Lesson Nine – Colours

Zeuxis and Arte walk in the lanes of the flower show of Epinal. The flower show is in Epinal’s great hall, somewhat outside the town. The hall is very high, and as the Epinal builders were very proud of their wood industry, its roof is supported by incredibly long wooden beams. The wood of the vast forests of the Vosges Region provide to the visitors an impression of warmth. The lanes in the hall, between the flower beds, are narrow, for all around are exhibition spaces. The horticulturists and nurseries of the environment, from France, Belgium and even Germany, have come together to show the spelndour and vigour of their flowers. Every exhibitor has competed with the rarest flowers, the most extravagant constructions of flowers, the finest colour combinations. In the beginning, Zeuxis does not talk. He and Arte simply admire the fabulous show of hues. Then, Zeuxis sighs and starts to talk.

Zeuxis: It is simply marvellous, Arte, to walk in here. I have rarely seen such effervescence of different colours, and the forms of all these flowers are enchanting. I wonder why red is red and why blue is blue. Do you know what colour is?

Arte: What do you mean what colour is? Look around you, Zeuxis.
Arte turns on herself to take in the wealth of colours.
All the flowers have different colours. I see each flower with another colour, even when the colours sometimes are almost the same. Colour is a property of the flowers; it is a property of the object.

Zeuxis: Is it now really girl? Look at this tulip. It has the fiercest red I have ever seen. Wait, I’ll take it a bit in the shade. What happens?

Arte: All right, wise guy! The tulip turns to dark grey! For colours we need light. But in the shade, I can still se the tulip even if it turns grey. Colour is not a property of the object in itself. I know now that I was stupid, of course! It is obvious, but I didn't realise! I learned at school of Isaac Newton’s experiments with prisms. Newton separated various kinds of light that all together constitute white light. When light falls on the tulip, the red kind of light is reflected on the petals and touches my eye. So that is how I see red. And when there is no light, I can see no colour. So in the shade, the tulip turns grey to my eye.

Zeuxis: Hmm. Let’s try another flower.
Zeuxis takes a blue tulip now and brings it in the shade.

Arte: Gee, Zeuxis, I continue to see that tulip blue. It should turn grey like that red one. OK, you got me again. You win! There must be more to colours than merely being a property of light, or to being an effect of interaction of light with an object. Stop teasing me by showing your knowledge. I know you know all, so be kind and explain.

Zeuxis: All right, Arte. Colour is indeed not really a property of the object, and neither is it really a property of light. Of course, the surface properties of the object and the properties of light have to do with colours. But we see colours with or eyes, and our organ comes into play too, and also what happens with the sensations of light transferred by our eyes to our mind.
Colour is a very simple concept with which we are constantly confronted, Arte. We cannot see forms without seeing colour. Yet it is one of the most complex phenomena of nature. Several physical and physiological processes have to be executed in sequence before we humans can assign the quality of colour to an object. And many if not all of these processes are still not well understood. They also are at work without us really being conscious of them. I’ll explain. But it will take a lot of theory, because we first have to consider the theory of seeing.

Arte: Between such a marvel of these flowers, Zeuxis, I can be patient.

Zeuxis: Light is a physical process. You knew that already from school, Arte. We ascribe the feature that we call "colour" commonly to a property of light, even though we just saw that there is more to it. Light is a physical phenomenon that consists of radiation of energy, and it is a property of our perceived universe. We describe light as particles, which have small, but not negligible, mass and thus energy for mass is energy. We can however also describe light as an electromagnetic wave. It is not a particular electromagnetic wave, but just those electromagnetic waves that interact with our eyes.

Arte: What is an electromagnetic wave?

Zeuxis: Nobody really knows exactly. Our scientists know that the waves exist; they can describe them with mathematical formulae and describe some of their properties, but why they exist and what they are is still very much a mystery. As an electromagnetic wave, light seems to be a property of the universe, a potentiality of the environment, a possibility created in the universe for certain phenomena to happen.

Arte: But you said light was also particles!

Zeuxis: Here already, in what we believe to be the most well understood property of the processes of colour perception, lies the paradox of a dual nature. The paradox that light can be both a wave and a particle has been well proven experimentally. Light manifests itself sometimes as particles, called photons, and sometimes as a wave. Both the photons and the waves are probably only a human's way of explaining certain phenomena that occur when the potentiality of interaction with the environment of the created possibility, realises to a tangibke effect. Light is transmitted through a medium, the air, and interacts with the molecules of air and with the particles of impurities suspended in the air. Then the light arrives on us. The light interacts with us, humans in a perceptible way through our eyes. Light of different wavelengths, or photons of different energy, fall on the retina of our eyes and stimulates there various molecules.
This excitation of the retina is a photochemical process. Scientists have largely, but not entirely, discovered exactly which molecules are stimulated, by which wavelengths or by which energies of the photons. It has remained unclear also what kind of signals are sent to our brain, either as a direct result of the waves or as a result of differences of wavelengths of adjacent waves. The formation and the sending of signals to our brain over the nerve cells are a physiological process. To which places of the brain the signals are sent is not entirely known. Unknown also is exactly which neurones are stimulated and how.
When the flux of light that reaches the retina diminishes, our physiological impression of the light of an object moves in the direction of violet. One can verify this by looking at coloured objects and diminish steadily the light of the environment. This is called the effect of Purkinje. It makes snow look blue at moonlight, and distant mountains also seem blue in the evening and at dawn. It lets you continue to see a blue tulip blue in the shade, but a red one dark. The way we perceive light is not linear with the intensity of that light.
Even less is known about the process that arises in our brain, the processes by which we recognise and give a quality of "colour" to the received signals. Why we "see", and call "green" the impression our mind has of a quality of the light perceived from a grass meadow, after that quality has gone through various successive transformations until it reached our brains as electric signals, remains a deep mystery. The mind-processes also are still unknown.
The perception of colours is particularly prone to illusions. But we have to be careful when it pertains to colour to speak of "illusion", since the real colour is only the colour that is perceived, and not one or other "physical" or intrinsic quality of the object. Viewing a colour just on its own on a flower will produce a certain hue in our eyes. Seeing just one colour is practically impossible, however. That same flower may present an entirely different hue when seen surrounded by flowers of other colours. Colours interact, and a colour seems to have only a quality of hue, as well as of intensity, relative to other hues.
Finally, the perception of the quality of colour of objects calls to our mind certain associated impressions. Red means danger, green safety and tranquillity, and yellow cheerfulness. These psychological processes of association can be explained by analogy, but why they exist is largely unexplained. The information transmitted to our brain and its nerve cells is linked to previous experiences, the memories of which are stored in our mind.
No wonder then, Arte, that so many scientists and artists alike were astonished at the nature of light, when they observed its challenging effects and wrote their interrogations into the nature of colour vision.
Josef Albers, a painter and lifelong teacher of art and colours expressed this as follows, "In visual perception a colour is almost never seen as it really is – as it physically is. This fact makes colour the most relative medium in art. In order to use colour effectively it is necessary to recognise that colour deceives continually" ^G94. I will even argue that there is no such thing as a physical reality of colour!

Arte: You know so much, Zeuxis. You are a Greek coming from the beginning of times, so far it seems to me, yet you know so much. How is that possible?

