Thoughts on Color Mixing Systems

So, I took color theory back in the mists of time. I never really thought about it much and followed my intuition along with the standard color palettes that were recommended to me at the time. Later, woodblock printmaking required me to think deeper- primarily in order to simplify and make more efficient my color choices- especially using shin-hanga techniques which involve layer over layer of colors and values to emulate watercolor painting.

During the 90s, I followed the traditional ukiyo-e palette using indian yellow, vermillion, ultramarine blue, indigo,  bengal red, carmine, and sumi. This system allows for a good range of colors- what I saw was a lack of a warm, light blue, so I later integrated cerulean blue.

After discussing color with Paul Ritscher of Salinas, California, I bought Blue and Yellow Does Not Make Green by Michael Wilcox. It’s a handy book that goes into detail, but what I gleaned most from it was something that I had intuited for some time but didn’t pursue enough:  That the single Red, Blue, Yellow primary palette wasn’t nearly enough and using a split-primary system: a warm red, cool red, warm blue, cool blue, warm yellow, and a cool yellow provided cleaner and broader options.

Below is a chart that I put together to help me look at split primary interactions. I am neglecting tints (lighter hue combinations) which, in woodblock printmaking are created by weaker ink mixtures and shades (darker hue combinations) which I use a combination of compliment hue mix to darken. Payne’s Gray is made this way I believe.


Things above might look a bit complicated, but I think for me, charting this out helps to cement my understanding- both intuitively and rationally. It’s interesting to see that the chart isn’t ‘symmetrical’ in that the pure green isn’t on the outer end of the secondaries as the pure purple and orange is. Also, there is some ‘criss-crossing’ going on between the reds and blues. I think that this is a result combining a ‘dualistic’ system (a single color mixing with another single color) with three 2-primaries. I feel that this palette allows for solutions to common problems and produces nice reds, pure purples, a range of greens, etc.

If I had continued with the 3 primary system, in order to get a nice orange (product of a warm red and a warm yellow), I would have to compromise with producing a muddy brownish purple which would have to be a product of a warm red and a cool blue (see the swatch above “pure purple” for an approximation). Incidentally, while I was a graphic designer, I was told that “Coke” (Coca-Cola) red required a five-color printing set-up since the CMYK color system couldn’t produce the proper red from Magenta and Yellow.

In the defense of mud, I think you will agree after looking at the middle of the secondary “Y” in the chart, that there are some rather handsome subdued colors that are a product of a cool and warm combination.

“Why does a mixture of a warm and cool primary appear muddy?”

Getting Geeky

I found a nice, albeit scientific explanation to how light is reflected within the visual spectrum. At the risk of comprehending this, it appears that molecules are ‘excited’ in different ways by photons hitting them. The wavelength and amount of the reflection depends on the molecule’s ability to revert back to its normal state…

Inversely, (from what my limited understanding can gather) the molecule’s inability to ‘calm down’ results in its absorption of the color’s wavelength- maybe the vibrations are interference… Anyway, if there is absolutely no reflectiveness, then black occurs. On a side note, vantablack uses texture to not reflect light. How can you see colored velvet then?…

It follows that if a wide area of the spectrum is absorbed, then a muted color is perceived by the eye. Inversely, if a narrow part of the spectrum is reflected, then a purer and lighter version of the hue is perceived.

So , how does this relate to warm/cool combinations? It seems to me that using say, a cool blue and a warm red covers too wide of the spectrum- therefore the wider the range of color, the less the reflection, the muddier the color. Whew! my feeble brain hurts, but maybe, just maybe this makes sense. Color is a complicated endeavor fraught with emotion and many, many variables.

Other Considerations

As far as the support (in my case, mostly-white thick mulberry paper) effecting color- I was told that the advantage of using the combination of white paper and transparent inks is that light first passes through the pigment, then bounces off of the white paper, then travels back through the pigment to create an addition effect. I’ve never printed on black paper, but I think that we can imagine the results.

An interesting reference for more detailed information (it’s free to download the .pdf) is The Color Book.

Another free .pdf treatise on color palettes is from an early western woodblock artist, F. Morley Fletcher: Colour-control: the organization and control of the artist’s palette, 1936

With all of this said, I am still a ‘valueist” rather than a colorist- if you can make the values work, then the hues tend to follow.

Happy mixing, folks! Maybe you’ve got your own system that works in a different way- please respond with a message below.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.