painted in an in-door or concentrated light. The painting of
figures out-of-doors you will find more difficult if you have had no
experience in painting them in the studio. The problems of light and
shade and color are more complex in the diffused light, and the
knowledge of structure and modelling, as well as of special values
gained by studio study, will be most helpful to you when you paint
out-of-doors. I should say, then, don't attempt any serious painting
of the human figure in the open air till you have had some experience
with its special problems in the house.

=The Nude.=--No good figure-work has ever been done which was not
founded on a knowledge of the nude. Whether the figure is draped or
not, the nude is the basis of form. The best painters have always made
their studies of pose and action in the nude, and then drawn the
draperies over that. This insures the truth of action and structure,
which is almost sure to be lost when the drawing of the form is made
through drapery or clothing. The underlying structure is as essential
here as in portrait. It is the more imperative that the body be felt
within the clothes from the fact that it cannot be seen. There must be
no ambiguity; no doubt as to the anatomy underneath; for without this
there can be no sense of actuality.

I do not say paint the nude. On the contrary, if you want to go so far
as that in the study of the figure, you must not attempt to do it with
the aid of a book. Go to a good life class. But I wish to emphasize
the principle that when you undertake to paint anything involving the
figure, you must know something of the structure of what is more or
less hidden, and must make allowance for the disguising of form which
the draping of it will inevitably cause.

And when you draw your figure, you should lay in your main lines, at
any rate, from the nude figure if you can. If you cannot command a
professional model for this purpose, you can only be more careful
about your study of the underlying lines and forms as they are
suggested by the saliencies of the draperies.

If this is the case, be most accurate in those measurements which
place the proportions of the parts which show through the covering,
and try to trace out by the modelling where the lines would run. By
mapping out these proportions, and drawing the lines over the drapery
masses wherever you can make them out, you can judge to a certain
extent of the truth of action in your drawing.