Summary
Charles Gustav Zander, chief of color printing at the Scottish ink manufacturer, A. B. Fleming & Co., Ltd., rose to prominence in 1896 as an authority on three-color printing after publishing his book, “Photo-Trichromatic Printing.” Zander promoted a scientific approach to three color printing inspired by the Young-Helmholtz theory of color vision and James Clerk Maxwell’s experiments in color photography. The “color scientists” of the day — Frederick Ives, Howard Farmer, and Arthur von Hübl, to name a few — agreed that the transmittance of three-color photographic separation filters, and the spectral absorbance of their corresponding printing inks, should be determined scientifically. However, it was not possible to create a system that adhered perfectly to the scientific ideals with the materials available at that time. Thus, compromises had to be made. Ives promoted the manufacture of precise photographic filters (to ensure an accurate record of the subject), followed by the manufacture ink to match, as best as possible, the respective filters. Zander promoted the manufacture of precise printing inks (to reduce inventory of ink manufacturers and printers), followed by the design of photographic filters to match the printing inks. Nevertheless, both camps agreed that three-color printing was the future (some experimentation with black and gray key plates was taking place in Germany, but was not widely used).
Another debate taking place was between those who believed in the Young-Helmholtz trichromatic color vision theory (red, green, and blue sensors in the eye), and those who believed in Ewald Hering’s color opponency theory (red/green, blue/yellow, and light/dark neural sensing channels). Maxwell’s early photographic demonstrations were inspired by Helmholtz’s work as were proponents of three-color printing. Hering himself did not dispute the fact that we had color-selective sensors in the eye, but he observed other phenomena that suggested neural color-opponency was at play.
Technical developments in three-color printing were guided by the idea that the full spectrum of colors could be produced using three-inks. However, Zander realized that the print gamut (a term not used in his day) was much smaller than he would like due to material limitations. Furthermore, neutrals were difficult to print consistently since they required three separate plates. Thus, in 1905, Zander made a surprising about-face and abandoned three-color printing. Inspired by Hering’s color-opponency theory, Zander created and promoted his Complementary Color Process, which included red, yellow, blue, and green inks. Among its benefits were the ability to print a wider color gamut (including bright greens and purples) than three-color printing and the ability to print neutrals using only two complementary inks (red and green or blue and yellow), as opposed to three inks. Unfortunately, the prospect of having to produce and manage an additional plate, all while learning a completely new color management workflow, did not sit well with Zander’s contemporaries. Furthermore, it seems Zander was not able to produce any samples that proved, without a doubt, that his process was any better than common three-color printing. He spent the next two years on the defensive and, by 1908, had retreated from the public discourse.
Zander’s Baby: Red, Yellow, Blue (and Green?)
Charles Gustav Zander was remarkable not just because of his important contributions to discourse on three-color color printing, but because, in the face of criticism, he had the courage to change his mind. An ardent evangelist of three-color printing and the use of scientifically rigorous color reproduction techniques, he realized, after more than 20 years in the industry, that he was wrong. He and his peers had long felt that printers could reproduce the colors of the natural world with the proper set of three-color inks and plates. The Young-Helmholtz theory of color vision stated that the eye had three-color receptors. Zander and his contemporaries felt that if humans could perceive all of the colors of the visual spectrum with three color receptors, then naturally those same colors could be reproduced using three-color photographic and printing techniques. The precise method for accomplishing this, however, was hotly debated.
Zander, chief of color printing at the Scottish ink company, A.B. Fleming & Co.,1 was a well regarded figure among those involved in the debate. A British subject born in Austria in 1861,2 Zander was educated in chemistry, before settling in England in 1880. He began working for A.B. Fleming in 1893 as a specialist in color ink formulation. He then rose from public anonymity to a position of authority in his field after publishing his book, “Photo-Trichromatic Printing” in 1896,3 in which he described a scientific approach to three-color printing with an explanation of human color vision and color theory. The image below, from Photo-Trichromatic Printing, shows a reproduction of the visible spectrum using three-color printing.
Color reproduction was driven by the public’s demand to hold works of fine art in their hands. Until the late 1800s, chromo-lithography was the primary means of reproducing art for the masses, but even chromo-lithographs were priced above what much of the public could afford, and the time intensive process made them unsuitable for mass-media. Then, halftone printing took off in the 1880s and three-color photography was rapidly maturing. The next step in this natural progression was the industrialization of color printing, the ability to produce full-color reproductions of art and the natural world at a price affordable to the common consumer.
The science of color reproduction is multifaceted, requiring an understanding of human color perception; the physics of light and optics; the chemistry of photography, pigments, and printing inks; and the mechanics of producing letterpress halftone blocks. Yet, the scientific community was only beginning to understand the color science of these various disciplines.
