[53] Of course substitutions of this kind considerably reduce the value of the results obtained, but it will be found that when full deduction has been made on this score, the coincidences remain overwhelmingly in excess of anything which could have been produced by chance.
[54] In some previous experiments a circular alphabet had been used, with a view of preventing any of those seated at the first table from learning by the movements of the operator's hand what point of the alphabet he had reached. The other precautions described seemed, however, as M. Richet points out, sufficient to exclude all considerations of this kind.
[55] Rev. Phil., Dec. 1884; see also S.P.R., vol. ii. pp. 247 et seq.
[56] It happened on another occasion under these conditions that the right little finger was slightly affected when the left little finger, which had been selected, was so in a more decided manner.
[57] An experiment of another kind, the description of which is here omitted, had been made on the morning of this day.
[58] An account of these experiments is also contained in an article by M. Richet in the Revue de l'Hypnotisme for February 1888.
[59] M. Richet also took part in these two experiments.
[60] It is not stated whether the hour of the experiment was chosen by lot, but this precaution was taken in many of the earlier experiments.
[61] An account of these experiments was also contributed by M. Richet to the Revue de l'Hypnotisme, Feb. 1888.
[62] Revue Philosophique, February and April 1886. A translation of these accounts is given in the Proc. S.P.R., vol. v. pp. 222, 223.