Part 1: the quest for colour
Today we take colour photography for granted. Taking pictures in full, natural colour is so easy that we don’t pause to consider how it all came about. Yet the search for a cheap and simple process of colour photography followed a long and difficult quest with many a wrong turn and dead end. In this search for photography’s ‘Holy Grail’, a few fortunes were made, but many more were spent before the dream was to become a reality.
In 1839, when photographs were seen for the very first time, they were greeted with a sense of wonder. However, this amazement was soon mixed with a tinge of disappointment. People didn’t understand how a process that could record all aspects of a scene with such exquisite detail could fail so dismally to record its colours. The search immediately began for a means of capturing accurately not only the form but also the colours of nature.
While scientists, photographers, businessmen and experimenters laboured, the public became impatient. Photographers, eager to give their customers what they wanted, soon took the matter, literally, into their own hands and began to add colour to their monochrome images. As the writer of A Guide to Painting Photographic Portraits noted in 1851:
When the photographer has succeeded in obtaining a good likeness, it passes into the artist’s hands, who, with skill and colour, give to it a life-like and natural appearance.
Several different processes and materials were used for hand-colouring, and it provided studio employment for many miniature painters who had initially felt threatened by the appearance of the new medium. Even after the emergence of the first practical colour processes, hand-colouring continued to be popular since it was often a cheaper and simpler alternative. Some of the results were very crude, but in the right, skilled hands, effects of great subtlety and beauty could be achieved.
Even at its very best, however, hand-colouring remained an arbitrary and, ultimately, unsatisfactory means of recording colour which could not reproduce the colours of nature exactly. What was required was a photographic process that could record colours directly in just the same way that it was already capable of capturing light and shade.
Before colour could be reproduced the nature of light and how we perceive colour had to be clearly understood. The scientific investigation of colour had begun in the seventeenth century. In 1666, Sir Isaac Newton split sunlight with a prism to show that it was actually a combination of the seven colours of the spectrum. Nearly 200 years later, in 1861, a young Scottish physicist, James Clerk Maxwell, conducted an experiment to show that, in fact, all colours can be made by an appropriate mixture of red, green and blue light.
Maxwell took three separate lantern slides of a tartan ribbon through red, green and blue filters. These slides were then projected through the same filters using three separate magic lanterns. When the three images were carefully superimposed on the screen, they combined to make a coloured image which was a recognisable reproduction of the original. While Maxwell’s experiments demonstrated clearly the basic principles of colour photography, in practice, his demonstration should not have worked at all. Although the physicist didn’t know it, the photographic emulsions that he used were insensitive to red light. Fortunately for Maxwell, the red cloth in the ribbon reflected ultraviolet light. This was invisible to the eye but did register on the emulsion.
While the fundamental theory may have been understood, however, a practical method of colour photography remained elusive. Some experimenters pursued the idea of a direct method of colour reproduction which did not rely on mixing primary colours. In 1891 Gabriel Lippmann, a professor of physics at the Sorbonne, demonstrated a colour process which was based on the phenomenon of light interference—the interaction of light waves that produces the brilliant colours seen in soap bubbles. This process won Lippmann a Nobel Prize in 1908 and was marketed commercially for a short time around the turn of the century. However, the extremely long exposure times required meant that the process was to remain little more than a scientific curiosity. The future of colour photography lay with three-colour processes that relied on mixing the primary colours of light.
Not long after Maxwell’s 1861 demonstration, a Frenchman Louis Ducos Du Hauron announced a method for creating colour photographs by combining coloured pigments instead of by mixing coloured light. Du Hauron’s process still required three black and white negatives taken through red, green and blue filters. These negatives were used to make three positive, separately-dyed images which, when superimposed, combined to give a coloured photograph. It is this method which forms the practical basis of today’s colour processes.
Initially, the work of Maxwell, Du Hauron and others, despite its theoretical importance, was to be of limited practical value. This was largely due to the fact that the photographic emulsions in use at the time were very limited in their colour sensitivity. Before their methods would work in practice, photographic materials that were sensitive to the whole colour range of the spectrum had to be introduced.