In 1861, James Clerk Maxwell showed that color photography was possible by projecting an image of a tartan ribbon in a lecture to the Royal Institution, shown below. His technique was to first capture three separate negative images of the ribbon through red, green, and blue filters, then reconstruct the color image by projecting the negatives through filters of the same three colors. The demonstration meshed well with the Young-Helmholtz theory of color vision.
Maxwell went on to derive a series of theoretically optimal red, green, and blue filter transmission curves needed to accurately photograph the visual spectrum using photographic technology available in his time, shown below.*
The scientists who followed Maxwell — the Frenchmen Charles Cros and Ducos du Hauron, Dr. Vogel of Berlin and his son Dr. E. Vogel, Dr. Eder of Vienna, and Frederick Ives of the U.S. — developed improved plates and filter technology for three color reproduction. Maxwell’s curves served as a baseline for developing the filter responses. By photographing objects through red, green, and blue filters,† negatives were created and used to make halftone printing plates. The absorption spectrum of the printing ink used with each plate had to be complementary to the absorption spectrum of the filter: a red filter absorbing green and blue should be paired with an ink absorbing red; a green filter absorbing red and blue should be paired with an ink absorbing green; and a blue filter absorbing red and green should be paired with an ink absorbing blue. The general hues of photographic filters were red, green, and blue, and the hues of three-color inks were a greenish blue, a purplish red, and yellow. As described by Zander,
Thus, taking the red negative, the positive made from it will be printed with a colour reflecting all the spectrum rays except red, which is cyan blue (a kind of peacock blue). Taking the green negative, the positive made from it will be printed in a colour reflecting all the spectrum rays except the green recorded by the negative, this positive or printing color benign ink (magenta). Likewise, the positive made from the blue violet negative will be printed in the colour reflecting all the spectrum rays except the blue-violet, which is yellow of the hue of the D line of the spectrum (primrose).4
This workflow is illustrated in the flow-chart below.
An object, such as a painting, is photographed three times, each time through a red, green, or blue filter. Sometimes the halftone screen was combined with the color filters. Other times, a second negative of each color was created by exposing the first negative through a halftone screen. Either way, the end result was a halftone negative for each filter. The halftone negatives were used to create halftone plates. The tone reproduction curve on each plate was adjusted by a process of mechanical modification and chemical etching referred to as “fine-etching.” The red, yellow, and blue inks were selected for each plate (perhaps after some trial and error in the proofing process), and the plates were loaded on the printing press, as part of the block that included text and other graphics. Only one color was printed at a time, allowing the sheet to dry between printings. The final print was produced after the third printing.
There was widespread agreement that the printing inks should be of colors complementary to the photographic filters. However, it was difficult to manufacture filters and inks that met the theoretical requirements of Maxwell’s curves, due to variability in photographic plate sensitivity and a still elementary understanding of human color vision. Thus, practitioners of three-color reproduction had to make compromises, and the specific nature of these compromises sparked fierce debate. Some practitioners argued that inks should be manufactured with scientific precision, then plates made to match the inks. Other practitioners argued that plates should be made first, using scientifically precise photographic filters and plates, with inks manufactured to match the plates.
Zander was firmly in the camp of scientifically derived inks, each complementary to the Young-Helmholtz primaries of color vision.5 His friend, Dr. Reginald Clay, sought to determine the precise formulations and absorption transitions (between high and low absorption) for theoretical inks. He concluded that the inks should absorb in either the red, green or violet parts of the spectrum, and that the luminosity of the colors should be as high as possible when matching the visible spectrum, since monochromatic light is quite dim.4
Zander and Clay found that the absorption of the theoretical inks was too low, causing the overprinted inks to have a dirty appearance (e.g. the green overprint appeared to have some black mixed in). The solution was to modify the filter transmittance curves by increasing overall transmittance of white light, as indicated by the small humps in the absorption regions of each filter in the plot below. As Zander described:
I found in practical work that in order to produce violet, emerald green and deep rich orange and scarlet reds by the superposition [overprint] of the theoretical minus-colours [subtractive primaries], which should in purity and luminosity equal the minus-colors themselves, it was better not to print both the minus-colors full strength. If printed full strength the resultant violet, emerald green, orange or scarlet is very dull of “dirty;” whilst if one of the two is printed a little more than half-strength, these mixed hues…are more like the unmixed printing colors in purity and luminosity.4
Neglecting to use the modified filters would require practitioners to adjust the plate tone reproduction curves with fine-etching, a labor intensive, less scientific process they were trying to avoid. The common block-making method of the time was to make photographic reproductions using a somewhat arbitrary set of red, green, and blue sensitive plates, and then tailor the blocks to them appropriately using fine etching.
