OSA's Digital Library

Journal of the Optical Society of America

Journal of the Optical Society of America

  • Vol. 6, Iss. 6 — Nov. 1, 1922
  • pp: 527–591


L. T. TROLAND  »View Author Affiliations

JOSA, Vol. 6, Issue 6, pp. 527-591 (1922)

View Full Text Article

Acrobat PDF (4010 KB)

Browse Journals / Lookup Meetings

Browse by Journal and Year


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools


No abstract available.

L. T. TROLAND, "REPORT OF COMMITTEE ON COLORIMETRY FOR 1920–21," J. Opt. Soc. Am. 6, 527-591 (1922)

Sort:  Author  |  Journal  |  Reset


  1. The Chairman of the Committee wishes to acknowledge his indebtedness to various members of his committee as follows: to Mr. Priest for a great mass of data, helpful criticisms and the suggestions drawn from his preliminary committee report of 1919 (not published, but copy may be borrowed from Bureau of Standards Library, Cf. J. Op. Soc., 4, p. 186; 1920); to Mr. Weaver for his elaborate computations of chromatic excitation values for many different classes of stimuli as well as for the development of the data upon which these computations were based; to Mr. Jones and to Dr. Ives for data and valuable criticisms.
  2. The definition of the term color which is advocated in the present report is the result of very careful consideration and protracted debate between various members of the Committee. It is unfortunate that in common speech the word color is employed, in different contexts, with at least two different meanings which are mutually inconsistent. The most common usage of the word makes it denote visual qualities which possess hue or have a finite degree of saturation, thus excluding all members of the gray series, including black and white. The second common usage of the word color is in harmony with the one recommended in the present report and causes it to embrace all visual qualities within its meaning. This second usage is most frequently found in the interrogative mood. For example, if we ask, "What is the color of a house?" it is as legitimate an answer to say "white" or "gray" as to say "red" or "green." On the other hand, the statement "the woman wore a colored dress" evidently excludes grays from the intended meaning. Such terms as color-photography, color-blindness, etc., have a similarly restricted meaning. It is scarcely admissible in a scientific terminology to employ one term in two distinct and closely allied senses; since this will inevitably lead to confusion. Consequently, it is necessary to reject one of the common-speech meanings of the word color. A careful study of the situation shows, however, that the rejection of either meaning must result, in the beginning, in certain perplexities. If we employ color in the broader sense we not only sacrifice a well recognized distinctive term for the hue-saturation aspects of visual experience but we also seem to discard a large number of terms derived from the Greek root chroma which have been used in the same sense. On the other hand, if we define color in the restricted sense to exclude the gray series we find it necessary to exclude all considerations of brilliance from the field of colorimetry.This means that if we are asked to specify the color of a gray object we must state that it has no color, and hence lies outside of our province. Similarly, we should be compelled to affirm that certain browns are identical in color with certain yellows, oranges, and reds because they possess the same hue and saturation, although their brilliances are quite different. The necessity of reactions of this sort on the part of the scientific colorimetrician would cause serious embarrassment in practice. It seems necessary to permit a certain degree of overlapping of the provinces of colorimetry and photometry, and possibly it would be desirable to include the latter under the former as a special branch. A way out of this dilemma appears possible to the chairman if we can decide to employ the Greek root, chroma, in a different sense from the Latin root, color. There seems to be no particular etymological reason for regarding these roots as exact equivalents, and it is in line with economy of terminology to differentiate between their technical meanings. We therefore propose that the root, color, and its derivatives be employed hereafter to designate all visual qualities, including those of the gray series as well as those possessing hue and saturation. (The German equivalent, Farbe, is already used in this sense.) The root, chroma, and its derivatives, on the other hand, will be used to designate visual qualities possessing hue and saturation and excluding the gray series, with its terminal members, black and white. Such a separation of meanings is far more defensible etymologically than many distinctions which have been formally established in scientific nomenclature; for example, the distinction between physics, the general science of material properties, and physiology, the special science of vital processes, both of which terms must be considered to have the same etymological significance because of the common Greek root which they contain. In harmony with the above general recommendation the following subsidiary developments may be indicated. The word, chroma, may be substituted bodily for the word, color, when the latter is intended in the restricted sense, thus red, green, pink, lavender, etc., are chromas, while black, any gray, and white are not chromas, although all of these qualities are correctly designated as colors. This usage of the term chroma may involve a slight confusion with its use by some authorities as a synonym for saturation but it will be noted that the change involved is only a small one, being simply the substitution of a qualitative for a quantitative meaning in practically the same context. The present report recommends that chroma be not used as an equivalent of saturation. If we recognize the suggested distinction between the Greek and Latin roots it is not a contradiction in terms to speak of an "achromatic color" nor is it a tautology to refer to a "chromatic color." (cf. German: bunten Farben.) The distinction in question has the advantage of preserving all of the derivatives of the root, chroma, in their accepted meanings, and there are so many of these derivatives so firmly fixed in scientific discussion as to make it practically impossible to eliminate or to modify them. All such terms as chromatic, achromatic, chromaphore, monochromatic, dichromatic, trichromatic, photochromatic, etc., will be taken to refer to colors possessing hue and saturation or to the stimulus or organic conditions underlying the production of such colors. The root, chroma, and its derivatives provide us with a well established and hence constantly available means for differentiating between color in a restricted sense and members of the gray series, while "color" and its derivatives provide us with a means for designating both of these meanings together. Some difficulties of course arise and must be met courageously by bold changes in usage. Fortunately the cases in question are not very important. For example, "chromatics" can no longer be regarded as synonymous with "colorimetry," chromatics being strictly the science of hue and saturation coördinate with photometry, if the latter is also regarded as a sub-division of colorimetry. The term "colorless" cannot be regarded as the equivalent of "achromatic" and must be taken to indicate complete transparency as well as achromaticity in an object. This is probably already the most common meaning of the term. The equivalent of the phrase "a colored object" in the common restricted usage of the term color would be "a chromed object." The phrase "color vision" becomes redundant and must be replaced by "chromatic vision." The terms relating to "color-blindness" may need some revision but the most common forms of this disorder are already designated as partial color-blindness, a designation quite in harmony with our usage of the term color. However, "total color-blindness" would be the equivalent of "complete blindness" on this basis and hence the word achromatopia, already in use, will be necessary in this instance. It is the opinion of the Committee that the above suggestions, although necessitating a number of radical changes, involve a minimum of such changes among the possibilities which are open to us in improving the nomenclature of color science. However, the recommendations of the present report are intended to be tentative and the Committee will be glad to listen to alternative proposals and to objections to the particular form taken by the present suggestions, which represent a compromise between strongly opposed factions, all well represented in the Committee.
  3. The word sensation is used here to stand for an elementary form of experience or consciousness normally depending upon the operation of a sense organ. Although the existence of any sensation rests upon the operation of the nervous system, this should not lead us to localize it in that system. Although color is not a physical entity, it obviously exists outside of us on the surfaces of objects as we see them, such visual objects or perceptions being themselves nothing but arrangements of color areas in space. This statement, however, should not be misinterpreted to mean that the colors are physical or are located on physical objects. There is no reason for supposing that visual objects are identical or coincident with the objects of physical science.
  4. As, e.g. in the English Translation of Planck's "Theory of Heat Radiation," (74).
  5. On the definition of color as a psychological entity see: (69, 1), (73, 21–23), (94).
  6. The substitution of the word "brilliance" for the commonly used "brightness" and "luminosity" is necessitated by the fact that both of the latter terms have received technical definitions in connection with photometric measurements. It is impossible either to discard these technical definitions or to identify them with the definition here offered for the term "brilliance."
  7. Many of the authorities mentioned fail to distinguish between the subjective attribute of color, which is designated in the present report by the term saturation, and the ratio of homogeneous to total radiation in the stimulus, or the purity.
  8. The term chromaticity as applied to a color is a natural substitute for the term quality which is sometimes employed to distinguish that aspect of a color which excludes its "intensity." The use of quality in this context is undesirable on account of the more general meaning which it possesses in psychology.
  9. Cf. (37).
  10. On the use of the symbolism mµ instead of µµ see C. E. Guillame, Unités et Étalons, p. 7, Paris, 1893; also Soc. Fran. de Phys., Recueil de constants physiques, p. 1; B. S. Tech. Pap. 119, p. 7.
  11. A further type specifies the stimulus conditions for just-noticeable (or otherwise standardized) differences between the colors evoked by compared stimuli. On certain assumptions, functions of this type can be integrated to yield those of type (1).
  12. For photometric concepts, see the report of the Committee on Nomenclature and Standards of the I. E. S. published each year in the Transactions of the Illuminating Engineering Society.
  13. Cf. JOUR. OPT. Soc. of America, 4, p. 58; 1920.
  14. For discussions of the Psychological Color Solid see (92, 60–64; 61, 22–25).
  15. See (5, p. 96, Table 1 and p. 97, Fig. 3).
  16. See (4, Table 38, p. 239, and Table 34, p. 17, Columns 7 to 9). Also (17).
  17. Three notable atlases of absorption spectra are (39), (96), (58).
  18. On Spectrophotometry in general see (45).
  19. Use of color temperature as a means of color specification has been developed extensively by E. P. Hyde and his collaborators (28)
  20. It is clear that since the black body colors form a single linear series the reverse conversion will seldom be possible.

Cited By

Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.

Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited