COLOR STANDARDS
AND
COLOR NOMENCLATURE

RIDGWAY

Fifty-Three Colored Plates

Eleven Hundred and Fifteen Named Colors

As stated in the Preface, the purpose of this work is the standardization of colors and color nomenclature, so that naturalists or others who may have occasion to write or speak of colors may do so with the certainty that there need be no question as to what particular tint, shade, or degree of grayness, of any color or hue is meant. Therefore, it is unnecessary to treat of the subject from any other point of view; it will be sufficient to say that this work is based on a thorough study of the subject from every standpoint, and that practically all authoritative works on the subject of color have been carefully consulted.[2]

Plan

Plan.—The scientific arrangement of colors in this work is based essentially on the suggestions of Professor J. H. Pillsbury for a scheme of color standards,[3] which have also been the basis of several other efforts toward the same end, as the plates in Milton Bradley's "Elementary Color" and educational colored papers, Prang's charts of standard colors, Klinkseick and Valette's "Code des Couleurs," etc.; but while all these present a scientifically arranged color-scheme and more or less adequate [2]number of colors they all fail to supply a ready or convenient means of identifying and designating the colors—the principal utility of a work of this kind. It is in the latter respect that the present work is believed to meet, more nearly than any other at least, this essential requirement, and in this consists whatever originality may be claimed for it.

The "key" to the classification or arrangement herewith presented is, of course, the solar spectrum, with its six fundamental colors and intermediate hues, augmented by the series of hues connecting violet with red, which the spectrum fails to show. If, with the red-violets and violet-reds thus added to the spectrum hues, the band forming this scale be joined end to end a circle is formed in which there is continuously a gradual change of hue, step by step, from red through orange-red and red-orange to orange; orange through yellow-orange and orange-yellow to yellow; yellow through green-yellow and yellow-green to green; green through blue-green and green-blue to blue; blue through violet-blue and blue-violet to violet; and violet through red-violet and violet-red to red—the starting-point—with intermediate connecting hues. In the solar spectrum, both prismatic and grating, but especially the former, the spaces between the adjoining distinct colors are very unequal; therefore for the present purpose an ideal scale must be constructed, so that an approximately equal number of equally distinct connecting hues shall be shown. Distinctions of hue appreciable to the normal eye are so very numerous[4] that the criterion of convenience or practicability must determine the number of segments into which the ideal chromatic scale or circle may be divided in order to best serve the purpose in view. Careful experiment seems to have [3]demonstrated that thirty-six is the practicable limit, and accordingly that number has been adopted.[5] If the number of intermediate hues were equal in all cases there would, in this scheme, be five between each two adjacent fundamental colors of the spectrum; but a greater number of recognizably distinct hues is obviously necessary in some cases than in others; for example, spectrum orange is decidedly nearer in hue to red than to yellow, and therefore the number of intermediates required on each side of the orange is different, being in the proportion of four for the red-orange series to five for the orange-yellow, and similarly six are required for the violet-red series, while four suffice for the blue-violet hues.

There is no known means by which we can measure the proportion of two or more pigments in any given mixture, "because color-effect cannot be measured by the pint of mixed paint or the ounce of dry pigment;"[6] but, fortunately, we have a very exact method, in the color-wheel and Maxwell disks, by which the relative proportions of two or more colors in any mixture may be precisely measured. This method has been used in the painting of every one of the 1115 colors of the present work, by means of one disk to represent each one of the thirty-six colors (both pure and "broken"), together with a black, a white, and a neutral gray disk, the last being a match in color to the gray resulting from the mixture of red, green and violet on the color-wheel;[7] the neutral gray disk, however, being used only for the making of disks for the broken series of colors (′, ′′, ′′′, ′′′′, and ′′′′′) and for the scale of neutral grays (Plate [4]LIII.) These colored disks are slit on one side from center to circumference, and therefore by interlocking two or more they may be adjusted so that either occupies any desired percentage of the whole area, which may be very precisely determined by a scale of 100 segments shown on the outer edge of a larger disk on which the colored disks are superimposed. When connected with the color-wheel and adjusted as may be desired, and then rapidly revolved, the two or more distinct colors resolve themselves into a single uniform composite color, whose elements are shown, in their relative proportion, by the scale surrounding the disks.[8]