Frederick Ives, the eminent innovator of both color photography and printing technologies, was strongly in the camp of deriving scientifically accurate negatives and blocks, then tailoring the inks to fall in line. Ives felt that photographic filters should adhere strictly to Maxwell’s curves, thus enabling accurate reproduction of the spectrum. These ideas were at odds with common thought of the day. According to Howard Farmer, Ives’ contemporary and ideological opposite, “Mr. Ives’s methods, in addition to being practically useless, are fundamentally and scientifically bad.” 6According to Ives, as quoted by Farmer from an Ives lecture:
We must employ colour-sensitive photography plates, and filter the spectrum rays through suitable coloured mediums, testing by exposures in the photospectrograph and modifying the colour filters until the density curves conform to Maxwell’s colour curves. If you can obtain by photography such a colour record of the spectrum, you can by the same means obtain true colour records in photographs from nature and works of art; but if the spectrum test in any way fails, no real accuracy can be guaranteed or should be expected.6
This statement from Ives illustrates a common idea of the time that accurate reproduction of the natural world was predicated on the accurate reproduction of the visible spectrum. Similar to the conclusions of Zander and Clay, Farmer described how the absorption bands of the Maxwell curves were too low and resulted in the introduction of black to the spectrum. Overlapping absorption regions of inks complementary to Maxwell’s curves would result in some areas of the spectrum with no light at all. Farmer’s feelings about Ives’ recommendations were evident in writing, as in one case when he referred to Ives’ “obstinate adhesion to Maxwell’s curves.”6
Countering Ives, Farmer described a theoretical case in which one was required to photograph a variety of colored pigments, rather than the optical spectrum. The resulting negatives, he postulates, would have three distinct, non-overlapping, transmittance curves, rather than completely overlapping curves proposed by Ives. In retrospect, neither Ives nor Farmer was 100% correct, but the debate between the two clearly illustrates the passionate devotion of these scientists to the development of three-color reproduction.
As mentioned previously, Zander felt that inks were the limiting factor of three-color printing, not the negatives, so it would be best to get as close as possible to the theoretically optimal curves with the inks, and then tailor the filters from there. In addition to adhering to these theoretical guidelines for ink color, inks were also required to be permanent (light-fast) and transparent. Pigments that could be used to formulate inks with these properties had yet to be invented. Only aniline colorants were capable of producing inks that got close, but they were notoriously “fugitive,” meaning they were prone to fading after exposure to light and under certain chemical conditions. Fugitive inks could be used for scientific demonstrations, but would never suffice for commercial use.
The transparency requirement was most essential for red and blue inks, the second and third printed inks in three-color printing. The most suitable pigments were transparent in two-thirds of the spectrum and opaque in the other third, with an “abruptly terminating absorption band,” in the part of the spectrum transmitted by the photographic negative.4 Absorption bands, when superimposed, should cover the whole spectrum to allow printing of blacks. Broad-band inks were required since narrow-band inks would be too dark and not produce intermediate hues. They should also not be too broad. Clay described how, for the reproduction of a monochrome subject using a three-color process, the resulting image should also be monochrome, “but if filters with graduate absorptions were used, the resulting picture would have highlights one colour, and shadows another; with filters with abrupt absorptions the monochrome would be reproduced as monochrome.”7
The common yellow inks were all opaque, and thus printed it first.‡ Zander recommended making the red printing ink with a madder lake, notable for its transparency and permanence. For blue, he recommended cyanide blue, transparent and permanent enough.3 For yellow, Zander recommended a permanent yellow lake, preferred over chrome yellows that dried shiny and hard, making it difficult for over-printed reds and blues to adhere to its surface. 8
Commercial inks were simpler to print due to lower luminosity and steeper absorption bands. When negatives were well matched to the inks, small variations in the amount of ink did not greatly affect absorption, making the print less susceptible to process variability. However, commercial reds did not have enough blue and commercial blues were too opaque in the blue and too dark in the green. Bright purples and violets were thus difficult to produce due to the incorrect reds and blues. These deficiencies resulted in a smaller gamut using commercial inks than with theoretical inks. This trade-off between a larger gamut and ease of printing is still present today, especially with regard to the use of Expanded Gamut inks.
Zander demonstrated the differences between the gamut of commercial inks and theoretical inks in the image below, reproduced in Penrose’s Pictorial Annual in 1901. Notice the bright magenta, green, and blue possible using the theoretical inks on the right hue scale, compared to the rather dull colors produced by the commercial inks on the left.