Zeuxis: I did not just sleep through the centuries, Arte. I saw many smart people think on art and on colours, and I read what they produced.
Greek philosophers studied vision and reasoned on how vision could happen. Empedocles wrote that vision came from an interaction between the exterior light and a light emanating from the human body. Plato endorsed this view of light and vision that seemed to come from the inside of people. Democritus gave a more physical interpretation of vision, but also thought of vision as an interaction between properties of the world and the properties of humans. We now know that these ideas were not that wrong, as we will see later.
Aristotle thought that colour arose from the transition from lightness to darkness. He explained vision as an action on the eyes. Aristotle’s works were known before the sixteenth century in Western Europe, so that in the Middle Ages colours were positioned between light and black, as gradations of brightness. Aristotle’s books were published in Italy in the late fifteenth century, and his theories were rediscovered and studied in the sixteenth century. These were commented upon in various books. Aristotle’s treatise of colours was first printed in our modern times in 1537 in Naples and had comments by Simon Portius, who sought old Latin and Greek books for the Medici rulers of Florence. In 1548 the same Portius commented Aristotle’s works in a Florentine edition dedicated to Cosimo I de Medici.
Many other works of the fifteenth, sixteenth and later centuries treated colour. Leon Battista Alberti (1404-1472) spoke of colours in his "Della Pittura", on the "Art of Painting" around 1435. Giorgio Vasari wrote of colours in his "Lives of the Artists" of 1550. Lodovico Dolci edited a "Dialogue on Colours" in Venice in 1565, and he also tried to explain some of the notions of artistic harmony of colours. Another treaty on colours with the title "Occolti Trattato de Colori" was edited in Parma in 1568. Leonardo da Vinci was of course interested in understanding colours and he spoke of colour effects in his "Trattato della Pittura". For Leonardo, the sequence of colours in the rainbow was the norm to compose harmonious colour combinations by. He wrote on aerial perspective, and noted that he saw objects at a great distance all with the same hue and tone. He noted already that certain colours strengthened each other when placed together. He also proposed to contrast darker tones with brighter ones, such as light blue with a red colour. Leonardo da Vinci thus perceived a change in colours when they were juxtaposed, a phenomenon that Michel-Eugčne Chevreul would describe three hundred years later and call "simultaneous contrast".
Guido Antonio Scarmiglione of Fuligno (d. 1620), a physician who first worked in Naples but later came to the court of Emperor Rudolph II in Prague, wrote "De Coloribus", a book published in Marburg in 1601, and this was equally an overview of the theories of colours of Aristotle. His work was dedicated to Emperor Rudolph II ^G76. According to the art historian John Gage, Scarmiglione argued that the Aristotelian views of colour, as activated by light on the surfaces was manifestly untrue ^G97.
The truly modern investigations into the physical phenomena of light started with Isaac Newton (1643 – 1727). Newton wrote in 1665 that a transparent prism bent light from the sun into a spectrum of colours ranging from red (least bent) to violet (most bent). Newton wrote about vibrations in the ether, which propagated through that ether to our eyes, thus effecting in humans a sensation of colours. He wrote a book published in 1730 called "Optics, a Treatise of the Reflections, Refractions, Inflections and Colours of Light". He already devised a colour wheel in which the primary colours red, green and blue were situated, and the compound colours in between them.
Newton founded the corpuscular theory of light, but Pierre Henry Fresnel (1788-1827) proved the wave theory of light. So did Thomas Young (1773-1829), who confirmed the wave theory with experiments on the interference of lightwaves. In the nineteenth century, several authors started to notice that colours changed when they were juxtaposed. Changes of colour due to fatigue of the eyes were noticed.
Michel-Eugčne Chevreul (1786-1889), a French chemist, edited a book in 1839, in which he treated the effects of the juxtaposition of coloured surfaces and the perceived changes in these colours when they were brought together. Chevreul worked much with the combination of colours to form other hues. Chevreul had worked with the notion of complementary colours and this notion gradually became one of the main interests of scientists and artists.
Hermann Grassmann (1809 – 1877) proved in 1854 that for every colour there exists another opponent colour in the spectrum which, when mixed with the first in the correct proportions, will produce white light.
Hermann von Helmholtz (1821-1894), a German physiologist, concluded in 1866 that the union of the impressions of two different colours to a single one is a physiological phenomenon, which depends solely on the reaction of the visual nerves. The title of his work was an indication that conceptions on colour were changing, as it was called "Physiological Optics" (1867).
James Clerk Maxwell (1831-1879) was a Scottish researcher. He studied electromagnetism and also wrote on colour. He decided not to use red, blue and yellow as primaries, as Chevreul and most other writers on colour had done, but red, green and blue. Maxwell and Helmholtz showed that a variety of spectral distributions of light could produce perceptions of colours, which are indistinguishable from each other.
Philipp Otto Runge (1777-1800) was a German painter of the early Romantic period. He corresponded with the poet Johann Wolfgang von Goethe on colours. With Philipp Otto Runge the search for rules defining harmony entered fully in the history of painting and of colour theory. Runge wrote a book on colours, called "The Colour Sphere", in which he showed a circle of six basic colours: yellow, orange, red, violet, blue and green. He varied these hues in tone and in intensity, and so came to a representation of all possible colours on a sphere. Indeed, to represent all variations of three parameters, one needs a representation in three dimensions, so with Runge also started a long series of three-dimensional representations of the variation of colours.
Runge thought that combinations of complementary colours were harmonious, combinations of contiguous colours were monotonous, and combinations of non-contiguous but at the same time non-complementary colours un-harmonious. In this last case he proposed using a third intermediary colour that could mediate between the two original ones. I will explain later what "complementary" colours mean, Arte.
Runge’s ideas on harmony were essentially close to Goethe’s perceptions of harmony, and the two men may indeed have exchanged ideas on this subject. Runge wrote also on the additive mixing of colours, and he stated that harmonious colours were those that created grey when seen from a distance and combined, which is indeed the case for complementary colours of the additive process. Harmony was for Runge the relief of tension between colours, in grey. As Runge found hues that were contiguous on his colour circle to be monotonous, he also found – contrary to Leonardo da Vinci – that the progression of colours in the rainbow was monotonous.

Zeuxis: I am talking too much, Arte, and I suppose I am telling now about confusing concepts. We will talk later of other writers on colours still.
I would like you to do something for me. I brought two books with me. One of the greatest and most unexpected writers on colour was Johann Wolfgang von Goethe. Goethe was a German aristocrat, a poet and a politician, so a very unlikely character to experiment with colours. I would like you to read his book. It is old, but Goethe showed many experiments. I just ask you not to believe him too much when he tries to explain why he saw the effects he saw.
And then, I have here a book of a painter, of Wassily Kandinsky. Kandinsky was a Russian, who eventually became one of the first abstract painters. He also talks of colours, and you will like to read about the emotions that Kandinsky felt to be associated with colours.

Arte: Thank you, Zeuxis. I will be a good girl and read your books. I will write what I think of this guy Goethe.

Zeuxis: Fine then let’s continue with our theory of light.
We think of light as being constituted of photons, packets of electromagnetic radiation. These photons fall on the retina of our eyes; they stimulate chemical molecules there, so that electrical signals are sent to our brain. These signals are transformed in our mind to a perception of colour. This kind of explanation is the easiest to work with, Arte.
Photons are particles, but it is a paradox of our physical universe that these packets of different energy can also manifest as electromagnetic waves.
That this dual nature of light is true can be proven by experiment.
We take a source of photons and release one photon after the other to send the photons at a screen in which a hole is bored. If we put a photographic plate behind the screen and hole, we will see a dot on the plate. That is quite normal, since we expect a stream of particles to pass right through the hole. We will not see a pattern of diminishing excitation on the photographic plate, such as the circular patterns of highs and lows of intensity of a wave. The absence of such patterns and the observation of a dot proves light in this experiment to consist of particles, of quanta of light.

