which might readily be committed spontaneously, the subject makes no comment upon it. If, however, he has been told to do something ridiculous, he is usually a little ashamed of his act, and looks silly and embarrassed; or, if asked why he did such a foolish thing, he invents a justification of some kind, and these excuses are often exceedingly amusing.
It is possible, moreover, to give rise to post-hypnotic illusions and hallucinations of the various senses. Hallucinations of sight are perhaps more readily provoked than any other kind. A person may be made to see a rose, a bright light, a cat, or a devil. In short, in some persons almost every imaginable visual hallucination may be provoked. Binet and Féré have tried to demonstrate the peripheral character of visual hallucinations, by showing that such hallucinations are doubled when the patient looks through a prism. If, for example, a person has an hallucination of a rose, he sees two roses on looking through a prism. Bernheim has shown conclusively that this discovery of Binet and Féré is the result of an imperfection in their manner of experimenting, and that in reality a prism produces no effect whatever on the hallucination. In experimenting on hysterical patients, however, it is very easy to be deceived with reference to this point, for when the prism is placed before their eyes they see at once that everything looked at through it appears double, and conclude, with hysterical shrewdness, that the hallucination ought to be doubled likewise. In other words, they either consciously or unconsciously apply their newly derived knowledge of the effect of a prism to the hallucination.
This assertion is supported by the following facts: If an hallucination be called forth by suggestion in a hysterical patient, in a room which is sufficiently dark to make the objects it contains nearly or quite invisible, the hallucination is not doubled in looking through a prism for the first time, for the subject is unconscious of the fact that the glass through which he is looking has the property of doubling the image of a real object. Repeat the experiment in a light room, and the patient will state that she sees the hallucination double. If the subject of the trial be an unsophisticated child, and not an hysterical woman, the hallucination is in every instance single.
The duration of post-hypnotic hallucinations varies considerably in different cases, but is usually not greater than a few minutes. Instances are, however, recorded where they have lasted hours, and even days. It must be remembered that post-hypnotic hallucinations can only be induced in a moderate proportion of somnambulists.
I can not omit a few words about the state of the circulation and respiration during hypnotic sleep. Braid noticed that, when
