Page:JOSA-Vol 06-06.djvu/31

From Wikisource
Jump to navigation Jump to search
This page has been proofread, but needs to be validated.
Aug. 1922]
Colorimetry Report 1920-21
553

but the results cannot be regarded as sufficiently final to justify their adoption in place of a maximally straightforward representation of the facts of color-mixture, such as is given in Table 6.

Fig 3 provides a graphical representation of some of the relations based upon Mr. Weaver’s analysis, in terms of trilinear coördinates or a so-called color-mixture triangle. It will be noted that the triangle here given is rectangular rather than equilateral, as is ordinarily the case. The latter form bears the simplest relation to the representation of the three color excitation values in the familiar three-dimensional Cartesian coördinates, being a

Fig. 4a. Wave-Lengths of Complimentary Hues


section of this system making equal angles with all of the reference planes. The rectangular form, however, is much easier to use either in plotting the loci of colors or in determining the excitation values of the colors lying on curves already plotted. The result of the mixture of stimuli represented by two or more points in the triangle is found by computing the position of the “center of gravity” or centroid, of the multiple-point system, in which the intensities of the components—expressed in relative units defined by the triangle—function as analogues of masses.