All these drawbacks are acknowledged to exist, and are allowed for, and, as far as possible, provided against, by the very fair-minded people who have conducted this inquisition.  Thus Mr. Henry Sidgwick, in 1889, said, ’I do not think we can be satisfied with less than 50,000 answers’. {195} But these 50,000 answers have not been received.  When we reflect that, to our knowledge, out of twenty-five questions asked among our acquaintances in one place, none would be answered in the affirmative:  while, by selecting, we could get twenty-five affirmative replies, the delicacy and difficulty of the inquisition becomes painfully evident.  Mr. Sidgwick, after making deductions on all sides of the most sportsmanlike character, still holds that the coincidences are more numerous by far than the Calculus of Probabilities admits.  This is a question for the advanced mathematician.  M. Richet once made some experiments which illustrate the problem.  One man in a room thought of a series of names which, ex hypothesi, he kept to himself.  Three persons sat at a table, which, as tables will do, ‘tilted,’ and each tilt rang an electric bell.  Two other persons, concealed from the view of the table tilters, ran through an alphabet with a pencil, marking each letter at which the bell rang.  These letters were compared with the names secretly thought of by the person at neither table.

He thought of The answers were

1.  Jean Racine 1.  Igard

2.  Legros 2.  Neghn

3.  Esther 3.  Foqdem

4.  Henrietta 4.  Higiegmsd

5.  Cheuvreux 5.  Dievoreq

6.  Doremond 6.  Epjerod

7.  Chevalon 7.  Cheval

8.  Allouand 8.  Iko

Here the non-mathematical reader will exclaim:  ’Total failure, except in case 7!’ And, about that case, he will have his private doubts.  But, arguing mathematically, M. Richet proves that the table was right, beyond the limits of mere chance, by fourteen to two.  He concludes, on the whole of his experiments, that, probably, intellectual force in one brain may be echoed in another brain.  But MM.  Binet and Fere, who report this, decide that ’the calculation of chances is, for the most part, incapable of affording a peremptory proof; it produces uncertainty, disquietude, and doubt’. {196} ’Yet something is gained by substituting doubt for systematic denial.  Richet has obtained this important result, that henceforth the possibility of mental suggestion cannot be met with contemptuous rejection.’

Mental suggestion on this limited scale, is a phenomenon much less startling to belief than the reality, and causal nature, of coincidental hallucinations, of wraiths.  But it is plain that, as far as general opinion goes, the doctrine of chances, applied to such statistics of hallucinations as have been collected, can at most, only ‘produce uncertainty, disquietude, and doubt’.  Yet if even these are produced, a step has been made beyond the blank negation of Hibbert.