Firstly, I have made no _voluntary_ movement; and secondly, I have made no _conscious involuntary_ movement. But, thirdly, I have made an _unconscious involuntary_ movement which directly depended on conscious ideation. I strongly thought of the bookshelf, and when the bookshelf was reached in our vague career about the room I made a movement--say rather a tremor occurred--in my hand, which, although beyond both my knowledge and my control, was enough to supply to the muscle-reader's delicate sensibility all the indication required. All this is now admitted, and, in a sense, understood; we formulate it by saying that my conscious ideation contained a motor element; and that this motor element, though inhibited from any conscious manifestation, did yet inevitably externalise itself in a peripheral tremor. But, fourthly, something more than this has clearly taken place. Before the muscle-reader stopped at the bookshelf he stopped at the rug. I was no longer consciously thinking of the rug; but the idea of the pin in the rug must still have been reverberating, so to say, in my sub-conscious region; and this unconscious memory, this unnoted reverberation, revealed itself in a peripheral tremor nearly as distinct as that which (when the bookshelf was reached) corresponded to the strain of conscious thought. This tremor, then, was in a certain sense a message-bearing automatism. It was the externalisation of an idea which, once conscious, had become unconscious, though in the slightest conceivable degree--namely, by a mere slight escape from the field of direct attention. Having, then, considered an instance where the automatic message passes only between two closely-adjacent strata of consciousness, externalising an impulse derived from an idea which has only recently sunk out of consciousness and which could easily be summoned back again;--let us find our next illustration in a case where the line of demarcation between the strata of consciousness through which the automatic message pierces is distinct and impassable by any effort of will. Let us take a case of _post-hypnotic suggestion_;--say, for instance, an experiment of Edmund Gurney's (see _Proceedings_ S.P.R., vol. iv. p. 319). The subject had been trained to write with planchette, after he had been awakened, the statements which had been made to him when in the hypnotic trance. He wrote the desired words, or something like them, but while he wrote them his waking self was entirely unaware of what his hand was writing. Thus, having been told in the trance, "It has begun snowing again," he wrote, after waking, "It begun snowing," while he read aloud, with waking intelligence, from a book of stories, and was quite unconscious of what his hand (placed on a planchette behind a screen) was at the same time writing.