The scales (both horizontal and vertical) of the present work are all prepared directly from definite color-wheel formulæ, based on carefully calculated curves; the thirty-six pure spectrum hues, represented [5]by the middle horizontal line of color-squares on Plates I-XII (together with an equal number of intermediates represented by blank spaces), requiring a separate curve and consequently different relative proportions of the two component colors for each series of hues—that is, the series from red to orange, orange to yellow, yellow to green, green to blue, blue to violet, and violet to red, respectively; but the progressive increments of white in the scales of tints, black in those of shades, and neutral gray in the several series of broken colors are exactly the same in every case. The first series of Plates (I-XII) shows the pure, full spectrum colors and intermediate hues (middle horizontal line, nos. 1-72),[9] each with its vertical scale of tints (upward, a-g) and shades (downward, h-n), the increments of white for the tints being 9.5, 22.5, and 45 per cent., respectively, those of black in the shades being 45, 70.5, and 87.5 per cent. The remaining Plates show these same thirty-six colors or hues in exactly the same order and similarly modified (vertically) by precisely the same progressive increments of white (upward) and black (downward), but all the colors are dulled by admixture of neutral gray; the first series (1′-72′, Plates XIII-XXVI) containing 32 per cent. of neutral gray, the second (1′′-72′′, Plates XXVII-XXXVIII) 58 per cent., the third (1′′′-72′′′, Plates XXXIX-XLIV) 77 per cent., and the fourth (1′′′′-72′′′′, Plates XLV-L) 90 per cent. The last three Plates (LI-LIII) show the six spectrum colors[10] (also purple, the intermediate between violet and red) still further dulled by admixture of 95.5 per cent. of neutral [6]gray, these being in reality colored grays; to which are added a scale of neutral gray and one of carbon gray, the former being the gray resulting from mixture of the three primary colors (red 32, green 42, violet 26 per cent., which in relative darkness equals black 79.5, white 20.5 per cent.); the latter being the gray produced by mixture of lamp black and Chinese white, and the scale a reproduction of that in the author's first "Nomenclature of Colors" (1886, Plate II, nos. 2-10). It should be emphasized that in all cases except the scale of carbon grays, only the disks representing the middle horizontal series of colors (both pure and broken) have been used, in combination with a black and a white disk, respectively, to make the colors of the vertical scales of tints and shades.

The coloring of a satisfactory set of disks to represent the thirty-six pure spectrum colors and hues was a matter of extreme difficulty, many hundreds having been painted and discarded before the desired result was achieved. Several serious problems were involved, the matter of change of hue through chemical reaction of the combined pigments or dyes[11] (especially the latter) being almost as troublesome as that of securing the proper degree of difference between each adjoining pair of hues. The method by which satisfactory results were finally secured was as follows: First, six disks were colored to represent each of the fundamental spectrum colors, [7]according to the author's conception of them.[12] These six disks were then placed against a suitable background (a neutral gray), in spectrum sequence, with wide intervals for the accommodation of connecting series of disks, which were then colored so as to represent an apparently even transition from one to the other. When this very difficult task had been done as well as the eye alone could judge, each intermediate was then measured on the color-wheel and the relative proportions (in percentages) of its two component colors recorded. After this had been done for all the intermediate hues each series (the red-orange, orange-yellow, yellow-green, green-blue, blue-violet, and violet-red) was taken separately and a curve constructed on cross-section paper from the recorded ratios. These curves were found to be in all cases more or less irregular or unsymmetrical, but nevertheless were sufficiently near correct to serve as a basis for a symmetrical curve; and after the points out of [8]proper line were suitably relocated the two component colors were correspondingly readjusted on the color-wheel and each faulty disk corrected (or a new one painted) until it exactly matched the required combination. The scales representing the tints and shades of each color, and also the gray or broken colors were similarly determined by corrected curves.[13]

By the method adopted of running each of the thirty-six spectrum hues through a scale of tints and shades, and repeating the combination through several series modified by increasing increments of neutral gray, practically the entire possible range of color variation is covered,[14] rendering it an easy matter to locate in the plates, either among the colors actually shown or in an intermediate space, any color which it is desired to match; and where short distinctive names have not been found (their place being, tentatively, supplied by compound names), as, necessarily, must often be the case, any color or intermediate between any two colors, either as to hue, tint, or shade, may be readily designated by the very simple system of symbols (numerals and letters) employed.[15]

In order to designate any color for which a satisfactory name cannot be found, or one not represented on the plates, it is only necessary to proceed as follows: Suppose the color in question is nearest 1 on Plate I; say, for example, is intermediate in hue between 1 (spectrum red) and 3 (scarlet-red), or in other words if represented in color its position would be in the uncolored [9]space designated as no. 2; and in tone between the full color (middle horizontal line) and tint b. Its designation, therefore, is 2a. Exactly the same method applies to any of the other blank spaces, as well as to the colors themselves, except that in case of the broken colors the "primes" (′, ′′, ′′′, ′′′′, or ′′′′′) are to be affixed to the hue number. First locate the hue, designated by number, then the tone, designated by lower case letter, the full, pure colors of the middle horizontal row being designated by number alone.

