Abstract
Smith VC, Pokorny J. Spectral sensitivity of the foveal cone photopigments between 400 nm and 500 nm. Vision Res.1975; 15: 161–171.
This is probably the most unread paper in vision science that has over 1000 citations! The purpose was to estimate correct cone fundamentals, the spectral sensitivities of the L, M, and S cones. The approach included colorimetric transformations, and the write-up and was very much directed to the specialist reader. After reading a manuscript draft, I commented to Vivianne that she had not included a tabulation of the L, M, and S spectral sensitivities. She said, “If someone wants the spectra, they can plug x,y,z values into the solution equations”. Not exactly a way to promote the derived spectra!
So, how did the fundamentals become widely accepted? A feature of the fundamentals was that Y, the relative luminous efficiency function was a linear combination of L and M, with no S-cone input. At the 1975 Optical Society Annual Meeting in Boston, Brian Tansley and Bob Boynton presented a paper on the blurring of minimally distinct borders when the pair of chromaticities was on a tritanopic confusion line. They modeled the data using the Vos-Walraven fundamentals, which have an S-cone contribution to luminance. After the session I chatted with Brian (Bob was not at the meeting) and argued that their analysis required fundamentals that had no S-cone contribution to luminance. Brian conveyed this to Bob, and in 1979 Bob and Don Macleod incorporated the “Smith-Pokorny fundamentals” in the MacLeod-Boynton chromaticity diagram. Bob tabulated the fundamentals in his 1983 book “Human Color Vision”, the tabulations became stored in many computers, they were used and found satisfactory, and citations followed.
Addendum: The fundamentals remain good descriptions of the human cone spectral sensitivities