Zeuxis makes a drawing in the sand of the lane. He draws plate 70 to illustrate the particle nature of light.

Zeuxis: In a second experiment two small holes are made in the screen. The holes do not even have to be put with one hole right in front of the photon cannon. The effect on the photographic plate will be dramatically different from the first experiment. Now, we will definitely see the interference patterns of two waves radiating from the two holes and combining their effects on the plate. This experiment proves the wave nature of light. It is as if the light "senses" another setting. By this experiment, only the quantum mechanical properties of the whole environment are changed, and since light is a property of our environment, the effect of light is different from the first experiment.

Zeuxis draws plate 71 to illustrate the wave nature of light.

Arte: Thus photons can be associated with a wavelength, and as well with the mass and other properties of a particle.

Zeuxis: Absolutely! Light, as a potential of the environment, is perceived differently, realises differently, according to the experiment staged! But let's not dwell too much on that. For our purposes, light is either a wave or a particle, and we will explain what we see, once by talking of light as waves, and then as particles.
As a wave, light oscillates in time, and the inverse of the length of a full period in time of this oscillation, a period that is endlessly repeated, can be expressed in our numbering system. The period is called the wavelength. The inverse of this period is called the frequency of the wave. Only light or photons with a wavelength of about 350 nanometre (nano means ten to the power minus ninth) to 750 nanometre stimulates our eyes and mind. Between these boundaries lies a full spectrum of waves of light, which shows in certain circumstances different colours, such as we can perceive them in a rainbow.
The first colours we perceive in a rainbow are shades of blue, then green and then red. Beyond the blue colour, into light of wavelengths we cannot perceive (that means our retina is not stimulated by them), lies the ultraviolet or UV radiation. Farther than red, with higher frequencies of wavelengths than we can perceive, is the infrared region of radiation or IR waves.
With our eyes, we only see all the shades between violet or dark blue and full red. Light of other wavelengths has no effect of colour to our eyes and mind.
Our skin absorbs more of the infrared rays than of the ultra-violet waves. Infrared rays therefore give us a sensation of warmth. Thus, humans have associated warmth with the colour red.
Within the visible spectrum of electromagnetic radiation only certain wavelengths give rise to certain visual sensations. The photons themselves have no property of colour however. Colours are representations in our mind of the stimulation of certain physiological phenomena in our eyes! Each photon of a different wavelength gives rise to one more stimulant, and these stimuli are even combined in our mind, so that stimulation by various photons give rise to the greatest variations of sensations we call colour.
When our stimuli are overloaded with photons of all wavelengths, we see the purest white. Absence of stimuli is black. In between we see shades of grey.

Arte: Zeuxis, now that we are talking about light, I think suddenly of something. Look at that man there, in the far. He smokes a cigar, and I just saw the colours of the flowers beyond the smoke change. We see light through the air, is it not, and not through the vacuum. Does that not affect light and hence also the colours of objects we look at?

Zeuxis, surprised, halts a while and then continues: It does, Arte, and not slightly! That was in fact an observation of which Goethe would have been proud, and not just Goethe! Let me explain.
Besides the corpuscular and wave properties of light and the direct effect of these on our retina, other physical effects indeed influence our perception, Arte.
Our eyes see light after the waves having passed through the medium of the air. Light may be modified on its way through the medium, affecting our vision of colours. The medium does affect colours. It does not so much change colours, however, when the distance between the viewer and a painting or between a painter and the model is small and when the air is pure. At longer distances however, for instance between a painter and a far landscape, and when the air is not pure, these effects become important and have been noticed since long.
Goethe observed that far mountains were seen in the blue colour. Before him, Aristotle and Leonardo da Vinci had studied this effect. Leonardo wrote in his "Trattato della Pittura" that distant mountains appeared blue and a more beautiful (deeper) blue in proportion, when the mountains were darker in colour. Aristotle said that the illuminated air between eyes and mountains was thinner towards the top of the mountains, and thus exhibited the darkness in deeper blue.
Goethe explained that the mountains were at so great a distance that we could not distinguish the real colours anymore, but merely a dark object, since no light was reflected from the mountains. For Goethe, this effect of mountains appearing blue was due to the interposed vapours, much as he had experimented with smoke between his eyes and light sources.
Let us look at a gas in which small particles are suspended. The colours we see differ with the angles of the source of light. When we put the source of light behind the gas, we see the light as red, because the particles in the gas absorb the blue light of the source. This effect explains the red hue of the sun at dawn and at sunset. When the source of light is however situated behind the viewer and projected onto the gas, when the viewer looks at a black background, a beautiful blue appears. The blue is reflected towards us by the particles of the gas. Absolutely clean air, or vacuum, should show us black. With a growing number of particles and with larger particles, the blue will become grey or even white, as with cigar smoke.
The colours of the sky are explained on this same principle, whereby the water vapour and the dust particles in the air play the role of the small particles of the gas. This physical effect also explains why we see the sky above us dark blue on a clear day, but lesser dark blue above the horizon. Indeed, the blue is more absorbed when we look horizontally, because we look through more particles than when we look straight up. And it explains why we see clouds in the sky white, whereas thunderclouds, heavier with water and close to rain, look grey to black.
Goethe observed at long the effect of light passing through a gas saturated with particles. This effect well explains why the sky is blue. The light of the sun comes directly at us. All waves but the waves equivalent to blue light indeed pass through the air unhampered and reach our eyes in straight lines. Blue light, however, is also dispersed by the particles in the air. The blue light interferes with the particles. It is reflected to all sides, and comes from us scattered from all sides to our eyes. Thus we see the whole sky in blue. So the same physical phenomenon, that is the scattering and absorption of blue light by the air, is responsible for the colour of the sky and for the red colour of the sun at dawn and at sunset.
When we look at far mountains, no colours come to us because of the distance. But blue is scattered in the air, so reaches us from longer distances. We see far mountains blue. The effect is thus responsible for the blue colour of far mountains and even for the colour of clouds in the skies. Painters knew the effect since Aristotle and of course by painterly tradition. If you look at old paintings of the thirteenth century, you will find far mountains in background landscapes painted invariably in blue, and not in black.
Painters and viewers are confronted with these phenomena of the natural elements seen through the air. At short distances, these effects are negligible. That means that the colours of a subject model that a painter is watching while working are not modified by the medium, as long as that is not saturated by impurities such as smoke. The air does not usually alter our perception of colours at short distances. The colours of a painting are those that a viewer perceives at distances of a few meters.
While we are at these colours of natural phenomena, we might as well explain why the night is dark. After all, in an infinite universe with an infinite number of stars, every line of our sight should eventually meet a star and thus light, so the sky should be very bright at night. The light of stars dim with distance, but this dimming should be exactly cancelled out by an increase in stars, as we look farther out. This is a puzzle called Olber’s paradox. It was solved in 1848 by no less than the poet Edgar Allen Poe. He argued in a poem that the stars had not had enough time to fill the universe with light. So if the sky is dark at night, this is so because our universe has not existed forever.

Arte: It is indeed marvellous, Zeuxis, what effects light can make on our eyes. You told me you would explain also how we see with our eyes. Well, how do we?

Zeuxis: All right. There are two phenomena to form colours: the additive process and the subtractive process. Let’s first talk about the additive process.