By the 1880s, plates that were sensitive to blue and green light were commercially available. It was not until the early years of the 20th century, however, following the work of Dr H W Vogel, that the first fully panchromatic plates, sensitive to all colours, were sold. At last it seemed that the way lay clear for the future possibility of a practicable and commercially viable method of colour photography.
Part 2: Additive colour
The first processes for colour photography appeared in the 1890s. Based on the theory demonstrated in the1860s by Maxwell, these reproduced colour by mixing red, green and blue light. The American photographer and inventor Frederic Ives devised a system based on three colour-separation negatives taken through coloured filters. From these negatives, positive transparencies were made which were placed in a special viewer, called a Kromskop. Mirrors in the Kromskop superimposed the images on the three transparencies and a second set of filters restored the colours. Kromograms, as the resulting images were known, were effective but prohibitively expensive, and Ives’ system was, ultimately, too complex to be successful.
Instead of making three separate exposures through red, green and blue filters, a simpler, alternative approach was to make just one exposure through a filter that combined all three primary colours. The first process to use this method was devised by Dr John Joly of Dublin, in 1894. Joly covered a glass plate with very fine red, green and blue lines (less than 0.1mm wide) in order to create a three-coloured filter screen. When taking a photograph, this screen was placed in the camera in front of the plate. After exposure and reversal processing, the black and white positive image was carefully placed in register with another filter screen. The result was a colour transparency which could be viewed by transmitted light.
The Joly process was introduced commercially in 1895, and remained on the market for a few years. However, the limited colour sensitivity of the plates then in use meant that the results were not very successful.
The first fully practical and commercially successful screen process—the autochrome—was invented early in the 20th century by two French brothers, Auguste and Louis Lumière, who had been experimenting with colour photography since the 1890s. They published their first article on the subject in 1895, the same year that they were to achieve lasting fame for their invention of the cinématographe. In 1904 they gave the first presentation of their process to the French Academy of Science and by 1907 had begun to produce autochrome plates commercially.
News of their discovery soon spread and examples of the new plates were eagerly sought. Critical reaction was rapturous. Alfred Stieglitz wrote:
The possibilities of the new process seem to be unlimited… soon the world will be color-mad, and Lumière will be responsible.
Realising there was no need to have the filter screen separate from the photographic emulsion, the Lumières combined their filter screen and light-sensitive emulsion on the same glass support.
Manufacturing autochrome plates was a complex process. First, pulverised starch grains were passed through a sieve to isolate individual grains between 10-15 microns in diameter. Many different types of starch were tried, but the humble potato gave the best results. These microscopic grains were then dyed red, green and blue-violet, mixed and spread over a glass plate, and coated with a sticky varnish. Next, charcoal powder was spread over the plate to fill any gaps between the coloured starch grains. A roller submitted the plate to a pressure of five tons per square centimetre to flatten out and spread the grains before the plate was varnished to make it waterproof. The final plate was a three-coloured filter screen, on every square inch of which were about four million transparent starch grains, each one acting as a coloured filter. The final stage was to coat the plate with a panchromatic emulsion.
Autochrome plates were simple to use. They required no special apparatus and photographers were able to use their existing cameras. Exposure times, however, were long—about 30 times those of conventional plates. Even in bright sunshine, an exposure of at least one second was needed, and in cloudy weather this could be increased to 10 seconds or more. Even in a well-lit studio, portraits could require an exposure of as long as 30 seconds.
Following exposure, autochrome plates were reversal-processed to produce a positive image. When viewed by transmitted light passing through the plate, the millions of tiny red, green and blue-violet grains combined to give a full-colour photograph, accurately reproducing the colours of the original subject. In theory, the grains were mixed and distributed randomly on the surface of the plate. In practice, however, mathematical probability meant some grouping of grains of the same colour was inevitable. While individual grains are invisible to the naked eye, these groups of clumps are visible; they are the reason for the autochrome’s distinctive beauty and for
comparisons with the work of Impressionist and Pointillist painters.