Lack of standardization was one of the biggest issues belying three-color printing, especially in Britain. Zander claimed that the progress of three-color printing in Britain was behind that of continental Europe and the U.S. A lack of agreement between inks and printing blocks forced many printers to mediate the differences themselves. As Zander recalled,
I asked a firm of printers not long ago, why they had given up three-colour work. The answer was “We had cause to be dissatisfied with the firm who supplied us with three-colour blocks and we tried another firm. On proving their blocks we could get no satisfactory results, and were told by them that the inks we were using then, and which produced good results with other blocks, were no good for their blocks.”8
Ink manufacturers had to keep a catalog of many shades of process colors to meet the demand of printers and block makers. Zander noted that the process yellows in the specimen book of one ink manufacturer ranged in hue from greenish-yellow to orange-yellow. The lack of standardization led to a back-and-forth between printers and plate-makers. American and German block makers seemed to adhere to a greater degree of standardization, as Zander described,
In America they manage things better as far as their colour scheme is concerned. I believe that the blocks of nearly all the American block-makers can be printed with one set of three-colour inks. They are by no means scientifically correct, nor is their red as permanent as it ought to be for commercial work, particularly show cards, but they are well selected and produce, in conjunction with the superior quality of the blocks, very excellent work.8
Printing techniques varied greatly from printer-to-printer. Zander observed that, even with the same ink, substrate, and blocks, no printer would produce the same image. Each printer manages ink differently, some printing heavy, others light; some more clean than others; some using different additives (like machine oil).7 The only way to overcome this variability would be through standardization of the photographic process, the block-making process, and the inks.
Zander advocated for the standardization of ink above all else. Reducing the ink inventory would help reduce process variability and overhead. The burden of matching blocks and ink sets would then be placed on the block makers. On this point, Zander’s contemporaries did not entirely agree. A Mr. Bull, who attended one of Zander’s lectures, recalled one experiment in which three-color ink sets from a variety of ink makers were selected to reproduce the same subject, and from them, blocks were created to match the ink sets. He found that colors could be reproduced accurately in some cases, but in others the reproductions failed. Thus, he reasoned that the negatives should be made first, thereby starting with an accurate record of the subject, and finding suitable inks later, with fine-etching filling in the gaps. Bull felt that standardization of inks was a good idea, but that it must be decided between the ink makers and the printers, since there was much competition between ink makers and they would produce whatever was demanded of them by the printers. The ink-makers themselves relied upon a consistent colorant supply. Variations in the pigments received made it difficult to produce colored inks with any consistency.
Other factors that influenced process variability included: the effect of changes in temperature and humidity on the inks; the inconsistent use of varnish, a key printing ink additive; the variable density of colored filter-glass; and variations in substrates. About varnish it was noted that “although ink manufacturers might do all that was required in matching the colours, they often failed to introduce the proper proportion of varnish.” If these nuisance factors could not be controlled, then process standardization would have little effect.
Those involved in color matching were also aware that proper ink formulation could not be achieved using visual color matching. As one printer stated, “the usual method with ink-makers when trying to match an ink, was to cut a hold in the block-maker’s proof and watch the effect of the colour through it. This was not a satisfactory method, because the eye could be so easily deceived and no two persons’ eyes could see colour exactly alike.”7 The use of color measurement instrumentation was understood as an important element in standardizing ink production. This point, interestingly, foreshadowed the development of electronic densitometers and spectrophotometers that would be developed over the next few decades.
In general, despite the many scientific arguments around the proper filters and ink, practical application of three-color techniques required a printer of great skill. Most commercial printers of the time did not have the technical skill to manage a three-color process. Standardization was the only way mass commercialization of three-color printing could take flight.
In 1907, some suggested that the Royal Photographic Society should create a committee to investigate standardizing three-color work. Standardization would require three-color block makers to change common commercial practices without a guarantee of increased profits. They would have to undertake investment in experimenting and account for a temporary increase in waste and reduction in quality until they became well practiced. Standardization would only work if there was agreement between block makers and ink makers on the most appropriate set of inks that can be accommodated by both disciplines.9
Until 1904, Zander was one of the most prominent supporters and lecturers on three-color printing techniques. However, around this time, he began expressing doubts about its general applicability. The gold standard for color reproduction was chromo-lithography. As he mentioned in a 1904 lecture,
We often hear it asserted that three-colour printing can reproduce anything in three printings, and will in time entirely supersede chromo-lithography with its numerous printings. I have a different opinion on that subject. The principal reason is the prominent fact that whilst the range of colours in three-colour work is strictly limited, it is absolutely unlimited in chromo-lithography.7
Zander was frustrated by the small gamut afforded by three-color reproductions and use of three-colors to produce monochromes. He felt the industry could do better.
