Illustrated books


Color Castle © 2000 David Kidd

Color Wheels

Everybody draws a color wheel differently: the other way round, or upside down, but every wheels contains the same truths:

In one half are warm colors, the other is all cool colors. This is proved by studying yellow and violet, the turning points. Is this yellow slightly orange, or is it slightly lemon? Is this violet getting blue or getting red? When the samples are neither warm nor cool they are the exact turning points that determine the division.


How can you make a color wheel that is correct? The way is to practice studying your eye's reactions to color as this will determine the opposite dualities. You can list the results by using a swatch book to match your after-images.

This was most elegantly described by Johann Wolfgang von Goethe in his Zur Farbenlehre section 5 he calls them 'physiological colors' that are displayed as an 'after-vision'. He felt that the impression of the colored object remains in the eye but the energy of the retina is stimulated to produce the opposite color, see figure 3 opposite.
Goethe writes further about the opposites of the color wheel when he explains that the human eye has an almost moral yearning for completeness and harmony: "To experience completeness, to satisfy itself, the eye seeks ... to produce the complental hue .... In this resides the fundamental law of all harmony of colors....(Johann Wolfgang von Goethe)



Opposite colors clash, for example mixing orange and blue produces a neutral gray.

or green and red as in my painting Neutrals.
Neutrals ©2000 David Kidd


In this painting see how the clash diminishes as you go away from the center. Perhaps this is how human conflicts can be diminished: dodge around the side of them. There is deeper significance in color wheels than just scientific curiosity or industrial pigment systems.

Dichotomy © 2002 David Kidd.


Most modern color wheels show two triangles, that is six points.

Color wheel 2002 David Kidd



Color wheel Pantone 1998 David Kidd

Everyone numbers their color wheel differently. The printers Pantone color system shown here starts with yellow as number 100, and then goes round the wheel up to number 399. But if you want to stick with artists pigment specifications, try this good website:


"Surprisingly it was not until about 1731 that the German theorist Le Blon made a discovery that many now say should have been obvious: that a comprehensive range of colors could be obtained by using just three basic colors, red, yellow and blue. Le Blon's discovery met with wide acclaim ... but it was not until some forty five years later that his work was presented in a practical form. In 1776 Moses Harris published the first organized colour wheel in an attempt to replace the then current understanding of colour which he described as being 'dark and occult'." (Wilcox)

RGB is the video and lighting system of additive primaries isolated by the Scottish physicist Sir James Clerk-Maxwell in 1861. When added together they make white light. This is how your computer monitor works.








Goethe would have called the R 'red-yellow', and the B 'red-blue', and Newton called that B 'indigo'.

CMY is the printing ink system of subtractive primaries isolated by The French pianist Louis Ducos Du Hauron in 1869. When subtracted from each other they make black ink. This is how printed literature works.









When old-fashioned artists like Goethe talk about primaries they agree with technologists on yellow but bend science by meaning by 'red' a hue in between R and M, and by 'blue' one in between B and C. These are what I have named Scarlet and Wedgwood, typified by the pigments Alizarin Crimson and Cobalt Blue, or Quinacridone Rose and Pthalo Blue. Scientifically however with Scarlet and Wedgwood they should use Lime instead of yellow to get a true primary triad that doesn't bend science. I named this my WSL system in 2002.








Unfortunately artists can't mix a bright yellow paint from lime and Scarlet, that's why they use yellow. But they will benefit by using a lemon yellow, not a warm yellow.


If we want Art and Science to stop conflicting I suggest that in schools we should stop teaching children that the spectrum is "Red, Orange, Yellow, Green, Blue, Indigo, Violet" and start teaching "Magenta, Red, Yellow, Green, Cyan, Blue". Unfortunately this omits the popular Orange and Violet because they are tertiaries. To include them we'd have to teach twelve colors: "Magenta, Scarlet, Red, Orange, Yellow, Lime, Green, Turquoise, Cyan, Wedgwood, Blue, Violet". This is the best option.