By 1913, the Lumiere factory in Lyon was producing 6,000 autochrome plates every day. The process’s commercial success prompted the appearance of many other colour processes based on the concept of screens made up of microscopic colour filters. These screens used either a random grain pattern or, more commonly, different geometric patterns of lines and squares.
Most of these processes are now long forgotten, but one remained popular for years. Dufaycolor first appeared in 1932 as a 16mm cine film, followed in 1935, by a rollfilm version. Devised by Louis Dufay, Dufaycolor employed a regular geometric screen of red lines alternating with rows of green and blue rectangles. Colour reproduction was good and it was comparatively fast—although only one-third of the speed of contemporary black and white film. Whereas autochromes appealed to photographers who liked to do their own processing, Dufaycolor was aimed at the snapshot market. A processing service which returned finished transparencies, mounted and ready for viewing, opened up colour photography to a whole new class of photographers. Dufaycolor, the last of the screen processes, remained on the market up to the 1950s.
Part 3: Subtractive colour
Most early colour processes worked on the principle of mixing, or adding together, appropriate combinations of red, green and blue light. These are usually grouped together under the general description of ‘additive’ colour processes. All additive processes share one great disadvantage—they rely on the use of filters which, by their nature, block out a great deal of light, resulting in long exposure times and very dense
transparencies. Moreover, the colour photographs made by using these processes can only be viewed by transmitted light—by projection or by using special viewing devices. However, there is an alternative method of reproducing colour photographically—‘subtractive’ colour synthesis.
The original theory for subtractive colour reproduction can be traced back to the fertile mind of Louis Ducos du Hauron who, as early as the 1860s, explained the method in his book Les Couleurs en Photographie. Du Hauron proposed that colour separation negatives be used to produce three positive images which were then dyed the complementary colours of cyan (bluegreen), magenta (blue-red) and yellow. Each of these complementary colours absorbs, or subtracts (hence the name), one of the primary colours. Cyan absorbs red light, reflecting a mixture of blue and green light. A cyan image, therefore, performs the same function as the red filter used in an additive process. Similarly, magenta absorbs green light and yellow absorbs blue light. By accurately superimposing these three complementary colours, all other colours can be reproduced. The colour in subtractive processes comes from dyes or pigments rather than coloured filters.
With subtractive colour, white, for example, is represented by clear glass or by white paper rather than by light passing through three filters. This means that subtractive processes are much less wasteful of light. More importantly, they work with reflected rather than transmitted light which means that they can be used to produce colour photographs on paper.
The development of subtractive colour processes followed two distinct paths. First, the design of specialised cameras, for taking sets of colour separation negatives and, secondly, the search for practical methods of making and superimposing three positive images in the complementary colours.
When taking colour separation negatives of subjects that did not move—for example, a vase of flowers—a conventional camera could be used. All that was needed was to change the colour filter after each exposure. If a great deal of colour work was being done, this procedure could be made simpler through the use of repeating backs. A number of different devices of this sort were marketed. The simplest type were long plateholders, fitted with three filters, which could be slid along the camera back in three steps. The most complex were fitted with clockwork motors, enabling three negatives to be exposed in rapid succession in as little as two or three seconds.
When photographing subjects where movement was likely to occur—such as portraits—even automatic repeating backs were not fast enough. For these, a camera that could expose all three negatives simultaneously was needed. Over the years, many designs for such ‘one-shot’ cameras were patented and a number were produced commercially. These used various arrangements of mirrors and prisms to split the light entering the camera into three separate beams, each of which went to a plateholder fitted with a different coloured filter. Among the most successful designs were the Jos-Pe, Bermpohl, Klein and Mirkut cameras.