Color Names

Color Names.—While it is true that the naming of colors as usually employed has so little to do with the purely technical aspects of chromatology or color-physics that, as Von Bezold remarks[16] "we are in reality dealing with the peculiarities of language," it is equally true that a collection of color standards designed expressly for the purpose of identifying and designating particular colors can best attain this object by the use of a carefully selected nomenclature. In other words, the prime necessity is to standardize both colors and color names, by elimination of the element of "personal equation" in the matter. In no other way can agreement be reached as to the distinction between "violet" and "purple," two color names quite generally used interchangeably or synonymously but in reality belonging to quite distinct hues, or that any other color name can be definitely fixed. Various methods of handling the matter of color in zoological and botanical descriptions, etc., by the avoidance of color names and substitution therefor of symbols, numerals, or mechanical contrivances (as color-wheel and spectrum analyses, color-spheres, etc.) have been devised but all have been found impracticable or unsatisfactory. The author has taken the trouble to get an expression of opinion in this matter from many [10]naturalists and others, and the preference for color-names very greatly predominates; consequently, whenever it has been possible to find a name which seems suitable for any color in this work it has been done, leaving as few as possible unnamed, and for these some other means must be devised for their designation. (See page 8). The selection of appropriate names for the colors depicted on the Plates has been in some cases a matter of considerable difficulty. With regard to certain ones it may appear that the names adopted are not entirely satisfactory; but, to forestall such criticism, it may be explained that the purpose of these Plates is not to show the color of the particular objects or substances which the names suggest, but to provide appropriate, or at least approximately appropriate, names for the colors which it has seemed desirable to represent. In other words, certain colors are selected for illustration, for which names must be provided; and when names that are exclusively pertinent or otherwise entirely satisfactory are not at hand, they must be looked up or invented. It should also be borne in mind that almost any object or substance varies more or less in color; and that therefore if the "orange," "lemon," "chestnut" or "lilac" of the Plates does not exactly match in color the particular orange, lemon, chestnut or lilac which one may compare it with, it may (in fact does) correspond with other specimens. Without standardization, even if arbitrary, color nomenclature must, necessarily, remain in its present condition of absolute chaos. Even the standard pigments are not constant in color, practically every one of them being subject to more or less variation in hue or tone, different samples from the same manufacturer sometimes varying to the extent of several tones or hues of the present work; indeed, in every case where two or more samples of the same color have been compared [11]it has been found that no two are exactly alike, the difference often being very great. For example: Of five samples of "vandyke brown" only two are approximately similar, each of the other three being widely different, not only from one another but from the other two, one being a blackish brown, another reddish brown, the third a yellowish orange-brown. Of eleven samples of "olive" no two are closely similar, the color ranging from a shade of dull (grayish) blue-green to orange-brown, dark brownish gray, and light yellowish olive; and the same or nearly the same degree of variation is seen in absolutely every color examined, showing very clearly the utter worthlessness of color names unless fixed or standardized.

In order to obtain as many color names as possible for standardization it has been necessary to draw from all available sources. Several thousand samples of named colors have therefore been collected, and for convenience of reference and comparison gummed to card catalogue cards, with the name, source, and other data thereon. These include the colors from many standard works, among them Werner's "Nomenclature of Colours" (Syme's edition, 1821), Hay's "Nomenclature of Colours" (1846), Ridgway's "Nomenclature of Colors" (1886), Saccardo's "Chromataxia" (1891), Mathews' "Chart of Correct Colors of Flowers" (American Florist, 1891), Willson and Calkins' "Familiar Colors," Oberthur and Dauthenay's "Repertoire des Couleurs" (1905), Leidel's "Hints on Tints" (1893), "Lefévré's Matieres Colorantes Artificiales" (1896), the Standard Dictionary chart of "typical colors," the educational colored papers of Milton Bradley and Prang, and many others; and besides these practically all of the artists' oil, water, and dry colors, manufactured by Winsor and Newton, F. Schoenfeld and Co., Charles Roberson and Co., [12]George Rowney and Co., Madderton and Co., R. Ackermann and Co., Bourgeois, Binant, Chenal, Le Franc, Devoe, Raynolds, Osborne, Bradley, Hatfield and others; also the coal-tar or aniline dyes of Dr. G. Grübler & Co., Continental Color and Chemical Co., and Henry Heil Chemical Co., and the well known Diamond Dyes; chromo-lithographic inks, embroidery silks, etc., etc.

