On Sense And The Sensible   

medium of the other, giving an effect like that sometimes produced

by painters overlaying a less vivid upon a more vivid colour, as

when they desire to represent an object appearing under water or

enveloped in a haze, and like that produced by the sun, which in

itself appears white, but takes a crimson hue when beheld through a

fog or a cloud of smoke. On this hypothesis, too, a variety of colours

may be conceived to arise in the same way as that already described;

for between those at the surface and those underneath a definite ratio

might sometimes exist; in other cases they might stand in no

determinate ratio. To [introduce a theory of colour which would set

all these hypotheses aside, and] say with the ancients that colours

are emanations, and that the visibility of objects is due to such a

cause, is absurd. For they must, in any case, explain sense-perception

through Touch; so that it were better to say at once that visual

perception is due to a process set up by the perceived object in the

medium between this object and the sensory organ; due, that is, to

contact [with the medium affected,] not to emanations.

If we accept the hypothesis of juxtaposition, we must assume not

only invisible magnitude, but also imperceptible time, in order that

the succession in the arrival of the stimulatory movements may be

unperceived, and that the compound colour seen may appear to be one,

owing to its successive parts seeming to present themselves at once.

On the hypothesis of superposition, however, no such assumption is

needful: the stimulatory process produced in the medium by the upper

colour, when this is itself unaffected, will be different in kind from

that produced by it when affected by the underlying colour. Hence it

presents itself as a different colour, i.e. as one which is neither

white nor black. So that, if it is impossible to suppose any magnitude

to be invisible, and we must assume that there is some distance from

which every magnitude is visible, this superposition theory, too [i.e.

as well as No. 3 infra], might pass as a real theory of

colour-mixture. Indeed, in the previous case also there is no reason