M. Richet[164] was, I believe, the first writer, outside the Spiritualistic group, who so much as showed any practical knowledge of this phenomenon, still less endeavoured to explain it. Faraday's well-known explanation of table-turning as the result of the summation of many unconscious movements--obviously true as it is for some of the simplest cases of table-movement--does not touch this far more difficult question of the origination of these intelligent messages, conveyed by distinct and repeated movements of some object admitting of ready displacement. The ordinary explanation--I am speaking, of course, of cases where fraud is not in question--is that the sitter unconsciously sets going and stops the movements so as to shape the word in accordance with his expectation. Now that he unconsciously sets going and stops the movements is part of my own present contention, but that the word is thereby shaped in accordance with his expectation is often far indeed from being the case. To those indeed who are familiar with automatic _written_ messages, this question as to the unexpectedness of the _tilted_ messages will present itself in a new light. If the written messages originate in a source beyond the automatist's supraliminal self, so too may the tilted messages;--even though we admit that the tilts are caused by his hand's pressure of the table just as directly as the script by his hand's manipulation of the pen. One piece of evidence showing that _written_ messages are not always the mere echo of expectation is a case[165] where _anagrams_ were automatically written, which their writer was not at once able to decipher. Following this hint, I have occasionally succeeded in getting anagrams tilted out for myself by movements of a small table which I alone touched. This is a kind of experiment which might with advantage be oftener repeated; for the extreme incoherence and silliness of the responses thus obtained does not prevent the process itself from being in a high degree instructive. Here, again (as in automatic writing), a man may hold colloquy with his own dream--may note in actual juxtaposition two separate strata of his own intelligence. I shall not at present pursue the discussion of these tilted responses beyond this their very lowest and most rudimentary stage. They almost immediately suggest another problem, for which our discussion is hardly ripe, the participation, namely, of several minds in the production of the same automatic message. There is something of this difficulty even in the explanation of messages given when the hands of two persons are touching a planchette; but when the instrument of response is large, and the method of response simple, as with table-tilting, we find this question of the influence of more minds than one imperatively recurring.