Every asylum doctor recognizes this phenomenon. In acute mental affections the discernment alters with the illness. At first there is a sort of duel between the healthy and diseased brain activity. As time goes on these two activities gradually become more reconciled to one another, to the detriment of lope. The diseased chain works more superconsciously and the healthy chain more hypoconsciously. Thus a patient imagining him-self to be God or a king is quite willing to undertake menial duties, and another who believes that he is starving or dying eats with a regal appetite.

These facts led me to study the negative hallucinations in the insane, and I was astonished to find how frequently this phenomenon really occurs. I first reported on this subject in 1889 in the Verein schweizerischer Irrenaerzte (the Association of Swiss Asylum Doctors), and later in the Congress on Hypnotism in Paris,1 and gave examples illustrative of this. One has scarcely paid any attention to this hitherto, since the patients mostly only speak of the positive parts of the phenomena, unless one asks them especially.

It is interesting to analyze the process of negative hallucination in the light of Semon's Mneme theory. The engram is that which I formerly termed dynamic trace (Ribot). 2 The engrams are not produced in reality by single, completely isolated stimuli, but by associated stimuli complexes, and they form in this way associated engram complexes. As is known, the association takes place either in point of place or in point of time.

Engrams can be associated simultaneously in point of place, according to Sernon. Simultaneous engrain complexes exist - as, for example, especially those which we receive by means of our sense of sight. They can also be associated in sequence, as is particularly the case with the sense of hearing and in onto-genia. The chief difference between simultaneous and successive engram complexes is that the simultaneous complexes are ambiguously linked as equivalents, while the engrains in sequence are linked polarly as unequivalents. In the sequence a-b, a acts much more strongly on b than b does on a, thus the backward action is weaker. If I ask "woh," instead of "how," for example, the person asked does not realize at once that "woh" is the reverse of "how." It often occurs in the sequence of engrams that two or more similar engrains are more or less equivalently associated with one that has preceded. In such a case Semon speaks of dichotomy, trichotomy, etc.; but as two succeeding engrams cannot be eepborized simultaneously from the one which has preceded, that which Semon calls "alternative eephoria " takes place. This is, that either the one or the other of the succeeding associated engrams are eepborized. In such an alternation it will depend on the frequent repetition of the one branch that this one is more frequently eephorized than the others, Thus, for example, in the second verse of the text of a song, if this has two versions, as is the case in Goethe's well-known song:

1 "Compte rendu," by Berillon, p. 122. (Paris; O Doin, 1800.) 2 A. Fnrel, "Memory and its Abnormalities." (Zurich: Orell, Fussli and Co., 1885.)

These forms of alternative eephorias play a most important part in the laws of ontogenia and of inheritance. The more frequently repeated, stronger engram branch is usually alone eephorized, while the other or others generally remain completely latent. The latter may, however, be eephorized under favorable conditions - e.g., in the following generation.

On regarding the place association, one meets with something of the same sort, in so far as the whole of the same space cannot possibly be filled at the same time by two different sensations or complexes of sensations in our consciousness. For example, I cannot see the same surface of one square centimeter red and blue at the same time. Either the sensations of blue and red fill two surfaces placed side by side, or the same surface can be first seen red and then blue, or the reverse. This law, which is obviously caused by the conditions of the outer world- - that is, by the projection of these conditions in the brain - repeats itself in the so-called impressions of memory (in the ecphorized engrams), no matter whether these be hallucinated or whether they be only imagined internally.

An intelligent person suffering from recurrent insanity hallucinated that she was in a subterraneous passage and witnessed an execution, while she was lying in a bed in the dormitory. She told me later quite clearly, on being questioned by me, that while she bad this hallucination she no longer saw the whole dormitory, including the beds, and did not even hear the noise of the patients in the neighboring corridor.

If we now regard negative hallucination in the light of the doctrine of the Mneme, we are bound to admit that dichotomies occur even in simultaneous engrams. These can be alternately ecphorized, especially when the same space is occupied by different impressions of the senses. For example, I can imagine a pine tree or a fir tree in the same imagined place, but not both at the same time and in the same place. I can therefore ecphorize the picture of the pine or the picture of the fir. The power of suggestion acts on the negative hallucination, so that the ecphoria of a formerly suggested sight engram, for example, acts more strongly than the stimulation of the light which reaches the brain of the person, hallucinating through his eye. Thereby even the original direct stimulation will be outweighed by the alternating ecphoria of an old engram in the brain. The same applies to the sense of touch.

But, just as with the alternating ecphorias in sequence, the branches of the dichotomies which are not ecphorized remain latent in the brain. Such a latency plays a large part, undoubtedly, in all hypnotic and also in all psychological phenomena.

I am of opinion that it would be worth while to investigate the Mneme theory of Semon further in this direction.

One need scarcely add that hallucination is a purely cerebral process, which is just as little influenced by the laws of optics, etc., as the area of distribution of a suggested anaesthesia is influenced by the area of distribution of the peripheral sensory nerves. It is well known that a man whose finger has been amputated often hallucinates his removed finger, and that a person whose optic nerves are destroyed can have visual hallucinations for many years after the destruction. I have observed a striking case of this in the Burghoelzli Asylum in Zurich. A man had his eye destroyed by a shot thirty years previously (March, 1865) by the North American Indians. The other eye was lost soon after from a sympathetic inflammation. The man bad the most marked visual hallucinations, although he had been retinally blind for twenty-eight years (since 1867). He had hie last hallucination of sight at the end of the year 1893. In all other respects he was quite normal, and gave very clear descriptions of his visions. The autopsy undertaken later revealed complete atrophy of both optic nerves.