By 1905, Zander had reached his breaking point, completely discouraged by the lack of progress made in three-color printing over the previous 15 years. “Three-colour work has proved itself an utter failure,”10 he wrote. Art reproductions using three-color printing were not satisfactory and “even with the best three-colour blocks under the most favorable conditions are disappointing, and no judge of art would class a three-colour print amongst artistic reproductions — they find no place in our best print and picture shops.” Pure greens could not be produced, but appeared as “an offensive juxtaposition of blue and yellow dots.” Blue skies, brilliant reds, violets, pure blacks and neural grays, were all difficult to achieve with three colors.
Some progress, though, was made in that 15-year period. Panchromatic dry plates with emulsions sensitive to the entire visible spectrum were developed and half-tone blocks had improved in their rendering of fine detail and gradations. Yet, the fine-etcher was still responsible for color corrections on the plate.
Because of their belief in the Young-Helmholtz theory of color vision, many scientists, including Zander, assumed that the entire world of color could be reproduced using pairs of three primaries: humans see the world with three color receptors in the eye, color photographs can be made with three filters, and color prints made with three inks. Zander even proposed that most custom ink formulations could be achieved using no more than three colorants.11,§
Zander’s disappointment is clear in his writings, but unlike his many of his peers, he would not rest on his laurels. Thus, he looked for an alternative. Writing in the third person, he stated,
It was with great reluctance that, failing everything else, [Zander] at last set himself to find our whether the Young-Helmholtz Theory of Colour Vision on which the three-colour process o based, was really the best basis for photo-mechanical colour reproduction, or whether some other colour scheme worked our, would produce better results.10
Zander’s proposal for an alternative to the three-color process was first articulated in his patent, “Process of Photomechanical Reproduction of Colors and The Resultant Article,” applied for on February 6, 1905, and granted April 7, 1908 (U.S. 884, 254).12 As he wrote in 1905, this new “process of photo-mechanical colour reproduction assumes and uses not three but four fundamental colours, viz.: red, yellow, green and blue, by mixtures of which in suitable proportions any colours in nature can be matched or reproduced.”10 This combination of four colors was based on the “assumption that there are four fundamental or mono-chromatic constituent colors of a continuous white light spectrum, by mixtures of which in suitable proportions any color in nature can be reproduced.” The inspiration for this radical new idea of color reproduction came from Ewald Hering’s color-opponency theory of color vision.
The color science community (to use a general term for people that studied color and its reproduction) was divided into two camps: those that believed in the Young-Helmholtz theory of color vision — that color perception was based on the signal from these three cells — and the Hering color-opponency theory — that there were neurons sensitive to either red or green; either blue or yellow; and either light or dark (see image below). These neurons, for example, could give a positive electrical response when stimulated by red, or a negative electrical response when stimulated by green. This type of bipolar nervous response could not be verified in Hering’s time, and that lack of hard proof caused many to be sceptical of Hering’s theory.
As Hurvich and Jameson noted in their introduction to their English translation of Herings “Outlines of a Theory of the Light Sense”
Physiologists struggling to make their discipline as rigorously “scientific” as physics and chemistry, which at that time still emphasized simple independent and dependent variables, were much more comfortable with concepts implying only simple, elemental stimulus-response processes. When the observed phenomena belied such simplification, the implied complexity was referred to intellectual processes of judgment and interpretation…The notion that physiological excitation in one are could bring about an increase or decrease in excitation in an adjacent area was simply too far ahead of the data at hand and the prevailing atmosphere of thought.13
According to Hurvich and Jameson, Hering understood that there must be some cells in the retina that generated signals by means of a photochemical reaction with light, and that it was the input from these cells that fed the opponent neural pathways he proposed. Unfortunately, many of Hering’s contemporaries misinterpreted his writing and understood his theory to be proposing that opponent cells were activated photochemically in the retina, like Young-Holmholtz’s cone cells. ¶
Zander reasoned that Hering’s theory was the correct visual model, over Young-Helmholtz.14 He understood that the color selectivity of the cones, key to the Young-Helmholtz theory, had also not been proven at that point, but that scientists were only aware of rods and cones. Young-Helmholtz, however, was the more reasonable theory by the logic of the day.
It is possible to interpret from Zander’s writings that he was among those who believed Hering’s theory proposed separate sensors in the eye for red/green, blue/yellow, and light dark opponent channels. In his aptly named “Zander Complementary Colour Reproduction Scheme,” Zander proposed the following four hues for his primaries: magenta red, lemon yellow, Emerald Green, and Ultramarine Blue. The system is based on the idea that complementary pigments– red and green, yellow and blue — will produce a neutral when over-printed.He listed several advantages over three-color work:14
- The four-color system expands the gamut, capable of producing magenta (a primary) and purples, emerald green (a primary), ultramarine blue (a primary), and violet.
- Dense, neutral blacks can be made with only two colors, and tinted by adding a third.
- Process variability will be improved by reducing the number of colors required to produce neutrals. He argues, citing the work of chromo-lithographers, that increasing the number of plates reduces variation.
