What do they prove? We must begin, as in all questions of this kind, by entertaining a certain distrust of the medium. It goes without saying that all mediums, by the very nature of their faculties, are inclined to imposture, to trickery. I know that Colonel de Rochas, like Dr. Richet and like Professor Lombroso, was occasionally hoaxed. That is the inherent defect of the machinery which we must perforce employ; and experiments of this sort will never possess the scientific value of those made in a physical or chemical laboratory. But this is not an a priori reason for denying them any sort of interest. As a question of fact, are imposture and trickery possible here? Obviously, even though the experiments be conducted under the strictest supervision. However complicated it may be, the subject can have learned his lesson, and can cleverly avoid the traps laid for him. The best guaranty, when all is said, lies in his good faith and his moral sense, which the experimenters alone are in a position to test and to know; and for that we must trust to them. Besides, they neglect no precaution necessary to make imposture extremely difficult. After taking the subject, by means of transverse passes, up the stream of his life, they make him come down the same stream; and the same events pass in the reverse order. Repeated tests and countertests always yield identical results; and the medium never hesitates or goes astray in the labyrinth of names, dates, and incidents.[5] Moreover, it would be requisite for these mediums, who are generally people of merely average intelligence, suddenly to become great poets in order thus to create, down to every detail, a series of characters differing entirely one from the other, in which everything--gestures, voice, temper, mind, thoughts, feeling--is in keeping, and ever ready to reply, in harmony with their inmost nature, to the most unexpected questions. It has been said that every man is a Shakspere in his dreams; but have we not here to do with dreams which, in their uniformity, bear a singular resemblance to fact?