emptible, with which the subject has been treated in modern
days.[64] I am not aware that I can refer to any instructive example of
this intermediate course, for I fear the reader is by this time wearied
of hearing of Turner, and the plate of Turner's picture of the deluge is
so rare that it is of no use to refer to it.

Sec. 23. Venetians and Florentines. Conclusion.

It seems exceedingly strange that the great Venetian painters should
have left us no instance, as far as I know, of any marine effects
carefully studied. As already noted, whatever passages of sea occur in
their backgrounds are merely broad extents of blue or green surface,
fine in color, and coming dark usually against the horizon, well enough
to be understood as sea, (yet even that not always without the help of a
ship,) but utterly unregarded in all questions of completion and detail.
The water even in Titian's landscape is almost always violently though
grandly conventional, and seldom forms an important feature. Among the
religious schools very sweet motives occur, but nothing which for a
moment can be considered as real water-painting. Perugino's sea is
usually very beautifully felt; his river in the fresco of S^ta.
Maddalena at Florence is freely indicated, and looks level and clear;
the reflections of the trees given with a rapid zigzag stroke of the
brush. On the whole, I suppose that the best imitations of level water
surface to be found in ancient art are in the clear Flemish landscapes.
Cuyp's are usually very satisfactory, but even the best of these attain
nothing more than the agreeable suggestion of calm pond or river. Of any
tolerable representation of water in agitation, or under any
circumstances that bring out its power and character, I know no
instance; and the more capable of noble treatment the subject happens to
be, the more manifest invariably is the painter's want of feeling in
every effort, and of knowledge in every line.

FOOTNOTES

[61] I state this merely as a fact: I am unable satisfactorily to
account for it on optical principles, and were it otherwise, the
investigation would be of little interest to the general reader, and
little value to the artist.

[62] Parsey's "Convergence of Perpendiculars." I have not space here
to enter into any lengthy exposure of this mistake, but reasoning is
fortunately unnecessary, the appeal to experiment being easy. Every
picture is the representation, as