Obtaining satisfactory negatives was only the first stage. These negatives then had to be converted into positive images in the complementary colours of cyan, magenta and yellow. Several different methods were used to obtain these images, the most popular being variations of the carbon process. These used sheets of carbon tissue, consisting of a gelatine coating, containing pigment, on a paper base. This tissue was sensitised before use by soaking it in potassium bichromate. Potassium bichromate hardens when
exposed to light and, after exposure in contact with a negative, the areas of unhardened gelatine could be washed away, to reveal an image.
Tissues could be produced using pigments of any colour—images on cyan, magenta and yellow tissues being superimposed to produce subtractive colour prints. A variant of the carbon process was the Trichrome Carbro process, first developed during the 1890s but made popular by the Autotype Company of Ealing, during the 1920s and 1930s. The Carbro process used a set of bromide prints made from separation negatives to make the necessary yellow, magenta and cyan pigment images on tissue for transfer in sequence on to a paper base.
While processes such as Carbro were available for amateur photographers to use, tissue assembly techniques were difficult and complex. Apart from the really dedicated, most amateurs preferred to use additive processes such as autochrome and Dufaycolor. Commercial colour photography was to become increasingly important during the 1930s and for professional colour printing at this time, one process was to reign supreme: Vivex.
Invented in 1928 by Dr D A Spencer, who later went on to become Managing Director of Kodak Ltd, Vivex was a modification of the Trichrome Carbro process in which sheets of cellophane were used as temporary supports for the pigment images. Any minor problems with registering the images could be corrected manually by stretching or squeezing the cellophane to ensure perfect superimposition. To exploit the Vivex process, a company called Colour Photographs (British & Foreign) Ltd was formed with a factory in Willesden, north London. This was the first laboratory to offer a colour print making service to professional photographers. It has been estimated that over 90% of all the colour prints made in Britain during the 1930s were produced using the Vivex process.
Subtractive processes such as Vivex used separation negatives produced on three separate photographic plates. It is possible, however, to combine all three films or plates into a single unit. Such combinations are known as tripacks. It was the use of integral tripacks that paved the way to the development of ‘modern’ colour processes such as Kodachrome.
Part 4: A quest fulfilled
Subtractive colour processes such as Vivex required colour separation negatives to be made on three separate photographic plates. However, if it were possible to combine all three plates into a single unit, or tripack, then there would be no need for specialised colour cameras or for repeating backs fitted with filters.
The basic idea was to construct a multi-layer unit, where each plate was coated with an emulsion sensitive to one of the primary colours. Light would pass through the first plate in order to reach the second emulsion layer and, in turn, pass through that plate to register on the third emulsion.
The first practical tripack system was introduced by Frederic Ives in 1916. His ‘Hiblock’ tripack consisted of a sheet of film sandwiched between two glass plates. The top plate was blue-sensitive, the film was green-sensitive and the bottom plate was sensitive to red light. After exposure the three layers were separated for processing, after which the negatives were treated as conventional separation negatives.
Other tripack systems followed, including the infamous Colorsnap process. In 1928, a new company, Colour Snapshots Ltd, was set up with massive financial backing in order to promote Colorsnap products. However, despite extravagant claims, the results were disappointing. The negatives from the second and third emulsion layers were so unsharp that the company was reduced to hand-colour black and white prints made from the sharpest, front element of the tripack. Unsurprisingly, Colour Snapshots Ltd went bankrupt in December 1929.
The Colorsnap process suffered from the same problem inherent to all tripack systems. Light was scattered and diffused as it passed through the various layers of emulsion and support, so one or more of the resulting negatives were blurred. Definition was too poor to allow much enlargement; tripack negatives were usually only recommended for contact printing. The solution to this problem was to coat all three emulsions on to the same glass or film support in direct contact with each other in ‘integral’ tripacks. Since it would be physically impossible to separate these emulsion layers, each would have to be capable of being chemically processed in isolation so as to produce an image in cyan, magenta or yellow.
