The existence of the self-styled intermediaries between sitter and communicator is another fact which does not fit in with the telepathic theory. Formerly Phinuit was the most common intermediary; then George Pelham collaborated with him; in Professor Hyslop's sittings, and, I believe, in all subsequent sittings since the installation of the Imperator _regime_, the intermediary is Rector. It is he who presides at the functioning of the "machine," because he is particularly competent--so say the communicators. These intermediaries have very defined and life-like characters. Phinuit, George Pelham and Rector are as unlike each other as possible. What, on the telepathic hypothesis, has had the power to create them? Mrs Piper's secondary personalities should have incarnated the communicator without intermediary. In order to understand this ephemeral reconstitution of a consciousness which has for ever vanished, we should have to allow that the scattered elements of this consciousness had temporarily grouped themselves around the _point de repere_ formed by the secondary personality of Mrs Piper. We should then see how difficult it is to explain the presence of these intermediaries. But if, on the other hand, we accept the spiritualistic hypothesis as well founded, we must admit that these intermediaries account for their presence very plausibly. Here is another argument, which, I think, is very strong, against the hypothesis of telepathy. Subjects in the hypnotic state, and the secondary personalities which appear in this hypnotic state, according to the precise and decisive experiments made by modern science, have an extremely definite notion of time. If you tell a hypnotised subject to perform an action in a year, at such an hour and minute, he will never fail, so to speak, although when he is awakened there remains in his memory no trace of the order. Now the communicators, in the phenomena we are studying, have an extremely vague notion of time, because, they say, time is not a concept of the world in which they live. How is it that telepathy, which can do so much, owns itself incapable, or nearly so, of determining the moment when an action has been performed? What prevents it from reading the idea of time, as well as any other idea, in the minds of the persons present, since the notion of time is as clear and precise in them at least as any other notion? To conclude, I should say that we are entirely ignorant of the point where the powers of telepathy begin and end. What I have just said makes the telepathic hypothesis an unlikely explanation; but, as Boileau said long ago, "Le vrai peut quelque fois n'etre pas vraisemblable"--Truth may sometimes be unlikely. CHAPTER XVII Some considerations which strongly support the spiritualistic hypothesis--Consciousness and character remain unchanged--Dramatic play--Errors and confusions.