- The four-color process will produce a more accurate reproduction.
- The four-color process reduces the amount of fine-etching required.
However, given the many process variables of the day, the success of his four-color process would require strict standardization of photographic and printing processes; otherwise, significant fine-etching would be required.
In addition to his three-color photographic demonstration, Maxwell was also noted for his use of spinning disks with superimposed color rings, presently called Maxwell’s disks, to study color mixing. Zander stated, “It is invariably stated in even quite recent text-books on colour that the colours of rotating colour-disks (often called maxwell’s Disks) mix or blend precisely the same as pigments do, if mixed on a palette.”14
Zander looked to the spinning disks as proof of the four-color theory’s viability. He noted that a spinning disk of chrome yellow and ultramarine blue made a tinted neutral rather than the green expected from additive color mixing. Pigment mixtures of these colors also produce a neutral. Similarly, he described how the combination of red and green on a spinning disk produced a greenish neutral rather than yellow and that this effect was similar to the color produced by mixing those same two pigments. Zander stated generally, “I have always found that the colours on rotating disks do not produce the effect of the optical combination of coloured lights, but exactly the effect of the mechanical mixture of pigments on a palette.”
In making this claim, Zander exemplified the continuing quest of his contemporaries to understand the nature of color mixing and color vision. In this case, Zander was neither correct nor incorrect. It is true that spinning disks do not produce the same sensation of mixing colored lights. However, it seems he did not understand that spinning disks are examples of a different type of additive mixing, called “partitive” mixing, or “additive-averaging.” Essentially, the perceived color of the spinning disk is an average of the individual colors. This is why a black and a white look gray when mixed on a disk, rather than white, as the mixture would appear in a strictly additive system. The effect of mixing the pigments of complementary colors (the effect of absorption and scattering of the pigments) is more similar to partitive mixing than to additive mixing since it produces darker neutrals. Subtractive mixing by way of overprints has a similar visual effect as pigment mixtures, but the physics of partitive mixing, pigment mixing, and subtractive mixing, is different. Zander’s observations of spinning disks likely influenced his conclusions and the design of his four-color system.
In describing his four-color color system, Zander sought to redefine the term “primary color.” He felt the theory of primaries should not be based on the spectrum (i.e. the three primaries of Young-Helmholtz are within the spectrum), since we can perceive pure colors that are admixtures of two ends of the spectrum, namely magenta
Zander’s new definition of primary colors was as follows: “A primary colour should mean a hue which cannot be produced by either optical combination or by mechanical (pigmentary) mixture of two other hues.” This definition applied equally to light mixtures as to pigment mixtures, and “does away at once with the confusion now prevailing as to whether ‘primary colour’ means a primary light-sensation, according to the Young-Helmholtz theory, or the so-called primary colours of the artists.” If we ascribe Zander’s thinking to those of his contemporaries, then this logic is a chief flaw of the contemporary thinking. Defining primary colors as applicable to only additive or subtractive mixing does not leave the possibility that both could exist.
The optics of pigmentary mixing is more complicated than simple addition or subtraction. In fact, there is an entire branch of physics devoted to the interaction of light and matter that is at the heart of pigment mixing. Today, there is no strict definition of what constitutes a primary in ink mixing, but over time, a set of a dozen or so base colorants has become commonplace. Similarly, primaries of additive mixing and subtractive mixing vary depending on the application. Zander’s definition, inclusive of both additive and “mechanical” mixing, was likely influenced by the availability of colorants and the confusion surrounding the exact nature of human vision.
Zander noted that contemporary understanding was changing rapidly, stating “It appears to me that our notions and theories regarding colour phenomena are gradually but surely changing, owing to the greater facilities and consequent greater correctness of observation both as regards apparatus and purity of pigments now available.” He invited readers to challenge the common teachings of the day, and rather than regurgitate what was already written, promoted “independent research and verification.” This willingness to challenge the common practice of the day, in spite of the possible criticism it might induce, was Zander’s greatest strength. He was a man of conviction.
Though the public agreed that certain chromatic colors were better reproduced with four-colors, Zander failed to win over public opinion and prove that the four-color prints were of a quality superior to three-color prints,.