The material from which to select suitable color names was greatly augmented, almost at the last moment, from two sources, as follows: (1) A very large collection of color-samples (unfortunately mostly unnamed) collected and mounted on cards by Mr. Frederick A. Wampole, a talented young artist, to whom was delegated, by a Committee of the American Mycological Society, the task of preparing a nomenclature of colors based upon spectroscopic determinations, but which, unfortunately, the untimely death of Mr. Wampole prevented from progressing beyond the accumulation of this collection. For the use of this material I am indebted to the courtesy of Dr. Frederick V. Coville, Botanist of the U. S. Department of Agriculture, and Mr. P. L. Ricker, Assistant Botanist, Bureau of Plant Industry, in the same Department. (2) A splendid collection of colored Japanese silks, taffetas, velvets, and other dress goods, kindly sent me by Mr. C. H. Hospital, of the silk department of the firm of Woodward and Lothrop, Washington, D. C. The very large number of colors represented in this collection are all named and have afforded a considerable number of the names adopted in the present work.

For obvious reasons it has, of course, been necessary to ignore many trade names, through which the popular nomenclature of colors has become involved in really chaotic confusion rendered more confounded by the continual coinage of new names, many of them synonymous [13]and most of them vague and variable in their application. Most of them are invented, apparently without care or judgment, by the dyer or manufacturer of fabrics, and are as capricious in their meaning as in their origin; for example: Such fanciful names as "zulu," "serpent green," "baby blue," "new old rose," "London smoke," etc., and such nonsensical names as "ashes of roses" and "elephant's breath." An inspection of the sample books of manufacturers of fancy goods (such as embroidery silks and crewels, ribbons, velvets, and other dress- and upholstery-goods) is sufficient not only to illustrate the above observations, but to show also the absolute want of system or classification and the general unavailability of these trade names for adoption in a practical color nomenclature. This is very unfortunate, since many of these trade names have the merit of brevity and euphony and lack only the quality of stability.

It has been difficult for the author to decide whether the standards of his original "Nomenclature of Colors" (1886) should be retained in the present work. Some of them are admittedly wrong (indeed, certain ones are not as they were intended to be); besides, owing to the method of reproducing the originals (hand stenciling) there is considerable variation in different copies of the book, one or more reprints, necessitating new mixtures of pigments, adding to this lack of uniformity.[17] Many persons, however, have urged the retention of the old standards, on the ground that they have been used by so many zoologists and botanists in their writings during the last twenty-five years that they have become established [14]through common usage. This very important consideration has induced the author to retain such of the old standards as can be matched in the present work, even though some of them do not agree strictly with either his own or the usual conception of the colors in question. An asterisk (*) preceding a color name indicates that the name in question is adopted from the older work, the variation between different copies of the work requiring the selection, in the new one, of a color representing as nearly as possible an average of the former.

In any systematically arranged scheme, unless the number of colors shown is practically unlimited, it will, necessarily, be impossible to find represented thereon a certain proportion of colors comprised among even a very limited number selected at random, or only roughly classified. Hence many (thirty-six, or more than five per cent.) of the colors shown in the old "Nomenclature of Colors" fall into the blank intervals of the present work, being intermediate either in hue or tone, or chroma, sometimes all. It is necessary of course to provide some means for the correlation of these with the present scheme, which is done by the list on page 41, where the position of each is shown.