The additive process

Zeuxis: In our eyes are molecules that are basically tuned to the wavelengths of red, green and blue. The characteristics of sensitivity to wavelengths of these molecules overlap. Most wavelengths generally stimulate combinations of these molecules. The signals generated by the molecules stimulate positive and negative, inhibitory signals in the neurones.
Thus we humans say that the other colours but red, green and blue are composed of combinations of these three. When all molecules are equally stimulated, we see white.
Yellow is the impression in our mind of a photon of a certain wavelength, but we know that this is not really so. No molecule of our retina is stimulated by yellow (that is by light of a wavelength we perceive as yellow). This wave stimulates the "green" and "red" sensitive molecules. Since the wavelength stimulates only the capturing elements on our retina of green and red, we say that yellow is constituted of green and red in certain proportions.

Arte: Isn’t that like in television sets? We learned about that in school. In a certain kind of television sets, I believe our teacher called them shadow-mask televisions, the electron guns behind the screen shoot electrons at small luminescent dots of green, red and blue phosphors on the screen. The dots transform the energy of the electrons into light. They emit light of a certain wavelength; thus they show a certain colour. That light of various wavelengths, hence colours, are combined in our eyes. The dots are so small that we only see the combinations of various luminosities of green, red and blue. Thus many combinations of colours can be obtained, but not all.

Zeuxis: Now you are teaching me, Arte. You are right!
All the colours of the natural spectrum and many more can be constituted by varying the intensities of the light of the sets of these three dots. Of course, one has to look from a distance so that the green, blue and red merge their lights, and so that one cannot perceive the individual dots anymore. Two combinations concur here. We indeed combine the three R, G and B colour tones in some ratio to obtain other colours. And we can vary the intensity of each colour to obtain still different colours and light and dark tones. The dots on the television screen have a precise colour. Suppose that we could change that colour tone (not possible on a screen) by adding more or less white to the dots. Then we would have three effects combined. The three effects are called colour hue, saturation (adding more or less white light) and tone or brightness (intensity of light). The number of colours that can thus be obtained is staggering, millions of different colours!
We will henceforth talk of pure colours as the colours of the rainbow. With these fully saturated colours, some of the combinations are also familiar. Red and blue together give magenta. In the additive process, colours produced as sensations on our retina, produced by light stimuli, add up and combine. In the additive process there are three colours that can thus form all others: green, red and blue. The addition happens in our eyes and brain.
We can represent the additive combination of colours schematically as a triangle.

Zeuxis draws plate 72.

Zeuxis: Here is a schema of the colours we talked of, a simple and first triangle of colours. Green and blue added give cyan; green and red produce yellow; red and blue produce magenta.

Zeuxis: White is in the gravity point of the triangle, and white can be formed either by a combination of green, red and blue, or by an infinity of combinations of two colours, some of which are indicated by the gravity lines of the triangle. For instance, yellow and blue can combine to white, as well as green and magenta, cyan and red. Two colours that in certain proportions form white are called complementary for the additive process. Thus red and cyan, green and magenta and blue and yellow are a few and the best known complementary colours.

Zeuxis draws circles as plate 73 to illustrate the effect in our eyes when light spots of different colours add up:

Zeuxis: In the art of painting, the additive colour process will take place in our eyes whenever colour strokes or colour dots on the canvas are small enough to give only one sensation of colour in the viewer. The effect changes with distance, and thus with the position of the viewer.
A viewer who stands at a greater distance from a painting will more readily see the additive process, as the colours of very small areas combine. When the viewer comes closer, he or she will perceive the colours of the individual areas separated again.
The Impressionist and Divisionist painters exploited this effect. These painters brought small brushstrokes or small dots of colour on their panels. Seen from a distance, colours of adjacent dots or stokes combine in the eye of the viewers in an additive process. The dots can be larger or smaller. With small dots, the distance for a viewer at which the additive process comes into effect may be less than half a meter. With larger dots, the viewer may have to take more distance.
Claude Monet (1840 – 1926) particularly used larger juxtaposed brushstrokes of different colours, so that a viewer needed to go to look at the painting from farther away. Advancing towards the painting changes the patterns a viewer perceives. This is why in some of Claude Monet’s paintings a viewer sees only abstract colour patterns from close by. From a distance of two meters the viewer will start to discern for instance a tree branch with blossoms.

Arte: That is not too difficult, Zeuxis. I thought colours were a more complex matter.

Zeuxis: Well colours are complex, Arte. We are only at the beginning of our theory. Now it gets more difficult. There is a second process of constructing colours, the subtractive process.

The subtractive process

Zeuxis: The process by which colours of paint, the colours of the dots or small brushstrokes are formed, is by the subtractive process. The way these colours combine on our retina is by the additive process.
A wooden panel or a canvas has no inherent light source such as an electron gun behind it. It receives white light from the outside. Natural white light of the sun contains many wavelengths. The paint on canvas receives this light, lets it pass a little, and then absorbs light and reflects some. An area covered with red paint thus absorbs all the wavelengths of the white light but red, and reflects the red towards the viewer. This process of absorption of photons is called a subtractive process. An area of red paint on paper or canvas absorbs all wavelengths but red and reflects only the wavelengths of light that correspond in our eyes to red. Therefore, the colours are formed by a subtractive process; we see them in the additive process.
In the printing process one could, like on a television screen, print very small dots of green, red and blue colour printed very closely together. These dots would be seen from close by to be indeed red, green and blue, so reflect light of wavelengths corresponding to red, blue and green and absorb all other. Seen from a distance a viewer would see once more in an additive process the combinations of these colours, that is white. Printing in fact uses still another process, a process of transparent paints, but that is not the subject of this text. In the previous chapters I talked about the particular technique of the Divisionist painters.
I need to remind you, Arte, that indeed an additive process is the reason for the combination of colours in our eyes. But at the basis of the existence of the individual colours in the dots on the canvas, that is the colours seen at close distance, lies the subtractive process.
Painters use large painted patches, and thus they use the process of subtractive colour mixing, which involves the absorption and the reflection of light.
Painters mix colorants, pigments or dyes. These pigments absorb colours and reflect other colours. A yellow pigment absorbs blue and violet light but it reflects green and red light and the green and red additively combine to produce yellow in our eyes.

Arte: You told me earlier that the additive process works on red, green and blue. So we may expect that the subtractive process would work on the absorption of red, green and blue.

Zeuxis: And indeed, it does, Arte. But now it gets really difficult and confusing!
The colour of an area of paint that absorbs red light but transmits all other is cyan. An area of paint that absorbs green will reflect blue and red light, which will combine in our eyes to magenta. A paint that absorbs the blue primary will reflect green and red; thus it’s perceived colour will be yellow. Thus, the subtractive primaries are cyan, magenta and yellow.
The additive process and the subtractive process thus lead to different basic colours. The situation is rather confusing.

Zeuxis draws plate 74 to show the subtractive colours.

Arte: So, the various paints that I could use all absorb most of the light except the light of one particular wavelength, and reflect that light to us?

Zeuxis: Right! And since you mention paints, Arte, the paints that painters use come mostly from mineral pigments. The paints used in oil painting, in tempera painting or in fresco painting were made on the basis of mineral pigments. Nowadays painters can buy paints prepared in metallic or plastic tubes and these paints can be as well organic as mineral compounds. For centuries, painters had to rely mainly on mineral pigments grounded to fine powders. They had to buy lumps of earth or rock, grind the minerals and mix the pigments with oils. Mineral pigments are insoluble, so the fine granules stay suspended in the medium or vehicle in which they are mixed.
For tempera painting, painters took the yellow core of chicken eggs. They gently squeezed out the egg yolk in a glass and mixed that with a little water. Then they mixed on the palette the yolk with the mineral powders, before applying the paint on a panel or canvas. This technique of tempera painting produces colours that are generally soft. Tempera painting is done in slight and short brushstrokes, because egg yolk does not flow so well and needs to be directed by the brush entirely. It also rapidly dries out, so that drawing long brushstrokes and slowly smoothing out paint over an area is difficult.
In oil painting, painters first mixed the colour powders with linseed oil, which allowed smoothening the surfaces of paint more. Oils better preserve fluidity of the paint, so that areas could be covered more evenly. When painters used poppyseed oils, their brushstrokes could remain a little more visible, not unlike in tempera painting, and also a sense of texture was obtained, since the paint agglutinated more.
In fresco painting, the pigments were mixed with a white and almost translucent plaster, and then applied on a plastered wall. I am digressing, Arte … what were you saying?

