Five hours after Duvanel's death an automatic message announcing it is written at the house of M. Kaigorodoff, at Wilna in Russia, by Mdlle. Stramm, who had certainly at that time received no news of the event.

From what mind are we to suppose that this information came?

(1) We may first attempt to account for Mdlle. Stramm's message on the theory of latency. We may suppose that the telepathic message came from the dying man, but did not rise into consciousness until an opportunity was afforded by Mdlle. Stramm's sitting down to write automatically.

But to this interpretation there is an objection of a very curious kind. The message written by Mdlle. Stramm was not precisely accurate. Instead of ascribing Duvanel's death to suicide, it ascribed it to a stoppage of blood, "un engorgement de sang."

And when M. Stramm, three days after the death, wrote to his daughter in Russia to tell her of it, he also used the same expression, "un engorgement de sang," thus disguising the actual truth in order to spare the feelings of his daughter, who had formerly refused to marry Duvanel, and who (as her father feared) might receive a painful shock if she learnt the tragic nature of his end. There was, therefore, a singular coincidence between the automatic and the normally-written message as to the death;—a coincidence which looks as though the same mind had been at work in each instance. But that mind cannot have been M. Stramm's ordinary mind, as he was not supraliminally aware of Duvanel's death at the time when the first message was written. It may, however, be supposed that his subliminal self had received the information of the death telepathically, had transmitted it in a deliberately modified form to his daughter, while it remained latent in himself, and had afterwards influenced his supraliminal self to modify the information in the same way when writing to her.

(2) But we must also consider the explanation of the coincidence given by the intelligence which controlled the automatic writing. That intelligence asserted itself to be a brother of Mdlle. Stramm's, who died some years before. And this "Louis" further asserted that he had himself influenced M. Stramm to make use of the same euphemistic phrase, with the object of avoiding a shock to Mdlle. Stramm; for which purpose it was needful that the two messages should agree in ascribing the death to the same form of sudden illness.

Now if this be true, and the message did indeed come from the deceased "Louis," we have an indication of continued existence, and continued knowledge of earthly affairs, on the part of a person long dead.

But if we consider that the case, as presented to us, contains no proof of "Louis'" identity, so that "Louis" may be merely one of those arbitrary names which the automatist's subliminal intelligence seems so prone to assume; then we must suppose that Duvanel was actually operative on two occasions after death, first inspiring in Mdlle. Stramm the automatic message, and then modifying in M. Stramm the message which the father might otherwise have sent.

I next quote a case in Appendix VIII. D which illustrates the continued terrene knowledge on the part of the dead of which other instances were given in the last chapter.

And lastly, I give in Appendix VIII. E a case which in one respect stands alone. It narrates the success of a direct experiment,—a test-message planned before death, and communicated after death, by a man who held that the hope of an assurance of continued existence was worth at least a resolute effort, whatever its result might be. His tests, indeed, were two, and both were successful. One was the revealing of the place where, before death, he hid a piece of brick marked and broken for special recognition, and the other was the communication of the contents of a short letter which he wrote and sealed before death. We may say that the information was certainly not possessed supraliminally by any living person. There are two other cases (Proceedings S.P.R., vol. vi. pp. 353-355, and op. cit. vol viii. pp. 238-242 [876 A and B]) where information given through automatists may hypothetically be explicable by telepathy from the living, although, indeed, in my own view, it probably emanated from the deceased as alleged. In one of these cases the place where a missing will had been hidden was revealed to the automatist, but it is not clear whether the will was actually discovered or not before the automatic writing was obtained (although the automatist was unaware of its discovery), and in any case, apparently, its whereabouts was known to some living person who had hidden it, and may not have been known to the deceased before death.