our drawing will not
for a long while be worth the crumbs. So use india-rubber very
lightly; or, if heavily, pressing it only, not passing it over the
paper, and leave what pencil marks will not come away so, without
minding them. In a finished drawing the uneffaced penciling is often
serviceable, helping the general tone, and enabling you to take out
little bright lights.

[3] What is usually so much sought after under the term "freedom" is
the character of the drawing of a great master in a hurry, whose
hand is so thoroughly disciplined, that when pressed for time he can
let it fly as it will, and it will not go far wrong. But the hand of
a great master at real _work_ is _never_ free: its swiftest dash is
under perfect government. Paul Veronese or Tintoret could pause
within a hair's breadth of any appointed mark, in their fastest
touches; and follow, within a hair's breadth, the previously
intended curve. You must never, therefore, aim at freedom. It is not
required of your drawing that it should be free, but that it should
be right; in time you will be able to do right easily, and then your
work will be free in the best sense; but there is no merit in doing
wrong easily.

These remarks, however, do not apply to the lines used in shading,
which, it will be remembered, are to be made as quickly as possible.
The reason of this is, that the quicker a line is drawn, the lighter
it is at the ends, and therefore the more easily joined with other
lines, and concealed by them; the object in perfect shading being to
conceal the lines as much as possible.

And observe, in this exercise, the object is more to get firmness of
hand than accuracy of eye for outline; for there are no outlines in
Nature, and the ordinary student is sure to draw them falsely if he
draws them at all. Do not, therefore, be discouraged if you find
mistakes continue to occur in your outlines; be content at present
if you find your hand gaining command over the curves.

[4] If you can get any pieces of dead white porcelain, not glazed,
they will be useful models.

[5] Artists who glance at this book may be surprised at this
permission. My chief reason is, that I think it more necessary that
the pupil's eye should be trained to accurate perception of the
relations of curve and right lines, by having the latter a