It sometimes happens that the hypnotic does not remember all that occurred during the hypnosis. Thus matters of no interest to him escape his notice just as they would if he were awake. In some cases the cause of the amnesia cannot be ascertained. I said to a hypnotic, "In five minutes time you will call out 'Ha!' three times." He did as he was told, but did not afterwards remember that he had called out. But apart from this, several states of consciousness may occur, so that a subject in one state does not know what occurred when he was in another. From this Gurney supposes two stages of hypnosis, distinguished from each other by completely different memories. The old magnetizers described such stages. Gurney distinguishes two stages, a and b. In stage a the subject knows nothing of stage b; and in b nothing of a. It cannot be disputed that in some persons several sharply divided states of consciousness may exist, apart from the waking consciousness; this is also affirmed by Krafft-Ebing, Max Dessoir, Pierre Janet, and others; but I think it erroneous to speak of it as universal. I cannot take upon me to decide whether Gurney, who was an excellent experimenter, may not have used suggestion unconsciously.

But I consider it necessary to emphasize the fact that I have recently seen cases in which separate states of memory have apparently arisen spontaneously; they may, of course, have been brought about by auto-suggestion or indirect suggestion. We shall see later on, when discussing change of personality, that such chains of memory can be easily induced directly by suggestion.

But apart from these cases of hypermnesia it is characteristic that in the deeper hypnotic states not only the events that have taken place in earlier hypnoses are remembered, but also the events of waking life. On the other hand, in the waking state the events of that state alone are remembered. This state of things is named "double consciousness" (double conscience in the broad sense of the term). It was evidently well known to the old mesmerists - Kluge and Deleuze, for example - and was also observed by Braid.

The state of double consciousness is also found under pathological conditions. One of the best-known cases was published by Azam. The life of the patient, one Felida by name, for nearly thirty years was divided into certain periods - a, b, c, d, e, f. In the periods a, c, e (normal condition) she remembered only what had happened in them; in the periods b, d, f (second condition) she remembered what occurred in these periods, as well as what had happened in the periods a, c, e. The normal state was a, c, e, while the pathological was b, d, f. Osgood Mason has published a similar case which was watched for ten years; a remarkable feature of this case was that the patient, a woman, was thoroughly acquainted with the medical history of Felida (Azam's case). This may, perhaps, account for her illness. Many other similar cases have been published. Max Dessoir's thoughtful work on the "Doppel-Ich" contributes much to the elucidation of this question of double-consciousness; he shows that indications of such a splitting of the consciousness are much more common than has hitherto been believed; he refers us to examples in dream-life and in pathological states.

I shall return to the double consciousness with more detail in the theoretical part of this book.

One phenomenon which I have often observed depends on memory in the later hypnoses. If a whole series of scenes is suggested to a subject in hypnosis a very slight impulse suffices to cause the whole panorama to pass before him again in a later hypnosis. A hypnotic imagines himself hunting a lion; he kills the lion and devours it; and then by suggestion he is turned into a general, and then into a child. On awakening he forgets all these things. In a later hypnosis he hears an unexpected noise, which he immediately believes to be the roaring of a lion. In consequence he goes through all the scenes again, without omitting the smallest detail. This incident may be accounted among the indirect suggestions, since the auto-suggestion was aroused by an accidental circumstance.

The case observed and quoted above by Mabille, in which a person induced haemorrhage by auto-suggestion, after it had once been induced by external suggestion, belongs to the same category. The subject separated herself, so to speak, into two persons, one of whom made suggestions to the other, as is proved by the conversation which she carried on with herself.

The subject's recollection of all that he has experienced in earlier hypnoses is most important. The possibility of hypnotic training depends upon the strength of the recollection, which is also a frequent source of error in new experiments, since they are easily spoiled by memory of the earlier ones. I say to the hypnotic (X.) "You will now raise your left leg." X. does so. While I make the suggestion I unintentionally take hold of his right hand. When, in a later hypnosis, I again take hold of his right hand, X. lifts his left leg. Evidently he remembers the first event, and regards the taking of his hand as an order to lift his leg. It is probable that the new reflexes which Born and others thought they had discovered, and which I have mentioned before, came about in this way.

Although opinions may differ as to whether hypermnesia may be induced in hypnosis, by means of suggestion, the possibility of inducing errors of memory (paramnesia), or failures of memory (amnesia), can hardly be doubted; Bertrand collected many observations on these points. These memories may consist of former perceptions; the suggestive influence of these former perceptions has often been observed: by means of them the subject may be completely deluded about his former experiences. Since these suggestions have a certain retroactive force, they are called retroactive suggestions; or as they are concerned with supposed sense perceptions, they are sometimes termed retroactive hallucinations, either of a positive or negative order, according as a new memory is created or an old one annulled.