A SPECTRUM COLOR-MIXER

BY C. E. FERREE AND GERTRUDE RAND

Bryn Mawr College

In the research work of our laboratory we have long felt the need of reducing color-mixing to a quantitative basis. By means of apparatus which provides a physical measure of the stimulus and of methods which permit a numerical comparison of responses to the stimuli employed, it has been possible to make in several fields a quantitative determination of selectiveness of response; to find out in an exact and definite way just what are the peculiarities of the visual responses as compared with each other and with other types of response to light-energy. The present apparatus was designed primarily that this type of investigation might be extended from the simple sensory and 'acuity' functions to the more complex function by which colors are combined. In view of a limited budget there has been need also to devise a plan sufficiently simple to fall within the mechanical capabilities of an instrument-maker of the college type and one that would not tie up too much money and equipment in a single fixed instrument serving only a limited purpose.

The apparatus to be described is in the form of a comparatively simple attachment which can be used in connection with a spectroscope of the usual type. By means of it the following combinations may be made: (a) Wave-lengths from any two parts of the spectrum with a range varying from identity to the widest possible difference. (b) Wave-lengths from three parts of the spectrum with a considerable range of choice in the colors selected. (c) Wave-lengths from four parts of the spectrum also with a considerable range of choice in the colors selected. (d) A pair of complementary colors with any other color in the spectrum, of use perhaps in colorimetry by the monochromatic method to furnish a variable achromatic component to be combined with the standard

color. In such a case not only the standard color can be specified in terms of energy, but also the achromatic component with which it is combined. And (e) a pair of complementary colors with a mixture of two other colors in varying amounts with a considerable range of selection of the colors forming the mixture; a combination serviceable in colorimetry by the two-color method. Here both the mixture used as the standard color and the variable achromatic component may be specified in terms of energy. In case, however, the instrument is used in colorimetry by the three- or the four-color methods the achromatic component, if needed, will have to be supplied in the usual way by white light reflected into the system from the surface of the prism of the spectroscope.

The device consists of two parts: the objective slits, with the mechanism for their control and adjustment, and the optical system.

The slits S mm., and of

The Objective Slits:-The slits with the mechanism for their control and adjustment are shown in Fig. 1. (Fig. 1A) are four in number of fixed breadth, variable height. Suitably spaced, they are cut in two plates P1 each of which, driven by a screw Sc1, may be made to travel transversely across the spectrum, one above the other. These plates, with their slits, serve to divide the spectrum horizontally into two equal parts. The plates are supplied with guides so placed that the lower edge of one plate is contiguous with the upper edge of the other, the joint being made light-tight by a slight right-angled overlap O1 (Fig. 1B) mm in breadth, cut into the opposite faces of the two plates. Two of the slits are cut in the upper plate and two in the lower. Each pair of slits may be made to travel from one end of the spectrum to the other in a fixed relation by means of the screw Sc, which moves the plate in which they are cut. Also the difference in distance between the slits may be made to vary over a considerable range. This is accomplished by making each of the plates in two parts, joined by a right-angled overlap O2 (Fig. 1C) cut into the opposite faces of the two halves of the plate. The difference in the distance between the two slits can thus be made to vary

by the length of the overlap. The slits are moved through this range of adjustment by a screw Sc2 one support for which is on one half of the plate, the other on the other half. Thus one of the slits may be placed in any part of the spectrum by turning the screw which moves the whole plate, and the distance of the other slit from that one may be varied by turning the screw which controls the adjustment provided by the overlap. Each of the two plates is provided with this means of moving one of the slits to any part of the spectrum and varying the distance of the other slit from the one so placed or located. If in the use of the instrument in connection with any particular problem a wider range of variability of distance between the two slits is wanted for one or both of the plates, this may be secured by using plates with a longer overlap. There are limitations, of course, to the use of an overlap for producing a variable distance between the slits. Different lengths of overlap, however, may be used for different purposes as the need arises. It would seem that most of the needs of three- or four-color mixing could be satisfied with a single overlap. In two-color mixing two of the slits are closed completely by a device to be described later, leaving one open slit in the upper and one in the lower plate. By means of the screws which move the two plates, the relation of the two slits can be made to vary from vertical continuity to the widest possible separation in the spectrum.

In order to be able to close entirely any one of the slits or to vary in continuous series the amount of light coming through, a thin narrow plate P2 (Fig. 1A) is made to travel over it, from top to bottom, by means of a screw Scз, the supports of which are mounted on the plate containing the slit and travel with the slit. These control plates move in a grooved track T1 (Fig. 1C) cut in the plate containing the slit. At the bottom of this track, at which place the plate is very thin, is cut the slit. Thus the desired thinness of edge is secured. An excellent slit may be obtained by cutting a wide opening in the bottom of the track and fastening to the back of the plate, in a suitable shallow insert, segments of a Gillette razor blade G (Fig. 1C) with the desired degree of separation

between the two edges. This method of securing a satisfactory slit has frequently been used by us. By means of these control plates the amount of any of the component lights used in the mixture may be varied from full intensity to total extinction.

Above and below the slits running in a transverse direction are two grooved pieces to form a track T2 (Fig. 1A and 1B) by means of which the thermopile needed to make the energy measurements is moved into place from either side. This track is fastened to the plate containing the four slits, and is made in sections to permit freedom of motion of the two halves of each plate. In measuring the energy of light one of the lower slits may be made continuous with one of the upper, if need be, thus utilizing the full intensity of the beam at that point.

The Optical System.-It has been our intention to provide two types of optical system, one designed to mix the colored lights on some diffusely reflecting surface, the other to focus the mixed beam on the eye. The latter of these, however, has not as yet been satisfactorily completed. The former, shown in Fig. 2, is extremely simple and is similar to that described

Color Mixing Spectroscope

Eye

Auxiliary Spectroscope

FIG. 2. Optical system