is the usual circumstance in nature. But point C
will be found to be a little above the half, and point D a little less
than a third of the way up the vertical line. How much above the half
and less than the third will have to be observed by eye and a
corresponding amount allowed in setting out your drawing. In the
horizontal distances, B will be found to be one-fourth the distance from
X to the height of C on the right of our vertical line, and C a little
more than this distance to the left, while the distance on the right of
D is a little less than one-fifth of the whole height. The height of B
is so near the top as to be best judged by eye, and its distance to the
right is the same as B. These measurements are never to be taken as
absolutely accurate, but are a great help to beginners in training
the eye, and are at times useful in every artist's work.

[Illustration: Plate XVII.

DEMONSTRATION DRAWING MADE BEFORE THE STUDENTS OF THE GOLDSMITHS COLLEGE
SCHOOL OF ART

Illustrating how different directions of lines can help expression of
form.]

It is useful if one can establish a unit of measurement, some
conspicuous distance that does not vary in the object (if a living model
a great many distances will be constantly varying), and with which all
distances can be compared.

In setting out a drawing, this fixing of certain salient points is the
first thing for the student to do. The drawing reproduced on page 90
[Transcribers Note: Plate XVIII] has been made to illustrate the method
of procedure it is advisable to adopt in training the eye to accurate
observation. It was felt that a vertical line drawn through the pit of
the arm would be the most useful for taking measurements on, and this
was first drawn and its length decided upon. Train yourself to draw
between limits decided upon at the start. This power will be of great
use to you when you wish to place a figure in an exact position in a
picture. The next thing to do is to get the relative heights of
different points marked upon this line. The fold at the pit of the
stomach was found to be exactly in the centre. This was a useful start,
and it is generally advisable to note where the half comes first, and
very useful if it comes in some obvious place. Other measurements were
taken in the same way as our points A B C D E in the diagram on page 87
[Transcribers Note: Diagram IV], and horizontal lines drawn across, and
the transverse distances measured in relat