Color Council and the Tobey Color Card Co., St. Louis, Mo., the National Bureau of Standards commenced issuing the ISCC-NBS Centroid Color Charts [109] in February 1965. Each set consists of 18 charts displaying samples closely approximating 215 of the 267 centroid colors, and fair approximations of 36 others, making 251 color samples in all. The samples are in the form of one-inch squares of paper coated with glossy-finish paint affixed to a variable gray background so that each color is on a neutral background of approximately its own lightness. The Munsell renotation of each color sample is supplied in the cover pages to the charts. Duplicates of each of the 251 ISCC-NBS centroid colors can be obtained in 9 by 12 in sheets from the Munsell Color Company, Inc. of Baltimore, Md. These centroid colors have been used as standards for a wide variety of purposes. They facilitate color description at level three, and have found their most important application so far in the recording of data for establishing trends of public acceptance of color in various lines of merchandise.
.jpg)
There are many tests analogous to the comparison of a solution of an unknown amount of a constituent with a series of suitably prepared standard solutions to find the concentration of the specimen. In these tests the colors of the unknowns exhibit a one-dimensional change with concentration; and, although this change may be complicated in terms of luminous transmittance and chromaticity coordinates [102], a suitably spaced series of standards over this range of colors will yield the desired concentration either by actual match with one of the standards, or by visual interpolation among them, Such a series of standards is said to constitute a color scale. The ideal material from which to make the standards is the constituent of the unknown itself; in this way there is guaranteed not only a perfect color match at some point along the scaled but also a perfectly nonmetameric match so that variation of the illuminant or individual-observer variations are generally unimportant.
However, if the unknown is impermanent, it may become necessary to try to duplicate the desired
34