The question of giving representations of metallic colors in this work was at one time considered; but the idea was abandoned for the reason that these are in reality only ordinary colors reflected from a metallic or burnished surface, or appearing as if so reflected; the actual hue is precisely the same, though often changeable according to angle of impact of the light rays, and relative position of the eye, this changeableness being sometimes due to interference.[18] Colors again vary, without actual difference of hue, in regard to quality of texture or surface; that is to say, the color may be quite [15]lustreless, appearing on a dull, sometimes velvety surface, while again it may be more or less glossy, even to the degree of appearing as if varnished. To deal with these variations, however, requires simply the use of suitable adjectives. For example: To indicate a color which has no lustre or brightness, the adjective matt (or mat) may be used, in preference to dull, which implies reduction in purity or chroma; other adjectives, appropriate in special cases, being velvety, glossy, burnished metallic, matt-metallic, etc.

Color Terms

Color Terms.—No other person has presented so forcibly the urgent need for reform in popular nomenclature nor stated so clearly and concisely its shortcomings and the simple remedy, as Mr. Milton Bradley, from one of whose educational pamphlets on the subject[19] the following is quoted: "The list of words now employed to express qualities or degrees of color is very small, in fact a half dozen comprise the more common terms, and these are pressed into service on all occasions, and in such varied relations that they not only fail to express anything definite but constantly contradict themselves.... Tint, Hue and Shade are employed so loosely by the public generally, even by those people who claim to use English correctly, that neither word has a very definite meaning, although each is capable of being as accurately used as any other word in our every day vocabulary"....

Certainly one would expect that men of learning, at least, would employ the broader color terms correctly; but some of the highest authorities on color-physics habitually use them interchangeably, as if they were quite synonymous; and even the dictionaries, with few exceptions, give incorrect or "hazy" definitions of these [16]terms. It is not strictly correct to say a "dark tint" or "light shade" of any color, because a tint implies a color paler than the full color, while a shade means exactly the opposite; and to say an "orange shade (or tint) of red," a "greenish shade (or tint) of blue," a "bluish shade (or tint) of violet," etc., is an absurdity, for the term hue, which specifically and alone refers to relative position in the spectrum scale, without reference to lightness or darkness, is the only one which can correctly be used in such cases.

Indeed the standardization of color terms is almost if not quite as important, in the interest of educational progress, as that of the colors themselves and their names; therefore, to make easy a clear understanding of the specific meaning of each, the following definitions are given:—

Color.—The term of widest application, being the only one which can be used to cover the entire range of chromatic manifestation; that is to say, the spectrum colors (together with those between violet and red, not shown in the spectrum) with all their innumerable variations of luminosity, mixture, etc. In a more restricted sense, applied to the six distinct spectrum colors (red, orange, yellow, green, blue, and violet), which are sometimes distinguished as fundamental colors or spectrum colors.

Hue.—While often used interchangeably or synonymously with color, the term hue is more properly restricted by special application to those lying between any contiguous pair of spectrum colors (also between violet and purple and between purple and red); as an orange hue (not shade or tint, as so often incorrectly said) of red; a yellow hue of orange; a greenish hue of yellow; a bluish hue of green; a violet hue of blue, etc.

Tint.—Any color (pure or broken) weakened by high illumination or (in the case of pigments) by [17]admixture of white, or (in the case of dyes or washes) by excess of aqueous or other liquid medium; as, a deep, medium, light, pale or delicate (pallid) tint of red. The term cannot correctly be used in any other sense.

Shade.—Any color (pure or broken) darkened by shadow or (in the case of pigments) by admixture of black; exactly the opposite of tint; as a medium, dark, or very dark (dusky) shade of red.

Tone.—"Each step in a color scale is a tone of that color."[20] The term tone cannot, however, be properly applied to a step in the spectrum scale, in which each contiguous pair of the six distinct spectrum or "fundamental" colors are connected by hues. Hence tone[21] is exclusively applicable to the steps in a scale of a single color or hue, comprising the full color (in the center) and graduated tints and shades leading off therefrom in opposite directions; or of neutral gray similarly graduated in tone from the darkest shade to the palest tint. Each one of the colored blocks in the vertical scales of the plates in this work represents a separate tone of that color.

