In this way persons who, by themselves, have never obtained automatic movements of a coarse calibre, sometimes assume the unconscious guidance of the table-movements, provided that the tremors are strong and that the medium understands their meaning. In this case the medium takes control of the slight oscillations and returns them considerably strengthened, but rarely at exactly the same instant, generally a few seconds later, in this way revealing the agent's conscious or unconscious thought. By means of this simple mechanism there may arise those cases of thought-reading so bewildering at first sight. A very simple experiment, that succeeds in many cases even with unpractised persons, will serve to illustrate this. The experimenter thinks, say, of the number _four_, and then waits, his hands quietly resting on the table, until he feels that the table makes the first inclination to announce the number thought of. He lifts his hands off the table immediately, and the number _four_ will be correctly tilted out. It is advisable in this experiment to place the table upon a soft thick carpet. By close attention the experimenter will occasionally notice a movement of the table which is thus represented. [Illustration: FIG. 2.] (1) Purposeful tremors too slight to be perceived by the subject. (2) Several very small but perceptible oscillations of the table which indicate that the subject is responding to them. (3) The big movements (tilts) of the table, giving the number four that was thought of. (_ab_) Denotes the moment when the operator's hands are removed. This experiment succeeds excellently with well-disposed but inexperienced subjects. After a little practice the phenomenon indicated is wont to disappear, since by practice the number is read and reproduced directly from the purposeful movements.[29] In a responsive medium these purposeful tremors of the experimenter act just as the intentional taps in the experiment cited above; they are received, strengthened and reproduced, although slightly wavering. Still they are perceptible and hence act suggestively as slight tactile stimuli, and by the increase of partial hypnosis give rise to great automatic movements. This experiment illustrates in the clearest way the increase step by step of auto-suggestion. Along the path of this auto-suggestion are developed all the automatic phenomena of a motor nature. How the intellectual content gradually mingles in with the purely motor need scarcely be elucidated after this discussion. There is no need of a special suggestion for the evoking of intellectual phenomena. From the outset it is a question of word-presentation, at least from the side of the experimenter. After the first aimless motor irrelevancies of the unpractised subject, some word-products or the intentions of the experimenter are soon reproduced. Objectively the occurrence of an intellectual content must be understood as follows:--