We may, therefore, say that there are three fundamental or primary colors in the spectrum, by the admixture of which we can form all intermediate colors and white. The positions of these fundamental colors are shown by the heavy vertical lines R. G. B., in Fig. 3, and the curves show how the intermediate colors are formed. It will be seen from the diagram that the fundamental red is at about 6700, and this excites but the one sensation; but the green at 5180 not only excites the green-sensitive nerve fibril but also the red and the blue, while the blue at 4600 excites the green and red also, though but slightly. If we now take any intermediate color, such as orange at 6000, we only have to draw a vertical line through this point; the distances at which this cuts the two curves R and G give the relative amounts of red and green light required to match this particular color, and, as will be seen, this is about 70 red plus 35 green. There is another and very striking experiment, which can be performed by the aid of three projection lanterns; if these are arranged to throw three superimposed circles, the correctly colored red, green and blue-violet filters placed in them, and the three objectives fitted with iris diaphragms, it is easy to show every color of the spectrum. For instance, if we project the red disc, and cap the green and blue lenses, we have the fundamental red color sensation; now, after closing the diaphragm of the green lens to its smallest aperture, we can gradually mix green light with the red by slowly opening the green diaphragm, when the red light on the screen will be seen to turn gradually more and more orange, until, when both lenses are at full aperture, we have practically a pure yellow. Now, by gradually closing the iris of the red lens, the light will become first yellowish-green, then greenish-yellow, and finally, when the red is completely cut off, pure green of the fundamental hue. In exactly the same way, by manipulating the blue lens diaphragm, we can show the gradual transition of the pure green through all the intermediate shades of greenish-blue and blue-green to pure blue. By manipulating the green diaphragm we can show the transition from pure blue to the fundamental blue-violet, which, for the sake of brevity, is generally called blue. By mixing with this the red, we can run through the whole gamut of crimsons, magentas, pinks or purples, for all these terms are used indiscriminately to designate these colors. Finally, if all three lenses are working at their full aperture, we obtain white, assuming that all three light-sources are of equal intensity; if this is not so, the white obtained will be more or less tinged with color, which may require slight adjustment of one or another of the diaphragms.
We have here the addition of light to light, for we start out with a black screen, that is one without illumination, project red light on it, then add the green and the blue-violet, with the final product of a white screen. It will now be easy to understand how we can show a picture in colors. Let us take the very simplest example, such as pieces of glass with opaque lantern-slide binding strips pasted thereon at different angles, and we can at once grasp the formation of the final colored pattern. Assuming that we have a white circular disc illuminated by the respective fundamental colors, we place in the red Iantern a sheet of glass with a vertical stripe of opaque material. Obviously this shuts out the red light, so that we have a result such as shown in Fig. 4. At the sides all three lights are present and give us white; in the vertical stripe the red is cut out, only the green and blue-violet being present, therefore the result is a pure blue stripe on a white ground. Now let us insert in the green lantern an opaque cross with the arms at an angle of 450 to the vertical stripe; then, the green being cut out, the result is a mixture of the red and the blue-violet, which as we have seen is crimson or pink. In the center, where the image of the cross passes over the vertical stripe both the red and green are cut off and only the blue-violet shows. If we now insert a glass with a horizontal opaque stripe in the blue-violet lantern, where this cuts out this color we have the admixture of the red and green, that is, yellow, and in the center where all three strips cross one another all light is cut off and we have black. The final picture is a white ground with a blue vertical stripe, a yellow horizontal stripe and a crimson cross with a black center. If instead of using opaque strips we use lantern slides, which range in opacities from clear glass to complete blackness, and place one in each lantern, we shall so cut out the respective lights as to give us innumerable tints and shades of color as well as greys and blacks, and as the silver image gives the forms of the objects we have a picture in its natural colors. This is the theory of the additive process.
It is not always convenient to use a lantern or an instrument in which the three images can thus be seen, and we naturally desire to see the results on paper. We have then another proposition, but one which can be explained similarly. We start in this case with an illuminated white surface, and if we place a coating of red pigment, such as carmine, on this, the whole surface will be red. If a green pigment be now washed over the red, the color becomes a dirty indefinite shade, which might be called olive-brown; and if we use over this a blue-violet wash, the result will be black. It is obvious that our fundamental colors as used for the additive process are quite unsuitable for paper prints. But if on top of the red wash we place a yellow one, we obtain a more or less pure orange, and if we superimpose blue, not blue-violet, we shall obtain black, because the blue absorbs the orange made by the red and yellow. Blue on top of yellow gives us a more or less pure green, and in conjunction with red it gives us the crimsons. Therefore, the printing colors are red, yellow and blue.
The action of the individual colors ought to be clear from Fig. 2, but we can deal with it in another way. Let A (Fig. 5) represent the spectrum reflected from a red pigment, and B that of a yellow pigment: then if these spectra are superimposed the only colors reflected are shown in C from D 1/2 E to the red, and the sum of these is orange. In the formation of greens we have a similar case: let D represent the absorption of a blue pigment, and E that of a yellow, then, superimposing these, the only light common to both is the green shown inF.
From what has been said it should be clear that we obtain our colors in the case of prints, and lantern slides as well when they are colored with pigments, by subtracting colors from white light; therefore this process is generally known as the subtractive process.