Scale.—A linear series of colors showing a gradual transition from one to another, or a similar series of tones of one color. The first is a chromatic scale[22] (or scale of colors and hues) and in the plates of this work is represented by each horizontal series; the second is a [18]tone scale, on the plates running vertically, growing from the full color, in the center, to a pale tint (at the top) and a dark shade (at the bottom). For clearer comprehension of these two distinct scales, each plate of this work may be compared to a sheet of woven fabric; the chromatic scale (horizontal) representing the warp, the luminosity or tone scale (vertical) the woof. A third kind of color scale is represented by adding progressive increments of neutral gray to any color. This is shown by the several series of Plates, of which the first (Plates I-XII, with colors numbered 1-71) represents each step in the spectrum scale unmixed with gray, followed by five other series in which the same colors[23] are shown dulled by gradually increasing increments of neutral gray, the first (Plates XIII-XXVI, colors 1′-71′) containing 32 per cent., the second (Plates XXVII-XXXVIII, colors 1′′-71′′) 58 per cent., the third (Plates XXXIX-XLIV, colors 1′′′-69′′′) 77 per cent., the fourth (Plates XLV-L, colors 1′′′′-69′′′′) 90 per cent., and the fifth (Plates LI-LIII, colors 1′′′′′, 15′′′′′, 23′′′′′, 35′′′′′, 49′′′′′, 59′′′′′ and 67′′′′′) 95.5 per cent. of gray, the last being in reality colored grays. Finally scales are shown (on Plate LIII) of neutral gray (in which all trace of color is wanting), and of carbon gray, a simple mixture of lamp-black and chinese white. It is not easy to find a suitable name for these scales of reduced or "broken" colors, but they may, for present convenience, be termed reduced or broken scales.

Full Color.—A color corresponding in intensity with its manifestation in the solar spectrum.

[19]Pure Color.—A color corresponding in purity with (or, in the case of material colors, closely approximating to) one of the spectrum colors.

Broken Color.—Any one of the spectrum colors or hues dulled or reduced in purity by admixture (in any proportion) of neutral gray, or varying relative proportions of both black and white; also produced by admixture of certain spectrum colors, as red with green, orange with blue, yellow with violet, etc. These broken colors are far more numerous in Nature than the pure spectrum colors, and include the almost infinite variations of brown, russet, citrine, olive, drab, etc. They are often called dull or neutral colors.

Fundamental Colors.—The six psychologically distinct colors of the solar spectrum; Red, Orange, Yellow, Green, Blue, and Violet.

Primary Colors.—Theoretically, any of the spectrum colors which cannot be made by mixture of two other colors. According to the generally accepted Young-Helmholtz theory, the primary colors are red, green, and violet: orange and yellow resulting from a mixture of red and green, and blue from a mixture of green and violet. There is considerable difference of opinion, however, as to this question, and further investigation of the subject seems to be required; at any rate, authorities fail to explain why red may be exactly reproduced (except as to the degree of luminosity) by a mixture of orange and violet, exactly as yellow results from mixture of red and green or blue from green or violet, green being, in fact, the only spectrum color that cannot be made by mixture of other colors.[24]

[20]Chroma.—Degree of freedom from white light; purity, intensity or fullness of color.

Luminosity.—Degree of brightness or clearness. The relative luminosity of the spectrum colors is as follows: [Yellow (brightest)?], orange yellow; orange; greenish-yellow, yellow-green, and green; orange-red; red and blue (equal); violet-blue, blue-violet, violet.[25]

Warm Colors.—The colors nearer the red end of the spectrum or those of longer wave-lengths (red, orange, and yellow, and connecting hues) "and combinations in which they predominate."[26]

Cool, or Cold, Colors.—The colors nearer the violet end of the spectrum or those of shorter wave-length, especially blue and green-blue. "But it is, perhaps, questionable whether green and violet may be termed either warm or cool."

Complementary Color.—"As white light is the sum of all color, if we take from white light a given color the remaining color is the complement of the given color." When any two colors or hues which when combined in proper proportion on the color-wheel produce, by rotation, neutral gray, these two colors each represent the complementary of the other.

Constants of Color.—The constants of color are numbers which measure (1) the wave-length, (2) the chroma, and (3) the luminosity.

In addition to the terms defined above there are many others, for which the reader is referred to the chapter on "Color Definitions" on pages 23-30 of Milton Bradley's excellent and most useful book "Elementary Color."

The following table shows the relative percentages, in color-wheel measurement, of the two components in each of the hues connecting adjacent pairs of the six spectrum colors as represented on the original Plates of this work; together with an equal number of exact intermediates (not shown on the Plates), the latter in lower-case type and not indicated by symbols.

