Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 exptools 1/6/84; site ihima.UUCP
Path: utzoo!linus!decvax!decwrl!amd!dual!zehntel!ihnp4!ihima!cmb
From: cmb@ihima.UUCP (Christine M. Buss)
Newsgroups: net.physics
Subject: Re: colors
Message-ID: <208@ihima.UUCP>
Date: Wed, 1-Aug-84 17:04:27 EDT
Article-I.D.: ihima.208
Posted: Wed Aug  1 17:04:27 1984
Date-Received: Fri, 3-Aug-84 01:59:35 EDT
Organization: AT&T Bell Labs, Naperville, IL
Lines: 37

>Human eyes cannot generally see "chords" of light. 
>Our ears are capable of doing a spectral analysis of sound, but we cannot
>do the same with light.  The ears can simultaneously detect a large range of
>frequencies, our eyes can only detect (!) three.  However, we are able to see
>one "chord", the combination of blue and red which makes magenta.
(Exclamation point mine)

As someone who studies human color vision, I find these statements 
rather misleading.  As you say, the ears can detect a large range of 
sound frequencies. They do a kind of Fourier analysis on the sound and
retain information about each of the constituent frequencies in the
sound.  

The eyes also can detect (and discriminate) a large range of light frequencies
(from about 400 to 700 nm), not just three.  Three light absorbing pigments 
with different spectral sensitivities (broad, overlapping, but not identical 
spectral sensitivities) report to the visual system the quantum catches 
they receive from any light.  The visual system compares their three outputs 
to determine the "color" of the light.  So the visual system
can discriminate any two single wavelengths in the visible spectrum 
(except a pair very close together), but any two mixtures of wavelengths 
that produce the same quantum absorption in the three pigments look identical.  

I have no idea what to make of your use of the word "chord" with respect 
to vision.  If a chord is simply a mixture of wavelengths, than of
course we can _see_ them.  Can we tell what the constituent wavelengths
are?  No.  In the case of magenta, we can tell approximately what two
wavelengths must be mixed to produce it, but only really on the basis of
empirical color mixture experience.  It's not at all like what the ear
does, where the components of a chord are experienced as distinct in the 
mixture.  Magenta looks like a single color, not like the superposition 
of two colors, red and blue.

I'm not at all certain that the author of the quoted lines
misunderstands color vision, but I think that one could easily read what
he said and come away with some mistaken ideas of vision.  I hope this
makes some of the processes clearer.