Arte: Oh, you answered my question, Zeuxis. What are the pigments that produce the nicest colours?

Zeuxis: Mineral pigments have various origins. The word "mineral" means that the basis atom that interacts with light is a metal. Metals crystallise with other atoms and molecules, to absorb and reflect light selectively.
For instance, for yellow to red colours natural earths or ochres of different compositions exist. These can be simply earths, rocks or sands containing iron oxides. The reddest earths contain hematite. The yellow ochres contain goethites. The yellow ochres also become red when heated, whereas hematite ochres take on more violet hues when heated. Siennas are orange-brown such earths.
Painters bought the mineral pigments in shops and grinded the pigments down themselves, then mixed them with egg yolk or oils.
Yellow colours also come from lead oxides. Masticote is a compound of lead and tin oxides, litharge a pure lead oxide. Yellow pigments also were the chromium minerals, which can tend to green too. Various cadmium compounds also produce bright yellow colours, to orange and red hues. Golden yellow comes from the pigment called orpiment, found near hot springs, which contains arsenic trisulfide. That was a highly toxic material, and it was abandoned except in a very fine-grained version called "King’s yellow". The cadmium pigments replaced orpiment.

Arte: And the other hues?

Zeuxis: A red hue has also the mineral called realgar, which contains arsenic disulfide, and was often found together with orpiment. But realgar deteriorates in light. Red also is cinnabar, which is a mercury sulphide and the most common mineral of mercury. It was mined in Almadén in Spain
Green hues can be obtained from copper minerals. Malachite green is a copper carbonate. Also copper chlorides and copper silicates were used for green. There were green clays, glauconites and celadonites. Celadonites give a bluish green colour, and much of that was for instance found near Verona in Italy, hence galled Veronese green. Glauconites and celadonites are mica minerals, iron-rich clays that contain aluminium.
Alexandrian blue was a mixture of other copper carbonates. It was a blue pigment, much used in ancient Egyptian and Roman paintings.
To obtain very fine blue hues, painters used a semi-precious stone mainly found in Afghanistan, called lapis-lazuli. Lapis lazuli contains sodalites, sodium aluminosilicates, also called feldspathoďds. This colour pigment was called ultramarine. It was an expensive colour, so painters also used the cheaper but not so beautiful azurite. Azurite was a basic copper carbonate, also called chessylite. Indigo was a material that yielded a nice purple-blue. This was not a mineral; it was extracted from a plant that mainly grew in India. So not all the pigments were found in nature as rocks or earths.
Mineral compounds can also be artificially prepared by chemists. Prussian blue, for instance, was discovered by a German chemist; it was a dark blue made from cyanides, potassium and iron. Lapis lazuli was made artificially in the beginning of the nineteenth century from china clay, sulphur and sodium carbonate with small amounts of silica and pitch.
Blue and green pigments are now made from phthalocyanines, and these are organic pigments, although they are still copper compounds in pigments.
Black could come from simple lampblack or from manganese oxides.
The whitest white colours, Rembrandt’s whites, come from oxidising lead, and also natural white clays could be used. Nowadays, titanium dioxides are used.
One could make brown hues by mixing the manganese oxides with yellow or orange to red ochres. But some ochres are naturally brown too, and then called umbers.
Many of the pigments that painters can find now pre-prepared in tubes, are artificially constituted mineral compounds or organic compounds. These organic compounds are the products of organic chemistry, the chemistry of hydrocarbon molecules.
From the 1960s on, acrylic paints were introduced. These are based on synthetic acrylic resins, in which the pigments are mixed. They dry rapidly and have a nice brilliant transparency, cover well, and are very resistant.
All kinds of artificial oils are also applied in the paints.
Mineral pigments are often toxic. They contain heavy metals, and some of those are dangerous to health. Arsenic, lead and mercury are very toxic to humans. Even in paints they may slowly poison a painter, especially when the painter uses his fingers and hands to apply the paint instead of the brush or other utensils like knives. These pigments are therefore avoided nowadays as much as possible.

Arte: So I should not eat paint.

Zeuxis shocked: By Zeus, no, girl! I have seen so many of my colleagues waste slowly away without knowing why. But the good painter keeps his or her hands clean! A good painter has no need to use fingers, and he or she doesn’t drop paint on hands and arms. A good painter is a clean painter! And always first read the warning on paint tubes or paint boxes!

Zeuxis: The two processes, the additive and the subtractive process are the two basic processes of colour formation. The subtractive process produces individual colours in paints and when the painter places small dots of paints together, very small and very closely together, the colours interact in our eyes by an additive process. It is somewhat confusing, but that is how nature works!

Arte: Yes, Zeuxis. We did not really need two principles to complicate matters like this. But I understand the two different origins: one a mixing of sensations in our eyes, the other a mixing of pigments or paints giving rise to one sensation that reaches our eyes.

The painter’s palette

Zeuxis: And it is even more confusing. Painters have changed the whole idea!

Arte: How can that be, Zeuxis? These are quite fundamental physical truths, it seems to me!

Zeuxis: Yes, Arte, but painters are humans, and humans tend to be stubborn, and self-cntred, and they tend to individualise and to complicate matters ever. Painters had their own view on colour, which was based on their instinct and intuition of colours that went well together. So, unsatisfied with the additive process, which they did not really know and understood in early times, and used to work with paint and therefore forced to admit the effects of mixing paints but yet also unsatisfied with the results and with the strange hues that were the basis of this process, they followed another road!

Arte: Why was that, Zeuxis?

Zeuxis: Because in painting artists talk of a third way of handling colour combinations, which although based on the subtractive process, has been so to say polluted by the principles of the additive process.
Painters use paints, and pigments produce colours by a subtractive process. The real primaries of subtractive colours as we have explained are cyan, magenta and yellow. But these are commonly called otherwise. Cyan, normally a blue-green, is usually taken as blue, whereas magenta is taken as red. Then the primary subtractive colours for painters become blue, red and yellow.
This third way of looking at colours deserves a name too. But its basis is not a physical process. It is more a choice made by painters of a basic set of primary colours that seem to go well together. We might call it a set of colours that painters believe to require each other in harmonious combinations: subjectively determined primary and complementary colours. These subjectively determined primary colours would be red, yellow and blue.
On the other hand then, we would have objectively determined primary and complementary colours, which would be the colours that are primary and complementary according to the additive process: red, blue and green.

Arte: So painters constituted their own palette of colours?