In 1912, Rudolph Fischer had patented a proposal to use what later became known as colour couplers. These are substances that react with chemicals formed during development to form coloured dyes. Fischer suggested that colour couplers for producing cyan, magenta and yellow dyes be incorporated into the appropriate layers of an integral tripack so that during development coloured images would be formed. Since, with integral tripacks, all three emulsion layers are in direct contact with each other, there would be no problems with registration and the result would be a full colour photographic image. Unfortunately, the colour couplers Fischer used tended to disperse between emulsion layers during processing. Fischer’s theory, however, was perfectly sound, and his work was to form the basis of research that was to lead to the first practical and commercially successful integral tripack system—Kodachrome.
Kodachrome was the brainchild of Leopold Mannes and Leopold Godowsky. Both earned their living as professional musicians (Mannes played the piano and Godowsky the violin) while spending their spare time experimenting with colour photography. Despite their best efforts, there came a point when they were unable to progress without outside support.
This support was to come from Dr C E Kenneth Mees, director of the Eastman Kodak research laboratories in Rochester, New York. In 1922, Mees met with Mannes and Godowsky and, impressed with the quality of their work, agreed to supply them with the materials they needed to continue their research. At this point, they were working on a two-colour subtractive system for colour photography but, after reading about Fischer’s work with colour couplers, they decided to abandon their previous methods and concentrate on developing a practical three-colour multi-layer film system. In 1931, the two Leopolds gave up their musical careers to work full-time in the Kodak research laboratories where, with the help of Eastman Kodak’s enormous resources they made rapid progress.
Like Fischer, Mannes and Godowsky had great difficulty in preventing the coloured dyes spreading between the emulsion layers. They overcame this by putting the colour couplers in the developer instead of in the emulsion. Kodachrome is, in effect, a black and white film to which coloured dyes are added during processing. Kodachrome processing, involving repeated development, dyeing and then selective bleaching, was extremely complex. In all it required at least 28 different stages that could only be carried out in laboratory conditions. For this reason, photographers were unable to process their own film but had to send it back to Rochester. On 15 April 1935, the first Kodachrome film went on sale—for use in 16mm cine cameras. American photographers had to wait until the following year before 35mm Kodachrome film was available. The first supplies of 35mm Kodachrome reached Britain in 1937.
Meanwhile, in Germany, Agfa also announced a multi-layer colour film in 1936. Agfa had been making additive colour plates since 1916, so they called their colour film Agfacolor-Neu—‘new’ to indicate that it was completely different from any earlier products. Agfacolor-Neu was the first commercial process to follow Rudolph Fischer’s theory of using colour couplers.
Agfa’s research chemists had discovered a way of anchoring couplers in the individual emulsion layers. This made Agfacolor film much easier to process. Unlike Kodachrome, it could even be done by the user at home. After the Second World War, the details of Agfa’s research became freely available and other companies—such as Ferraniacolor and Gevacolor—introduced colour films based on the same principle.
With the perfection of dye-based multi-layer colour films such as Kodachrome and Agfacolor-Neu, a new era of colour photography had dawned. The quest for colour—a search that had begun with the announcement of the invention of photography nearly one hundred years earlier—was over.
- Brian Coe, Colour Photography: The First Hundred Years. Ash & Grant, 1978
- Jack H Coote, The Illustrated History of Colour Photography, Fountain Press, 1993
- Joseph S. Friedman, History of Color Photography, American Photographic Publishing Co, 1945
- Pamela Roberts, A Century of Colour Photography, Carlton Books, 2007
- E.J. Wall, The History of Three-Color Photography, Focal Press, 1970
- John Wood, The Art of the Autochrome: The Birth of Color Photography, University of Iowa Press, 1993.
For an extensive bibliography on the autochrome process, see:
- John Wood, The Art of the Autochrome: The Birth of Color Photography, University of Iowa Press, 1993, pp165-185
- John Wood, The Art of the Autochrome—A Supplemental Bibliography, History of Photography, Summer 1994, pp140-142