e colors of Turner,--in
those very colors which perhaps you have been laughing at all your
life,--the fact being that he, and he alone, of all men, _ever painted
Nature in her own colors_.

Sec. 10. "Well, but," you will answer, impatiently, "how is it, if they are
the true colors, that they look so unnatural?"

Because they are not shown in true contrast to the sky, and to other
high lights. Nature paints her shadows in pale purple, and then raises
her lights of heaven and sunshine to such height that the pale purple
becomes, by comparison, a vigorous dark. But poor Turner has no sun at
his command to oppose his pale colors. He follows Nature submissively as
far as he can; puts pale purple where she does, bright gold where she
does; and then when, on the summit of the slope of light, she opens her
wings and quits the earth altogether, burning into ineffable sunshine,
what can he do but sit helpless, stretching his hands towards her in
calm consent, as she leaves him and mocks at him!

Sec. 11. "Well," but you will farther ask, "is this right or wise? ought
not the contrast between the masses be given, rather than the actual
hues of a few parts of them, when the others are inimitable?"

Yes, if this _were_ possible, it ought to be done; but the true
contrasts can NEVER be given. The whole question is simply whether you
will be false at one side of the scale or at the other,--that is,
whether you will lose yourself in light or in darkness. This necessity
is easily expressible in numbers. Suppose the utmost light you wish to
imitate is that of serene, feebly lighted, clouds in ordinary sky (not
sun or stars, which it is, of course, impossible deceptively to imitate
in painting by any artifice). Then, suppose the degrees of shadow
between those clouds and Nature's utmost darkness accurately measured,
and divided into a hundred degrees (darkness being zero). Next we
measure our own scale, calling our utmost possible black, zero;[18] and
we shall be able to keep parallel with Nature, perhaps up to as far as
her 40 degrees; all above that being whiter than our white paper. Well,
with our power of contrast between zero and 40, we have to imitate her
contrasts between zero and 100. Now, if we want true contrasts, we can
first set our 40 to represent her 100, our 20 for her 80, and our zero
for her 60; everything below her 60 being lost in blackness. This is,
with certain modifications, Rembrandt's system. Or, secondly, we ca