Zeuxis: Yes! What we have absolutely to well understand, is that these "painter’s" colours are not based on any physical or physiological process (even not really the subtractive process though they might have been derived from this) but purely on matters of artists’ taste.
Wassily Kandinsky for instance based his whole treatise on colours in "On the spiritual in art" on red, blue and yellow. And so have done most other painters until our days.
Today, we understand why Kandinsky used these three colours red, blue and yellow whereas in reality he might have better spoken of cyan, magenta and yellow.
The subtractive and additive processes of forming colours are very different, as are the effects of their processes. These processes were not known to be distinct processes before the middle of the nineteenth century, and even after that time scientists and artists often confused understanding of the two processes.
For the painter, this means that the laws of mixing paint and obtaining colours therefore are the laws of the subtractive process.
Mixing paints means usually more absorption of light by the resulting paint, thus less brilliant hues. Mixing pigments or paint of the three primary colours of the subtractive process equally together does not give white as in the additive process. It will yield a dark, dull grey, or a muddy brown, and if the process of mixing were in perfect accordance with theory, it should give black. Painters have not mixed so many colour pigments or paints together in order to obtain the hues of their paintings because of this difficulty. They usually prefer pure colours of one pigment only, maybe added with white or black. And when painters did mix paints, they preferred to mix bright coloured paints so that not too much light be absorbed. Viewers can notice how certain pictures are dark in overall tone while other may have a dark background but remain in very bright hues in the main subject.
Pigments used by painters also vary in strength in mixing. Paints or pigments of high tinting strength will dominate mixtures on the painter’s palette, whereas others will have less significant effects. This is an issue for painters. They will have to take care with which paints they mix to obtain a certain hue on the canvas. But this is less an issue for viewers.
When a viewer looks at very large areas of colour on a canvas, the colour perceived is the result of the subtractive process of the paint and the viewer will perceive that one colour. When a viewer stands very close to a picture on which many dots or strokes are juxtaposed, the viewer will perceive the individual colours of the dots in the same way and these colours are equally the result of the subtractive process of the paint or pigment. When, however, the colour areas are small and when the viewer looks fro ma distance, the additive process comes into play in a more pronounced way, to change what we see of the colours.

Zeuxis now also draws plate 75 to illustrate the painters’ choices of primary and complementary colours:

Arte: That was indeed complex, Zeuxis. It did teach me however that the ways painters can make colours are very many. There are slight but important differences between the colours as ordered according to the subtractive process, and as ordered according to painters.
But look at those blue flowers there. Some of them show the full colour blue, others have a fainter colour blue. How does theory explain these differences?

Zeuxis: In our colour theory we do not talk of "full" and "faint", Arte. We use other names, and define the properties of variation of one colour more strictly. Painters talk of hue, tone and intensity. These terms, however, and their definitions, have changed in history. Different authors used and still use other definitions for these notions.

Hue

Zeuxis: The painters’ basic (subtractive) hues are red, yellow and blue (we know it should be magenta, yellow and cyan but we grant the names red, yellow and blue). When these are mixed, then the secondary hues of painters are formed, which are orange, violet and green. The tertiary colours are obtained when the amount of the primary hues is increased. Thus we obtain yellow-orange, orange-red, red-violet, violet-blue, blue-green and green-yellow. These colours can be represented in a circle or on a wheel, as I have shown. Families of colours, usually denoted as harmonious colours, lie closely together in that wheel. Colours on opposite sides are complementary colours.
Black, white and greys are said to have no hue. These colours will almost never be presented on colour wheels. They are also called achromatic colours.

Tone

Zeuxis: The tone is a colour’s relative degree of lightness or darkness. A painting in dark colours is said to be in low tonal key. A painting in pale colours is said to have a high tonal key.
Painters have indicated shadows on draperies, for instance, by painting lower tonal colour hues (darker tones) in the parts of the cloth that should receive less light. These colours are then situated next to the paler hues (lighter tones) of the same colour in the parts lighted by the sun.
More complex perceptions are at play, for each hue has also a tonal value in relation to other hues. Pure hues do not all have the same tone. For instance, orange is lighter than red, and violet is darker than green. The tone of yellow is much brighter than the tone of purple. The sequence of the above colour circle is the most natural one. Changes in this order are often disturbing.
Differences in tone are very difficult to perceive and people have different impressions of this characteristic.
Here follows one possible ranking of pure hues according to tone (see the colour wheel): yellow, yellow-orange, orange, orange-red, green-yellow, red, green, red-violet, blue-green, blue, blue-violet, violet.
Thus, yellow is the brightest, most luminous pure hue. Adding white can brighten it further. By adding black, we darken the tone but also dull the hue, thus also change another property of colours called "intensity" (see next paragraph)
When one darkens a light-toned hue, for instance by adding black to yellow, one obtains other colours that are called discords. Discords are also called colours obtained from dark toned hues that are lightened by adding white, like for instance when adding white to blue.
Other names given to this notion are value, lightness and luminosity.

Intensity

Zeuxis: The intensity of a colour then is its degree of hue saturation. The colour vermilion of which Kandinsky often wrote, is orange-red at high intensity. The brown earth pigment called sienna has less orange-red saturation. Intense hues are called chromatic colours. Colours of less intensity are then hues in which either white or black is added. The resulting colours are called "tints". So tints are derivations of a pure hue.
Another word often used for intensity is "saturation".
Black and white, as well as greys, are the truly achromatic colours. By adding white, the colour fades, and this is loss in saturation. So, intensity is a degree for how bright or dull a colour is. All pure hues are also fully saturated, and thus have maximum intensity. But even among the pure hues, intensities seem to vary. Thus yellow is perceived as being more intense than violet. Similarly, red is more intense than green. So we also might think of a scale of intensity among the pure hues, with yellow most intense. A ranking is almost impossible here, but we do view orange as more intense than blue, red more intense than green, and yellow more intense than blue.

Primary colours

Zeuxis: I talked about primary and complementary colours earlier. This is an interesting and strange concept, worthy of some more information for you.
In all the theories of colour and colour contrasts, the concept of primary colours is central. A few scientists did not really rely on the concept, as they probably found it fragile and unproved, but most researchers and writers do depart from certain colours that they took as fundamental, basic colours, from which the others were derived. However, very little consensus has been built in history over just which these primary colours could be and over just how many should be used as basis.
Red, green and blue were the primary colours for James Clerk Maxwell, Ogden Rood and more recently Frans Gerritsen. These were all physicists.
Also three primaries, but then red, yellow and blue, were chosen by Wilhelm Ostwald, Johannes Itten, Josef Albers, Faber Birren, Moses Harris, Johann Wolfgang von Goethe, Philip Otto Runge and Michel-Eugčne Chevreul. Most of these, but not all, were painters.
Harald Küppers also privileges three primaries, but proposed to take orange, green and violet as basic primaries.
Ewald Hering and Leonardo da Vinci preferred four primaries: red, yellow, blue and green. These could be the colours we intuitively would name as the ones we can best distinguish and know in our normal lives.
Albert Munsell proposed to use five hues: red, yellow, blue, green and purple.
We know that Newton thought or chose to think that the rainbow was constituted of seven colours.
And of course, we should add white and black to all these as fundamental colours.
So, when it comes to primary colours, opinions have differed. Most writers accept three colours plus white and black as the primary colours. From this analysis, the most plausible and natural choice goes to only three colours, to red, green and blue or in more precise terms to agree with Harald Küppers on orange, green and violet as more scientific. But John Gage wrote, "’Basic’ sets of ‘simple’ or ‘primary’ colours are a great gift to structuralists, but offer little comfort to those of us who are concerned to interpret the use of colour in concrete situations" ^G97 .

The creation of colours by varying the colour qualities

Arte: That makes the comprehension of colours pretty complex, Zeuxis, because we can choose from a limitless variation of hues with tones and intensity.