All of the vertical scales in the original Plates of this work (the scale of carbon grays alone excepted) contain the following percentages by color-wheel measurement:

One of the most serious difficulties encountered in the preparation of the Plates of this work was the apparent impracticability of reproducing satisfactory shades of pure colors. This originated in the fact that there seems to be no substance (pigment, dye, or fabric) which represents a true black, all reflecting more or less of white light, and consequently producing shades which are dull [24]or broken. The difficulty is increased by the additional fact that any black pigment mixed with almost any color falls short of even the color-wheel mixture in purity of hue in the resulting shades, owing to the very considerable amount of gray in all black pigments. Chromolithography can be made to produce clearer and better shades of the pure colors, but is distinctly objectionable for the purpose of a work of this kind owing to eventual oxidation of the oil or varnish with which the pigments are combined in lithographic inks, causing a change of hue; reds becoming more orange, blues more greenish, etc., in course of time.

While the absence (in large part) of pure chromatic shades is much to be regretted, the defect is not so serious, from the standpoint of utility, as might appear at first sight; for while saturated or darkened pure colors are not uncommon in the animal, vegetable, and mineral kingdoms, more or less broken dark colors are infinitely more so; and since the latter are greatly increased in number by the defect mentioned the actual result is rather an advantage than otherwise.

It will doubtless be noticed that there is a conspicuous difference in relative darkness between shades of yellow and contiguous hues on the one hand and corresponding ones of violet and adjacent hues on the other, as if the percentage of black in each were very different. This, however, is entirely the result of difference of luminosity of the two sets of colors, that of yellow being between 7000 and 8000 while that of violet is only about 13;[28] for the percentage of black in corresponding tones of the vertical scales is precisely the same for each color throughout the chromatic scale of this work.

Every Plate in each series of broken colors (′ to ′′′′′) contains exactly the same percentage of neutral gray in each color, the relative amount increasing progressively in the several series, as shown in the following table. The percentages of white in the tints and of black in the shades of the tone scales are in all cases exactly the same as in the tone scales of pure colors.

Note.—The percentages given in the preceding tables may not in all cases be precisely those actually contained in the colors on the Plates, since absolute precision in reproduction is hardly possible. All that can be claimed is a reasonably close approximation to the ideal.

Red.—Devoe's geranium lake (dry), its orange hue neutralized by a wash of rhodamin b. (Crocein scarlet b. washed with rhodamin b. produces practically the same fine red.)

Hues between red and orange.—Crocein scarlet b. with gold orange.

Orange.—Gold orange with orange g.

Hues between orange and yellow.—Orange g. with auramin.

Yellow.—Auramin, rather dilute. (The best substitute among pigments is a fine quality of zinc yellow, as Hatfield's.)

Hues between yellow and green.—Auramin washed with light green.

Green.—Auramin (very dilute) washed with light green. (The auramin should be applied first, because it "sets" or becomes fast quickly, while the light green does not, but is largely removed by overwashes of the yellow, thus rendering it very difficult to get the desired hue.)

Hues between green and blue.—Methyl green; the same washed with light blue (Diamond Dye); for the hues nearer blue, light blue washed with Winsor and Newton's permanent blue or new blue (the least violet-hued of the artificial ultramarines).

Blue.—Light blue washed with permanent blue or new blue. (Although the color is nearer that of the artificial ultramarines named, it is useless to apply the latter first, [27]for overwashes of the light blue merely sink through and darken the color without improving the hue. A moderately saturated solution of the light blue should be applied first, and when this is dry covered with one or more rather thin washes of the permanent blue or new blue).

Hues between blue and violet.—Winsor and Newton's permanent blue and some of the more violet-hued artificial ultramarines, the hues nearer violet washed with crystal violet or gentian violet.

Violet.—Crystal violet.

Hues between violet and red.—Methyl violet 1b. washed with rhodamin b.; for hues nearer red, rhodamin b. with Devoe's geranium red (dry) or crocein scarlet b.

While more or less similar in hue to rhodamin b., several other aniline dyes, as acid fuchsin, rubin s., rosein, magenta, etc., do not combine satisfactorily with the violets, the mixture soon becoming dark or dull and none of them are quite as pure a purple or red-violet.

It is most important to remember that disks thus colored must be carefully protected from light when not in actual use and never exposed to direct sunlight. The artificial ultramarines are, of course, permanent, and so, practically, are crocein scarlet, gold orange, orange g., and auramin—that is to say, are not materially affected by the action of light except after very prolonged exposure, though the last named undergoes a change of hue; but the green and violet aniline dyes are all very evanescent, rapidly fading and eventually disappearing; light blue and rhodamin, while sensitive to light, are far less so than the greens and violets.