A source of illumination with good color properties, daylight or electric, should reveal a full range of colors, should
enable good color discrimination between objects of similar spectral reflectance, and should not distort colors. We
presently have only one recognized measure of color rendering in the lighting industry, color rendering index (CRI),
developed in the early 1960s. However, CRI should not be used alone as a predictive measure of the color rendering
properties of a light source. First, CRI is a poor predictor of color discrimination. Gamut area index (GAI), another
measure of color rendering, is consistently better at predicting performance on the Farnsworth-Munsell 100 Hue test than
is CRI. GAI is also better at predicting subjective judgments of "vividness" than CRI. On the other hand, when
measuring the ability of a light source to display colors "naturally," neither the GAI nor the CRI performs consistently.
In fact, sometimes GAI is a better predictor of "naturalness" than CRI, and sometimes the opposite is true. When GAI
and CRI are used jointly in characterizing the color rendering characteristics of a light source used for illumination, high
values on both metrics appear to ensure subjective impressions of both "naturalness" and "vividness." In general, this
two-metric system appears to be predictive of an average individual's "preference." A priori tests of this two-metric
system of color rendering were conducted, lending support to the validity of this approach for characterizing the color
rendering properties of electric light sources.