I did notice that there wasn't any good "blue" on the grid they provided: https://theconversation.imgix.net/files/186280/original/file-20170917-25673-migist.pn

I did notice that there wasn't any good "blue" on the grid they provided: https://theconversation.imgix.net/files/186280/original/file-20170917-25673-migist.png?ixlib=rb-1.1.0&q=45&auto=format&w=1000 https://theconversation.imgix.net/files/186280/original/file... My two monitors display the grids a bit differently, but A3 is pretty obviously red, B4 is orange, E7 is the best yellow, H4 is the best green. None of the blues look like the primary color blue though. I took a sample of the blue colors in a paint program and they all seemed pretty far away from a proper blue. The hue is off. N4, which was presented in the article as blue, is RGB (35, 131, 172). A3 red by comparison is (190, 27, 50) and H4 green is (42, 144, 48). As an additional test, I edited the grid, drawing rectangles of red (255, 0, 0) and blue (0, 255, 0) over each of the proximate color squares. Pure red is fairly close to A3 whereas pure blue is very different from all blue-like squares on the grid. It feels to me that something is off about this grid, like it doesn't cover enough of the color space to encompass a proper primary blue, but I don't know enough about color theory to characterize what's wrong with it.