In industry the CIE chromaticity diagram is used to match colors of light in spectrometers. Zero corresponds to Blue, the vertical axis to Green, and the horizontal to Red. The visible wavelengths start at bottom left at 380nm and arc over to the right at 770nm. Although this is no circle, and the wavelength scale extends then shrinks, the primaries of the two color systems are still opposite each other.
My version of the CIE chromaticity diagram above is cubist to emphasize the key colors under discussion.


In the CIE chromaticity diagram the colors and tints really blend together without borders so it is impossible to agree on precise numbers for the primary wavelengths. Even experts disagree on what they should be, but that matters little because exact wavelengths are impractical to reproduce. Every color whether pigment or light proves to be merely a range that peaks in intensity around the desired wavelength. In fact pigments usually have multiple peaks thereabouts.

The color Magenta is not on the arc of colors because it falls in the purple department i.e. it is not one wavelength but two. See my purple page for a full discussion. However it can be specified on the CIE chromaticity diagram by using the opposite wavelength with the suffix c, that is 550c.


To an artist who wants to paint every subject the function of a chosen palette is to enable the painting of a full spectrum. To paint all my twelve key colors at maximum purity and intensity I find I need twelve different pigments. However opening and closing twelve tubes of paint is cumbersome, and wastes the time that I'd rather use for painting. It's inefficient. Knowing the idiosyncrasies of twelve paints is like having too many guests at a dinner party: you just don't get to know some of them.
If we cut back to six paints we can mix each adjacent pair to produce a good semblance of the color that lies between. Not quite at full intensity, but who needs full intensity of every part of a painting? Only the Fauves.
Let's try the minimum of pigments. What happens if we cut back to three? For example what happens if we try to paint a full spectrum by mixing the triad SWL, Scarlet, Wedgwood, Lime?








= 9


The purples (2 3 4) and greens (6 7 8) are dull but good enough for many purposes however the closest we can get to Orange (11) is a brown. And the closest to Yellow (10) is like Olive. I do like olive, and 11 is a very lovely brown, and for many paintings they would do.















When we mix pairs in different proportions to get my twelve key colors we find the middle ones dull. If I wanted to add daffodils or a golden sunset this triad is inadequate. I would need to add fourth pigment: Yellow.
If we started another triad with Yellow as the printers have done, we get the classic CMY which printers acknowledge cannot produce a bright purple. In fact they always add black (K) to cover up the holes left by CMY. For artists three pigments is inadequate for painting most subjects, although a very good exercise in mixing.


If the tertiary colors are added to make twelve colors, then it can also be easily divided into quarters. Using a set of four pigments would be an improvement. Lets try using four key colors. Since can't mix yellow from any other pigment, lets start with that irreplaceable hue. The quadrate would be YSBT (Yellow, Scarlet, Blue and Turquoise). That is Benzedrine Yellow, Quinacridone Rose, French Ultramarine and Cobalt Turquoise. That could be an excellent minimum palette with a novel pigment, Turquoise, unfamiliar to most artists. Try it out and see what I mean.





Since scientifically the pairs in a quadrate are opposites, yellow and blue should neuralize each other; so should scarlet and turquoise. Mixing these greys is a good test to see if you have the right pigments










Quinacridone Rose v. Pthalo Green, and Pthalo blue v. Gamboge.





   Our ancestors saw human health and temperaments as colors. They saw four types of personalities that I see as being these colors: See also my Color and Elements.

PMS 130

PMS 232

PMS 347

PMS 300



Billmeyer, Fred W and Max Saltzman. Principles of Color Technology. New York: John Wiley, 1981
Goethe, Johann Wolfgang von. Zur Farbenlehre 1810. Trans. as Goethe's Theory of Colours. London: Murray, 1840. & Cambridge: Massachusetts I.T., 1970.
Wilcox, Michael. Blue and Yellow Don't Make Green. Cincinnati: North Light, 1987.

All other material is copyright ┬ęDavid Kidd