Here the problem becomes complicated. I will not distinguish, as Ochorowicz does, fraud committed in the waking state from fraud committed in the trance or second state. The psychological mechanism is the same in both cases, and appears to me to depend upon self-suggestion, or what has been called _monoïdeism_, that is to say, the mind is invaded by one idea, which ends by stifling all others, and by realising itself: it is, in reality, a phenomenon analogous to that determined by suggestion. It is in unconscious or involuntary frauds, that the psychological disaggregation of the medium which Janet has studied, is best observed. These frauds present phenomena which are without interest from a medianic point of view. What is the mechanism of unconscious and involuntary frauds? It appears to me to be the following: the subjects—they may have been good mediums in their day—who commit this kind of fraud sit down to the table, or give a seance in view of obtaining supernormal phenomena. But the production of these phenomena is often difficult, sometimes impossible. Immobility, expectation, and obscurity act powerfully upon the nervous system of these mediums, and particularly so when they are hysterical. They determine the trance; the desire for the phenomenon becomes a fixed idea, and then a self-suggestion. If the supernormal phenomenon delays, the inferior strata of the consciousness—_whose morality often differs greatly from that of the active personal consciousness_—realises it normally. In the same way, even if the sensitive does not fall into the trance state, there is, nevertheless, a particular state manifested which is not sleep, neither is it the full, genuine waking-state. The active and voluntary personal element of the consciousness, as well as the judgment, becomes weakened. The sphere of the personality is reduced, and personal activity gives place to automatism. Every degree between conscious and involuntary fraud and pure automatism is to be met with. Therefore, it is prudent to take measures to guard against fraud with all subjects who become entranced, or with those who become somnolent in obscurity, silence, immobility, and expectation; but we should be frank with our sensitives: let us not offer, in ourselves, an example of dissimulation to the medium; neither must we let him have the impression of not being controlled: this would be to expose him to a temptation, all the greater in that his personal power of volition is weakened. Add to this, that we do not in the least know what influence the mental state of the experimenters has upon the medium, although some kind of influence appears to me to exist. We do not know to what extent an ill-founded certitude of fraud can be responsible for its birth. Ochorowicz says on this subject:—