In 1906, Stephen Horgan, a respected printer and writer for the Inland Printer, wrote: “All allowances must be made for Mr. Zander’s enthusiasm over his own baby, but in reproducing a specially selected subject, as he does in the [Penrose’s Pictorial] Annual, he must show more of an improvement by comparison than he does.”15 Horgan wrote again in the following year, “Those who have invested capital in the three-color process need not fear that Mr. Zander’s four-color method was going to supersede their system. Mr. Zander was simply overenthusiastic about that baby of his.”16
Horgan’s primary criticism of Zander’s four-color process was the increased material and labor expense that the production and make-ready of a fourth plate would require. It seems that Zander was unable to produce any tangible evidence that the quality of his four-color exceeded those of three-color processes. Some samples that Zander did provide were over-etched, which is counter to one of Zander’s claims that his process would reduce the need for fine-etching. Horgan continued to state that the four-color process “would therefore take more time than ordinary three-color [printing], and depends on extremely skilled labor, having very little or nothing to do with the particular method suggested by Mr. Zander, which, if it is any improvement at all, must justify itself by its mechanical superiority.”
Zander was not the first to propose printing with red, yellow, blue, and green inks. Dr. Eder first suggested printed with four chromatic colors in 1896. He stated, as quoted by Baron Arthur von Hübl in 1906, “working according to [Hering’s] theory, all the manipulations necessary for three-colour printing may be adapted without difficulty to a four-colour system.”17
Von Hübl believed the four colors should be located equidistant around the hue circle so the absorption spectra of each photographic filter would be evenly spaced, unlike the four colors Zander’s process which were not evenly spaced. However, it is clear that von Hübl was not a fan of four-color printing, regardless of the primary color hues. He argued that, every color in the spectrum was covered by the absorption regions of two filters, rather than one in three-color printing, which gives color a dirty appearance. Von Hübl also stated the yellow of Zander’s four-color system was closer to orange, and thus the system would not produce a pure yellow. It’s unclear where von Hübl acquired this information. Zander clearly specified in his writings that his recommended primary was a “lemon yellow.” Von Hübl felt that, on top of the inherent weakness in four-color systems due to overlapping absorption regions, Zander’s system would only make the problem worse with it’s uneven hue spacing.
Zander indicated that he was accused by Dr. Eder himself of being a copycat, an accusation he firmly denied, stating “Dr. Eder alleges that my colour process is based on Hering’s theory, and was anticipated by him in a lecture he gave…ten years ago, permit me to state that, to express it mildly, Dr. Eder is mistaken. My colour process was arrived at by empirical methods independently of, and is not based on Hering’s theory.”18 In the face of this criticism, Zander sought to dissociate his ideas from Hering’s. Regarding von Hübl’s claims about Zander’s primary hues, Zander stated that they did not agree well with his research. He also disputed von Hübl’s claims that his four-color process produced dirty colors, arguing that his process is in commercial use and is “reproducing colours not only far brighter than three-colour work, but also the range of colours obtainable is far more extensive an accurate.”19
Some of Zander’s contemporaries thought he was proposing to replace three-color printing with his four-color process, but those who read his work knew he understood that “if a printer can satisfy the requirements of his customer with three-color prints, or even two-color prints, it would be folly to use four printings.”20 It was also understood that Zander never claimed to do away with fine-etching, only to minimize the etching process with proper application of his technique.
Another observer noted that implementing the four-color process would require extensive re-training of engravers: “The initial difficulty is that the fine-etcher who is use to three-color work, and the way in which mixed color are built up in it, is almost sure to make a hash of Zander’s process the first time he works it, and has to adapt himself to a different method of correcting etching.”20 The editors of the British Journal of Photography, in examining a sample image of hat provided by Zander, noted that “a few hats show the whole of the four-colours printed on them, some of the work being either deep, etched away, or engraved away by hand.”21 They concluded, as did others, that manual work was still required to implement Zander’s process, and that manual work for four plates was less desireable than manual work for three.
William Gamble,# looking back on Zander’s process in 1912, noted that some of Zander’s specimens produced brilliant colors, such as the one below, but wrote that “the results were not entirely convincing,” and, coming to the same conclusions as other critics, “probably, through the engravers not having sufficient practice with the new method. Printers did not view with favour the idea of a fourth printing, and on the whole the process was received so coldly that the inventory has not pushed it further.”22
After 1907, as Gamble suggested, Zander seemed to stop writing and back away from the public discourse. No mention of him in newspapers, journal articles, or other sources could be found after that year. Zander’s chief argument against his critics was that his four-color process was in commercial use and could produce brilliant colors. However, it required a precise execution. Perhaps it is true that four-color printing was simply too complicated to justify its replacement of the three-color processes. The extra costs associated with training, plate making, make-ready, and materials, outweighed any benefit Zander’s four-color process might provide.
It must have been hard for a man who so ardently preached the gospel of three-color printing to suddenly change course and preach a completely new philosophy so at odds with contemporary thinking and public opinion. Although there’s no indication that he was respected any less after proposing his four-color process, he was certainly subjected to harsh criticism by his colleagues and placed in a defensive position from which he does not appear to have freed himself. However, despite his lack of commercial success, Zander’s contributions to the discourse on color printing, including what we now refer to as Expanded Gamut Printing, should not be overlooked. He was a prominent thought leader, contributing to the improvement of three-color printing, and helping stretch the imaginations of his peers by simply suggesting four colors be used instead of three.