Zeuxis: Yes, Arte! The combinations of colours are endless since green, blue and red can be combined in various proportions, and each with varying tones.
Equal proportions of red and green yield yellow. But two proportions of red plus one proportion of green yield orange. Two proportions of green plus one red gives a colour called lime. A blue, a green and four proportions of red give brown. And so on.
The endless combinations of colour hues, tones and intensities of colours is also an endless source of variation for painters. Painters can alter the hues by choosing another colour. They can heighten the intensity by choosing a more saturated hue. They can add layers of the same paint to darken the tone. And they can dilute the saturation of their colours by adding white, or by dulling the hue by adding black paint.
The visual system of humans can in theory distinguish between millions of different colours, but we have only names in daily language for a few tens of colours.
We may talk of secondary colours as of colours obtained by mixing primary colours. Tertiary colours then would be colours obtained from mixing a secondary in unequal proportions with a primary. And we can continue, as quaternary colours would be a mixture of a tertiary colour and a primary, and so forth.
Primary colours are more readily recognised by viewers, so that they tend to dominate a picture. Areas of pure hues powerfully attract the attention of viewers. We can find these effects very strongly for instance in the paintings of the English artists of the Pre-Raphaelitic Brotherhood, like in the pictures of William Holman Hunt.
Secondary colours are more discreet, are often used for backgrounds or for pictures in which the dominance is granted to the content or the lines or the structure of the forms. Secondary colours support the mood of a painting.
Tertiary and quaternary colours seem artificial, rare, expressly chosen and thus less natural in a painting. They will create feelings of unease, and thus will contribute to feelings of tension in a picture.

Other qualities of colour

Arte: So now I know all there is to colours, don’t I?

Zeuxis: I am afraid not, Arte. There are still quite other features of colour to charm you!
Colours can be objectively and physically characterised by hue, tone and intensity, as I explained. But two other features can be attributed to colour, which are not physical but psychological characteristics. These are the spatial value and the temperature value of colours. We will talk in later chapters more on these psychological values of colour. These features are for us, humans, as viewers, as important as the physical characteristics of colour.

The spatial value

Zeuxis: Colours like blue and white seem more distant to us than the colours orange or red. The spatial value of colours is correlated with hue. Blue or violet diluted colours, that is colours of less intensity (in which white is added), seem farther away than colours that are highly saturated. Dilution indicates distance. Let me show you.

Zeuxis looks around and since nobody is in sight at this hour, he makes his magic screen appear, and he projects a painting on it.

-> Maria Elena Vieira da Silva (1908-1992). The Theatre of Gérard Philippe. Le Musée d’Unterlinden. Colmar. 1975.

Zeuxis: It is no wonder that pictures such as Maria Elena Vieira da Silva’s "Theatre" were painted in black and blue lines that delineate the white surfaces. The blue lines induce feelings of space in viewers, as orange or red lines would not have been able to evoke.
Johannes Itten noted that yellow seemed close to viewers, blue kept a distant to the viewer. Itten argued that in the spatial value of colours, the colour orange is interposed between yellow and red, red between yellow and violet, green between yellow and blue, and so on.
Blue and violet always represent distance, the far. Such features of colours have of course to do with our perception of nature, and the psychological knowledge that we have of our world. The cosmos is blue or violet, far mountains are blue; these colours thus have become irrevocably related to the concept of distance in our mind. Even black, that most dead of all colours, has a spatial value of seeming to be more close to us than the blue colour, and this effect was exploited by Maria Elena Vieira da Silva for the black lines, representing maybe the curtains of the scene, look closer to us than the blue background lines. Imagine this painting in other colours. Then you will understand why this painter used blue. No other colours could have better created feelings of space.
The blue colour definitely creates a distance between the viewer and the subject. This quality has not just been used to denote physical distances, but also to enhance – or diminish - effects of empathy between a viewer and the figures in the painting. When a painter wants to create privacy, a distance of emotions between his figures and the viewer, he or she will return to blue or violet hues.

The temperature value

Zeuxis: The second other property of colours that I wanted to talk you about, Arte, has to do with feelings of warmth or coldness.
Colours can be perceived as being cold or warm. Johannes Itten called red-orange the warmest and blue-green the coldest colours. Combinations and gradations of these yield paintings based on cold-warm antagonisms of colour.
Cool hues are blue, blue-violet, yellow-green and blue-green. Warm hues are for instance yellow, yellow-orange, orange, red-orange, red and even red-violet.
This perception has a very relative value. A red spot on a black background gives a warm feeling, and it will radiate its warmth. The same red spot on a white background will look darker and will less radiate warmth, as it seems to be dulled, subdued. Red used in conjunction with other hues creates different feelings of temperature. The temperature feeling of a hue thus varies according to the ambient colours. A warm colour surrounded by a cooler one will appear warmer and vice versa.
We can sense some of these properties, as well of temperature as of spatial value, in old religious, Christian representations. Jesus was always a figure of love, as this was his main message. Therefore, painters usually depicted him in warm colours, in a red robe or cloak. As a result, Christ was perceived as a figure close to the viewer, close in compassion and sympathy. On the other hand, his mother, Mary, was the Immaculate Virgin and a highly respected Saint who had to impersonify purity. From the sixteenth century on, painters dressed her in a white robe and the blue cloak or maphorion. Her dress in early Byzantine picture was usually a dark red and her maphorion was dark red. In later centuries, however, her maphorion was painted in ever-lighter blue, a colder colour that was a symbol of her distance to humans as a Virgin and Saint.

Arte: I am indeed going to call my tortoise after you, Zeuxis. I am becoming puzzled. There are so many things to learn about colours. You explained me all the physical effects and characteristics of colours. Is that then the whole theory?

Zeuxis, sighing: I am afraid not, Arte. We have not finished by a long way! Until now we indeed mostly talked about the physical features of light to explain colours. As I said earlier however, we see by our eyes, and the properties of our eyes also influence the way we perceive colours. Let me just explain a little about these effects. First, I have to explain how complex the eye is.

Zeuxis: Light strikes our eyes. The cornea and lens of our eye focus the image onto the retina, the light-sensitive part of the eye. The iris adjusts in width to partially regulate light levels. The fovea is the central focal point of the eye, the place where the images are cast. There is a filter in front of the fovea, called the macula, which serves to limit the damage that might be done to the fovea by the too intense light sources such as when we look straight at the sun. The retina is linked to the primary visual cortex via a structure known as the Lateral Geniculate Nucleus. The retina produces wavelength-sensitive stimuli that are processed by different areas of the brain cortex. The optic nerve is the bundle of nerves, which carries the visual signals to the brain. Thus the eye adapts to various levels of brightness, correcting for too high levels of lighting energy to a certain extent.
Scientists and artists gradually became aware of the importance of the physiological effects in colour perception. Modern neurobiology emphasises the physiological origin of colour.

Arte: Can you give me an example of such an influence, Zeuxis?

