Before passing to a more specific interpretation of the data, it may be interesting to illustrate more fully the scope of individual variations; for the great difference in availability of subjects to the muscle-reader is equally prominent in tests with the automatograph. Some movements are direct and extensive, others are circuitous and brief. Fig. 10 is a good type of a small movement, but of one quite constantly toward the object of the attention. This may be contrasted with an extreme record, not here reproduced, in which there is a movement of six and a half inches in forty-five seconds; or with a fairly extensive movement as in Fig. 11. In some cases the first impulse carries the hand toward the object of thought, and is followed by considerable hesitation and uncertainty; a marked example of this tendency may be seen in Fig. 12. There is, too, an opposite type, in which the initial movements are variable, and the significant movement toward the object of thought comes later, when perhaps there is some fatigue. This tendency appears somewhat in Figs. 11 and 13. [Illustration: FIG. 13.--THINKING OF A LOCALITY. Time of record, 120 seconds. Direction of the attention ←. Subject facing ←. Illustrates initial hesitancy followed by a steady movement toward the object thought of.] IV What is the origin of the movements involved in these records? To what extent are they movements of the hand, of the arm, or of the entire body? Casual observation is sufficient to show that with a given position of the arm, certain movements are much more readily made than others; and the involuntary tendencies will naturally follow the lines of least resistance. If, for instance, you hold your arm nearly on a level with the shoulders and in line with them, you perceive at once that movements of the hand to the front are much more readily made than to the rear, and movements toward the body more readily than those away from the body; the tendency of the hand is to move forward in a circle of which the shoulder is the centre. What we require is a position in which movements in any one direction are as readily made as in any other; and this may be approximated, though only approximated, by holding the hand at an angle of about 45° with the line joining the shoulders, and with the elbow bent at an angle of about 120°. This was the position in most of the tests, and the usual result was a movement toward the object of attention; but when the object attended to lies in back of the subject, this tendency is sometimes outweighed by the natural tendency for the arm to move forward, and the result may be a movement _forward_, but a less direct movement forward than when the object of attention is to the front. In a good subject, however, the involuntary tendency is strong enough to prevail, and a movement _backward_ results. An instance of this, obtained under other but comparable circumstances, appears in Fig. 14.