3

The Munsell color sphere is described on Wikpedia. But what does the inside of the sphere look like? There is some conjecture on Wikipedia (most of which is original research), and I can only find text descriptions and the one painting of the exterior that seems to pop up everywhere:

enter image description here

Here is the best image of the interior I could find:

enter image description here

It is suggestive, but doesn't show all the internal grid lines, which you would expect in a detailed schematic.

So, I had a theory that a vertical cross section of the sphere would look like this:

enter image description here

But a comparison with Munsell's "Color Tree" would indicate the vertical cross section should look like this:

enter image description here

Which of these two seem most likely to you?

Improve this question
  • 1
    Munsell color system is a color space, which has a tag here. I'm really having a hard time finding the image. – posfan12 Aug 10 '18 at 9:23
  • 1
    Google "Munsell Color Chart Book" – user120647 Aug 10 '18 at 9:33
  • Thanks, but the Chart Book is more like a broken, irregular cylinder. The sphere he devised is older (he ditched the concept later on), but I'm still interested in it. – posfan12 Aug 10 '18 at 9:54
  • 2
    @BillyKerr: Color spaces are on-topic here. As for asking people to find stuff, I do not consider this much different from most other questions, in particular such as font-identification questions: Either somebody here knows the answer or is better at searching it. Also note that this site (like almost any other SE site) has a tag for reference requests. – Wrzlprmft♦ Aug 10 '18 at 10:07
  • 1
    In all honesty, the difference between your two diagrams is nil, as the terms "lightness and value" are largely interchangeable (although opposite in polarity - value is darkness) and chromicity or chroma is directly analagous to saturation. However, there is one significant departure: although your origin for the chroma scale is correct, even in the original Munsell sphere sketch the value axis origin (100% value=black) is at the south pole of the sphere, not at the centrepoint, as in your illustrations. – GerardFalla Aug 13 '18 at 15:34
4

Well, looks like the conceptual Munsell solid has... evolved since that first globular image he sketched to notate the concept: first with his own research and development back then, and further since under the auspices of the Munsell color company.

This is an image of the current iteration of the Munsell solid, linked from the Munsell website:

enter image description here

And as you can see, the conceptual sphere didn't account for differences in the highest possible perceptual intensities of specific hues, which differ, and so the "exterior surface" of the solid has morphed to account for that difference - but this does show the internal variation and gridlines to which you refer in your question.

Here is the underlying Munsell hue wheel:

enter image description here

And here a depiction of the range of value as the vertical axis throughout the Munsell solid:

enter image description here

And last, this depicts the variation in chroma being the horizontal plane outwards from the Munsell solid's origin:

enter image description here

Improve this answer
  • Does this indicate that the color wheel is conceptually wrong in that maybe red should have a bigger slice than blue? – joojaa Aug 10 '18 at 20:10
  • 1
    Thanks for your response. However, I am interested in the early color solids because they failed. I want to focus on Munsell's "Color Sphere", not any later iterations. To that effect, I amended my question with two additional images. Which of the two white and black wireframe images do you think most accurately captures what Munsell was trying to do with his early sphere? – posfan12 Aug 10 '18 at 23:19
  • 1
    I also found a scanned copy of the Atlas of the Munsell Color System on the Smithsonian Libraries website. ([Link][1].) The information I needed is on page 13 of 42. [1]: library.si.edu/digital-library/book/atlasmunsellcol00muns – posfan12 Aug 15 '18 at 16:36
  • @posfan12 - fascinating - you notice that even in that early iteration, the per-swatch matrices not only don't fit in the conceptual sphere but in fact already give rise the forms now recognised by Munsell and showing up in my answer? One notices particularly on the spread relating to value how close to the current form they were even then. THANK YOU for posting this! – GerardFalla Aug 15 '18 at 17:01
  • 1
    It depends on your (or Munsell's) criteria. For instance, if human perceptual uniformity is important to you, then you're going to appreciate the arrangement the Munsell company has settled upon. If dogs and cats learn to paint one day, the Munsell Color System is not going to be very useful to them, since they have very different visual systems. This applies to CIELAB, CIELUV, etc. as well. If you don't care about perceptual uniformity at all, then a color solid could take virtually any shape. – posfan12 Aug 15 '18 at 18:05
2

This may disappoint you but there were never any colours inside the Munsell sphere! The sphere Munsell patented in 1900 was not a colour space but an educational device consisting of a physical sphere painted with colours on the surface comprising six "spectral" hues at middle value around the equator, grading up and down to white and black respectively at the poles. The sphere was mounted so that it could be examined by rotating it on various axes. Munsell's original patent is online here: https://patents.google.com/patent/US640792A/en

Later Munsell manufactured and sold another version of his sphere based on his five principal hues that could be spun rapidly by means of a motor to optically produce three latitudinal grey bands. Several of these spheres still exist: https://www.1stdibs.com/furniture/more-furniture-collectibles/collectibles-curiosities/unusual-decorative-color-theory-sphere-circa-1900/id-f_480374/

Improve this answer
  • Munsell chose the spherical shape because he (before his color experiments) initially thought all colors could be arranged inside a sphere. – posfan12 Aug 15 '18 at 18:19
  • 1
    Nearly right. When he patented his colour sphere model in 1900 he was treating colour intensity as a relative dimension, just as Runge had done in 1810 and as Itten would do again in 1961. In May 1901 he began thinking of chroma as an absolute scale and established a distinction between a theoretical sphere of all colour sensations and an actual "irregular spheroid" of colours of pigments and dyes. He then conducted his experiments to establish the shape of this "irregular spheroid". If you haven't seen these you're in for a treat!: rit.edu/cos/colorscience/ab_munsell_diaries.php – David Briggs Aug 17 '18 at 2:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.