Zeuxis: Well, Arte, there is the Purkinje effect. I already showed you that before. Now I can explain the effect.
Johannes Evangelista Purkinje (1787-1869) was a Bohemian (Czech) physiologist and a Professor at Prague University. He described in 1825 a strange physiological effect on colours, which bears his name.
He proposed to look at two brightly lit colour surfaces, red and blue, so that one obtains the impression that both surfaces are equally bright. This last is already a very subjective feature, so that for different persons the brightness or tone of the surfaces needs to be adjusted to obtain the same effect. But, suppose we have done that. Now one diminishes strongly the light that falls on the surfaces, for instance to one-thousandth of the previous light. The result will be that a viewer, who before saw both surfaces equally light, now sees the blue much lighter than the red. As we said earlier, that is purely an effect of our eye!
The effect of Purkinje is important for the lighting of paintings in museums. We may perceive colours differently, according to the level of brightness in museum rooms. Especially aquarelle pictures, and also crayons drawings, and pictures in chalk, are often exhibited at low level of lighting in order to preserve the colours. In doing so, we may perceive the colours differently. And of course if this light is not pure white, the effect on the rendering of colours is even more different, but that is due to still other effects.
Another example is that of the red tulip in the green grass. At sunset, when light is much diminished, the tulip that was so bright in daylight is now almost black amidst the green background. So the red colour is very dark and its hue is almost not perceived. At dawn, the first hues that humans distinguish in the new light are the blue colours.
The reason for this effect of Purkinje is that there are in the retina two types of light-sensitive molecules: the rods and the cones. The rods only get activated at very low brightness levels; the cones work at high brightness. Seeing with cones and rods proceeds according to certain characteristics of sensitivity to wavelengths and wavelength differences. When one chooses two colours for which the cones have the same sensitivity, then this sensitivity does not remain the same at lower brightness - when the rods come in. The rods, used in the dark, are only sensitive to brightness levels, so that in the dark we only see shades of grey.

Arte: That is really surprising. So we see in fact only in three colours? You must tell me more about the cones and the rods!

Zeuxis: It seems so, Arte. But remember, scientists are only at the beginning of really understanding human vision. They call their theories now trichromatic theories.
The theory of colour perception emphasises the physiological origin of light perception.
The first theory to explain successfully the effects of colour vision was proposed by Thomas Young around 1801-1802. Young thought that the retina could not have a different colour receptor for every wavelength of light. He proposed that colours were perceived by a three-colour code. This is called the trichromatic theory.
According to this theory, there are two different sets of photoreceptors in our eyes. One set operates during the day. This is constituted of the cones, so-called after the physical form of the molecules. The cones mediate colour vision. They are highly concentrated in the fovea, but the fovea is less sensitive to light than the surrounding retina.
Humans have three types of cones, which are differently sensitive to wavelengths of light. There are short-wavelength sensitive ones or S-cones, middle wavelength sensitive cones or M-cones and long-wavelength sensitive ones or L-cones. So we perceive colours through a code of three signals. Our colour vision is derived from comparisons between the amounts of light received by each type of cone. These comparisons occur in the retina, but continue to be processed in the cerebral cortex of the human brain.
The other set of photoreceptors operates in the dark, and this set comprises the rods. Rods mediate night vision, at very low levels of illumination, and then we only see shades of grey. The rods function only in dim lighting. During the night we see shapes quite well, but we will discern no colours. Rods can be excited by a single photon, so they are very sensitive to low levels of light. They also outnumber the cones in a proportion of more than ten to one.
These cones and rods are chromophores, vitamin A derivatives, bound to a membrane protein called an opsin. Light triggers the isomerisation of the chromophores and the opsin. When light is absorbed by these photoreceptors, the light energy is converted into electrical and chemical signals that are processed by the neurones in our eye and brain. Other theories but the trichromatic theory have been studied and proposed.
The German physicist Ewald Hering (1834-1918) proposed another approach in 1878. He proposed that colour vision indeed is based on three stimuli for the brain, but that these stimuli were each a pair of opponents, that is the opponents between dark and light, red and green and blue and yellow. Colour vision was an effect of excitation and inhibitory effects in various successive layers. Specialised red-green opponent cells in the second layer compare signals from the cones in the first layer. These opponent cells compare the proportion of red and green light coming from an area in our visual focus. Then, other opponent cells compare the signals from the blue cones with the combined second-layer signals of the red-and-green cones.
This theory is called the Opponents Theory, and several of the special effects of vision like contrasts and after-images can be better explained by this theory. For our observations and conclusions, the trichromatic theory will suffice however. Moreover it seems that the trichromatic and opponent theories are not necessary mutually exclusive.
It is currently believed that there are two levels of neurones in our mind, and signals from the green and red cones in the first layer are compared by other "opponent" green-red neurones in the second layer. These opponent cells compute the proportion between red and green light coming from a particular part of our vision. The three cone types as described in the Young-Helmholtz theory send signals to the opponent cells of the Hering Theory. The S-cones (blue cones) send excitatory signals to the blue-yellow opponent cells. The L-cones (red-yellow) send inhibitory signals. If the strength of the excitatory signal is greater than that of the inhibitory signals, then yellow is seen. The red-green opponent cells receive excitatory signals from the M-cones. Green is seen if the strength of the excitatory signals exceeds the strength of the inhibitory cells. And red is seen if the strengths are the contrary.
Several modern, so-called "Zonal" theories, have been developed, encompassing both older theories.
Different animals have different numbers of receptors, and their colour vision is therefore very different from ours. Even the animals that have the same number of eye-receptor sets, that is four, have different vision because the wavelength sensibility of their vision receptors differs. Moreover, some animals are receptive to the polarisation of light, whereas humans are not. Nothing indicates that the composition of the colour perception is a linear function of the stimuli of the three kinds of cones of the retina. These functions may be non-linear, and they may differ among animals and humans. The colours that we see are thus not merely a property of the object we see, but much more a physiological property of living beings.
Current trichromatic vision theory says that sensors in the eye capture light. Light is captured on sensors by a spectral distribution of wavelengths of light. The spectral response of each sensor is a function of sensitivity for a particular wavelength and of the energy distribution over the various wavelengths, the energy of the illumination. A colour is then perceived through a combination of the responses of the three sensors of the human eye to external stimuli on the sensors. And our vision can adapt to a certain extent to various levels of brightness.
The modern "Zonal Theory" of vision assumes that there are several stages in colour vision. In the first stage, light is indeed captured and the stimuli are transformed into electrical signals. Here the trichromatic model works entirely. But then these signals are processed in a visual network of neurones to produce three new signals, one achromatic (bearing no colour information) and two antagonistic signals bearing information of colour. Judd developed a mathematical model for this process.
In this way, modern science came closer to explaining many of the effects and puzzles of vision.
We have seen that with the research into colour since the middle nineteenth century, the awareness grew that the original view of philosophers stating that colour was not an intrinsic quality of an object, was a true proposition. Colours changed under various circumstances that had nothing to do with the object itself. And scientists found that animals see the aspect of a surface of an object that we call of a certain colour very differently.
More than anything, colour as we see is a quality of humans. Colour as a concept exists because of humans. Colour exists because humans exist. The early scientific theories of colour as purely a physical phenomenon are not the right models of colour perception.

Zeuxis: That is why, Arte, I asked you to read Goethe’s book! Goethe was right when he sensed that there was much more to colour than a physical phenomenon, and on this notion I will continue to explain you more about colour, but that must be in next lessons. I am tired, and we have to leave the exhibition now. Please read your book on colours of Goethe.

Arte: I will, Zeuxis, though I have the impression you have taught me all there is to know about colours by now.

Zeuxis: Oh no, Arte, there is still a lot more to come! Colour perception is among the very most difficult puzzles of mankind.

Back to the navigation screen (if that screen has been closed)

Last updated: June 2010

Book

Copyright: René Dewil - All rights reserved. The electronic form of this document is copyright. Permission is granted for electronic copying, distribution in print form for educational purposes and personal use. If you do reduplicate the document, indicate the source as 'René Dewil - The Art of Painting - Copyright'. No permission is granted for commercial use and if you would like to reproduce this work for commercial purposes in all or in part, in any form, as in selling it as a book or published compilation, then you must ask for my permission formally and separately.