anches, so as to leave the whole only
about a foot in length; get a piece of paper the same size, fix your bit
of branch in some place where its position will not be altered, and draw
it thoroughly, in all its light and shade, full size; striving, above
all things, to get an accurate expression of its structure at the fork
of the branch. When once you have mastered the tree at its _armpits_,
you will have little more trouble with it.

73. Always draw whatever the background happens to be, exactly as you
see it. Wherever you have fastened the bough, you must draw whatever is
behind it, ugly or not, else you will never know whether the light and
shade are right; they may appear quite wrong to you, only for want of
the background. And this general law is to be observed in all your
studies: whatever you draw, draw completely and unalteringly, else you
never know if what you have done is right, or whether you _could_ have
done it rightly had you tried. There is nothing _visible_ out of which
you may not get useful practice.

74. Next, to put the leaves on your boughs. Gather a small twig with
four or five leaves on it, put it into water, put a sheet of
light-colored or white paper behind it, so that all the leaves may be
relieved in dark from the white field; then sketch in their dark shape
carefully with pencil as you did the complicated boughs, in order to be
sure that all their masses and interstices are right in shape before you
begin shading, and complete as far as you can with pen and ink, in the
manner of Fig. 6, which is a young shoot of lilac.

75. You will probably, in spite of all your pattern drawings, be at
first puzzled by leaf foreshortening; especially because the look of
retirement or projection depends not so much on the perspective of the
leaves themselves as on the double sight of the two eyes. Now there are
certain artifices by which good painters can partly conquer this
difficulty; as slight exaggerations of force or color in the nearer
parts, and of obscurity in the more distant ones; but you must not
attempt anything of this kind. When you are first sketching the leaves,
shut one of your eyes, fix a point in the background, to bring the point
of one of the leaves against; and so sketch the whole bough as you see
it in a fixed position, looking with one eye only. Your drawing never
can be made to look like the object itself, as you see that object with
_both_ eyes,[17] but it can be made perfectly li