyes are often directed to quite a different point. When the extinction for the whole eye is made, the readings are really much easier, as then the eye roams where it likes, and a final disappearance is noted. When the eye has once been invested with a roving commission, it is hard to control it. In making these observations it was therefore advisable to have data for the first branch of the curve, before commencing to observe for the later. The main cause of difference between the two branches of the curve is due to the absorption by the yellow spot.

It might be thought that with the curves ([Fig. 28]) before us, we have learnt all we can regarding the extinction of light, but is it so? Surely we ought to know something as to the reduction necessary for extinction of the different parts of the spectrum when they are all of equal luminosities and of ordinary brightness.

We arrive at this by simple calculation. Supposing we have two luminosities, one double the other, it does not require much thought to find out that you have to reduce the greater luminosity twice as much as the other in order for it to be just extinguished. In other words, if we multiply the extinction by the luminosity, we get what we want. Now, in the curves before us, we have taken the luminosity of the yellow light near D as one amyl-acetate lamp, and that has a height in the curve showing the spectrum luminosity very closely approaching 100. We may, therefore, multiply the extinctions of a ray by the value of its ordinate in the luminosity curve and divide the result by 100, and this will give us the extinction of each colour, supposing it had the luminosity of an amyl-acetate lamp. A portion of the curve so calculated is shown in the same diagram ([Fig. 28]) as a dotted line. It appears at the violet end as an approximately horizontal line, and then starts rapidly upwards, and would, if carried on to the same scale, reach far out of the diagram; but at the extreme red it would be found to bend and again become horizontal. I would have you notice that the same is true not only for the extinction observed with the centre of the eye through the yellow spot, but also for the whole eye. Such straight, horizontal parts of the curve must mean something.

Fig. 30.

In the diagram ([Fig. 16]) of colour sensations we see that in each of these two regions there is but one sensation excited, viz. the violet and the red. Now, if these sensation curves mean anything, the reduction necessary to produce the extinction of the same sensation when equally stimulated should prove to be the same, for there is no reason to the contrary, but exactly the reverse. Primâ facie, then, taking the Young theory as correct, we may suppose that these horizontal parts are due to the extinction of one sensation. Let us treat it as such, and go back to the original extinction curve shown in the continuous lines. The parts of the curve which lie over the fairly horizontal dotted line, at all events, should be the extinction curve of the same sensation, but more or less stimulated or excited. As before explained, if we have double the stimulation at one part of the spectrum to that we have at another, the reduction of the greater luminosity to give extinction will be double that of the lesser. If, then, we take the reciprocals of the extinction, it ought to give us a curve which is of the form of some colour sensation; and when we arrive at the maximum, we may for convenience make that ordinate 100, and reduce the other ordinates proportionally. This has been done in [Fig. 30] in the curves C and D. For the sake of a name my colleague and myself have named such curves “persistency curves.” Perhaps some other name might be more fitting; but still a poor name is better than none at all.

When the persistency curve was scrutinized to see what might be taken as its full signification, I must confess that the result astonished us somewhat, though we ought not to have been surprised. The persistency curve C, when applied (in a Euclidean sense) to the curve of luminosity recorded for the men who had monochromatic vision, almost exactly coincided with it. In other words, by far the largest part of the extinction was due to the extinction of the sensation which in the monochromatic vision was alone excited. If this be not the case, there is something in colour vision which no theory which I am acquainted with can account for. Then, again, the persistency curve agrees with the curve of luminosity when the intensity of the spectrum is very feeble, which is another coincidence of a remarkable character which some theory should explain. [[Fig. 30] gives, besides the persistency curves, the luminosity curves of the normal eye, of monochromatic vision, and of the violet-blind; and an exaggerated curve of the difference between the normal luminosity curve and that of the violet-blind, and others which I think will be found useful for general reference.]

What sensation is it that is last extinguished, and which is possessed by a certain class of colour vision? In the Young theory it can only be the violet sensation. It is certainly not the green, and much less the red. It does not correspond, however, very well with the violet sensation shown in [Fig. 16], but more with one which should be in the blue.

In making the extinctions of light, it is quite necessary that certain precautions should be taken to avoid error. All my audience know that when going from bright daylight into a cellar, in which only a glimmer of light is admitted, but little can be seen at first, but that, as the eye “gets accustomed” to the darkness, the surroundings will begin to be seen, and after several minutes what before was blackness comes to be invested with form and detail. So it is with the extinction of light in the apparatus described. Observations carried on before the full sensibility of the eye is attained are of no value. A recorded set of observations will show this. A light of a certain character was thrown on the extinction box, to be extinguished, and the observer entered the darkened room from the full glare of daylight. The eye was placed at the eye end and kept there, and the extinctions were made one after the other till they became very fairly constant. The following is the result:—

Times of Observation.	Readings.
At the commencement	1·0
After 38 sec.	3·2
After 53 sec.	4·9
After 1 min. 11 sec.	6·9
After 1 min. 44 sec.	10·5
After 2 min. 43 sec.	17·0
After 3 min. 44 sec.	27·5
After 4 min. 52 sec.	43·0
After 5 min. 59 sec.	63·0
After 6 min. 41 sec.	78·0
After 7 min. 28 sec.	89·0
After 8 min. 32 sec.	96·0
After 10 min. 46 sec.	103·0
After 12 min.	103·0

(For convenience the first reading is unity; the other numbers are the inverse of the extinction value.)

The eye apparently, under the conditions in which these observations were made, was at least 100 times more sensitive to very faint light after twelve minutes than it was at the beginning, and that then concordant readings could be made. It will now be quite understood that before any serious measures can be made this interval must elapse, and also that the light, finding its way to the end of the box to illuminate the spot, should never be strong, otherwise the eye might lose its sensitiveness.