OCR Output

COLOURS IN RIPPLED WATER oz

doubted brilliancy, and watch the breeze rippling the
water in streaks or patches a few yards from us, and
thus disturbing the image of the mountain on the
opposite shore, we look in vain in the ripples for the
added local colour, which we might have expected,
and can detect nothing but a gleam from the sky
above. Follow one of these streaks sideways, how¬
ever, into the reflexions of the tall trees that line the
bank close at hand, where the upturned faces of the
ripples reflect the tree-tops and not the sky (so that
there is no interruption of the dark image by bright
sky light) and there the colour of the water itself, blue
or green, as the case may be, can again be detected,
affording an illustration of the fact that (as pointed out
" on page 78) the local colour of water reveals itself most
in the reflexion of dark objects. In Plate XX XIX,
if the water were not too clear, its colour would in all
probability appear in the more ruffled parts. But in
Plate XX XVIII we could hardly expect the rippled
water, which reflects the sky, to show more local
colour than the smooth water, which reflects the dark
hillside. If, however, the hill were high enough, or
near enough, to cast an interrupted reflexion over the
whole water, rippled and smooth, then again the local
colour might be discerned in the rippled surface.
These instances show how, under the complex con¬
ditions at which we have arrived in this final stage
of our inquiry, it is impossible to give to the artist
hard and fast directions, one effect often overpower¬
ing the other to such an extent that the latter is
hardly perceptible. ;

a i K s