[Illustration: FIG. 19.--THINKING OF A BUILDING. Both hands hold record plates, the pencils being held fixed above them. Time of record 35 seconds. Direction of the attention ↓. Subject facing ↓. I., left hand: II., right hand.] The outline presented in Fig. 19 was obtained in a test in which the movements of the hands were separately recorded, in order to determine the degree of correspondence between them. The result shows a marked general resemblance, indicating in part a common origin of the two movements. The next figure, Fig. 20, shows that this correspondence is dependent in part upon the similarity of the positions of the two hands. The hand that is held away from the body moves more extensively; but the form of the movements remain similar. The records reproduced in Figs. 14-22 and 26 were obtained upon the same subject, though with slightly varying conditions, and are fairly comparable with one another, and thus illustrate the analysis of the resultant movements into their component factors. [Illustration: FIG. 20.--THINKING OF A BUILDING. Each hand holds record plate. Time of record, 35 seconds. Direction of the attention ↓. Subject facing ↓. I., left hand held extended far out. II., right hand held close to body.] [Illustration: FIG. 21.--THINKING OF ONE'S FEET. Record plate vertical. Time of record, 45 seconds. Direction of the attention ↓. II., thinking of a point overhead. Time of record, 45 seconds. Recording plate vertical. Direction of the attention ↑.] Involuntary movements are not limited to the horizontal plane; vertical movements may be recorded by holding the recording device in a slanting position, and fixing the record plate upon the wall. The main characteristic of such a record is the sinking of the arm through fatigue; the movement is rapid and coarse (tracing I. of Fig. 20). If the attention be directed to the front, we obtain a resultant of the tendency to move towards the object of attention, and of the sinking of the arm, as appears in the diagonal line of Fig. 22. Fig. 21 illustrates an interesting point similar to that illustrated in Fig. 14. When the attention is directed downward, the hand falls rapidly (tracing I.); but when the attention is directed upward, very little movement at all takes place,--the tendency to move towards the object of attention constantly counteracting the tendency for the arm to fall (tracing II.). [Illustration: FIG. 22.--COUNTING THE STROKES OF A METRONOME. Record plate vertical. Pencil held in extended right hand. Time of record, 20 seconds. Direction of the attention ←. Subject facing ←.] V