Zander passed away, according to one source, in 1921, at the age of 51.23 No obituary could be found.
Disclaimer
This article was written by Brian Gamm in his personal capacity. The views, thoughts, and opinions expressed in this article belong solely to the author, and not necessarily to the author’s employer, organization, committee or other group or individual with which the author has been, is currently, or will be affiliated.
- *The letters marking points on the x-axis represent Fraunhofer lines of the spectrograph, the most common color measurement device of the day.
- †The filters were made from either colored glass or troughs filled with a colored liquid.
- ‡Zander did claim to have found a good candidate for transparent yellow, a yellow lake, but it does not appear he ever made use of it or publicized it.
- §Still today, ink manufacturers try to reduce the number of components included in ink formulations, though to save cost and complexity, not because of a strict adherence to trichromacy.
- ¶It was not until the 1950s that the debate was finally settled when Hurvich and Jameson proposed a dual process theory in which both the Young-Helmholtz and Hering theories were correct.
- #Editor of Penrose’s Pictorial Annual
- 1.“Mr. C. G. Zander.” Inland Printer. 1906;37:719.
- 2.England and Wales Census, 1901, Charles G Zander, Islington, London, England, United Kingdom; from “1901 England, Scotland and Wales census,” database and images, findmypast (http://www.findmypast.com : n.d.); citing Highbury subdistrict, PRO RG 13, The National Archives, Kew, Surrey. Family Search. Published May 20, 2020. Accessed September 7, 2020. https://familysearch.org/ark:/61903/1:1:X98F-GFQ
- 3.Zander CG. Photo-Trichromatic Printing: In Theory and Practice. Raithby, Lawrence & Co., Ltd.; 1896. Accessed September 7, 2020. https://archive.org/details/phototrichromati00zand
- 4.Zander CG. Colour Curves and Pigments. Penrose’s Pictorial Annual. 1901;7:17-24.
- 5.Zander CG. Yellow, Red, and Blue. Penrose’s Pictorial Annual. 1899;5:17-19.
- 6.Farmer H. The Optics of Trichromatic Photography. British Journal of Photography. 1901;48(2126):68-69.
- 7.Zander CG. Trichromatic Printing Inks. The Photographic Journal. 1904;28:311-321.
- 8.Zander CG. The Confusion of Colours. Penrose’s Pictorial Annual. 1900;6:61-63.
- 9.Zander CG. Standardizing. Penrose’s Pictorial Annual. 1907;13:121-122.
- 10.Zander CG. The Complementary Colour Reproduction Process. Penrose’s Pictorial Annual. 1905;11:9-12.
- 11.Zander CG. Practical Colour-Mixing. Penrose’s Pictorial Annual. 1904;10:65-68.
- 12.Zander CG. Process of Photomechanical Reproduction of Colors and The Resultant Article. Published online 1908. Accessed September 7, 2020. https://patentimages.storage.googleapis.com/a0/eb/67/24f0928cae19dd/US884254.pdf
- 13.Hering E. Outlines of a Theory of The Light Sense (Introduction). Harvard University Press; 1964.
- 14.Zander CG. Chromatic Aberations. Penrose’s Pictorial Annual. 1906;12:17-20.
- 15.Horgan SH. Zander’s Four-Color Process. Inland Printer. 1906;36:897.
- 16.Horgan SH. Zander’s Four-Color Process Again. Inland Printer. 1907;38:403.
- 17.von Huebl A. Three-Colour and Four-Colour Photography. The British Journal of Photography. 1906;53:693-695.
- 18.Zander CG. The Zander Four-Colour Process: To the Editors. The British Journal of Photography. 1906;53:397.
- 19.Zander CG. The Zander Four-Colour Process: To The Editors. The British Journal of Photography. 1906;53:719.
- 20.Zander’s Four-Color Process. Inland Printer. 1907;39:236-237.
- 21.The Zander Four-Colour Process. The British Journal of Photography. 1906;53:832.
- 22.Gamble W. Modern Colour Process. In: Burch R, ed. Colour Printing & Colour Printers. Baker & Taylor Co. ; 1911:261-262.
- 23.“England and Wales Death Registration Index 1837-2007: Charles Zander, 1912; from “England & Wales Deaths, 1837-2006,” database, findmypast (http://www.findmypast.com : 2012); citing Death, Whitechapel, London, England, General Register Office, Southport, England. Family Search. Published December 31, 2014. Accessed September 7, 2020. https://familysearch.org/ark:/61903/1:1:2N1W-8Y4
