tained that the primary function of Rector, the control in Mrs. Piper's work, was to inhibit certain thoughts from coming through. Those thoughts might be those of communicators in the vicinity, so to speak, who either wished to send messages or could not hinder their passage, if any attention was paid to them.

The telephonic analogy is especially apropos here. The telephone wire cannot prevent the carriage of whatever voice is used near it and a communicator in rapport with the "open door" or organism that serves as a telegraph or telephone wire will inevitably transmit his thoughts and it requires some one or more personalities to prevent this, or to prevent the irrelevant thoughts of a given communicator from going through. Every nerve center might act as a medium in a purely mechanical manner unless some one can prevent it. Either the communicator or the control has to inhibit this tendency, and probably it is most frequent that others have to perform this function while the control and the communicator manage to convey only the right thoughts of one trying to transmit messages.

This rapport of which I speak may consist in the medium's attention to a given personality. I had a sitter present with Mrs. Chenoweth. All at once a certain personality began to make love at a great rate to the sitter, if I may interpret affectional messages in this manner. I thought it might be some deceased lover of the lady present, as I knew her husband was living and at home only a few blocks away. The medium did not know who was present. I remained silent and let things take their own course. At the end of

the sitting the pet name of the man who was communicating came as he signed it to his living wife, and I saw who it was, as he had been a communicator before. He was a well known man. I asked the sitter if it was relevant and she said it had no meaning whatever to her and did not know who it was. I did not

tell her, but wrote to the communicator's wife and found that all the incidents mentioned by him were correct and good evidence of his identity. I afterward learned that he had been an intimate friend in the family of the sitter, so that his appearance was relevant. But what he said was wholly irrelevant to the sitter, though relevant to his wife and characteristic of previous messages. The next week I asked the control, Jennie P, why this person was communicating, and her answer was that he had not communicated, but was "in the room" looking on. She said the sitter's father and mother were communicating. I told her that not a word had come from them, but that this well known person not only gave his pet name but told incidents in proof of his identity. The situation surprised the control and she said they did not know this on their side at all, but thought that the father and mother were communicating.

Before explaining this take another incident. I was trying to get the contents of a sealed letter. Mrs. Chenoweth started into the trance and was only in the subliminal stage of it when she saw and named the lady who had left the sealed letter, an absolute and unknown stranger to her, and then she remarked that she saw Dr. Hodgson standing by her and giving her directions as to how to proceed with her message, repeating the words she thought or heard Dr. Hodgson saying. Suddenly she finished the sentence with a sort of jerk, having repeated it very slowly, and reached for the pencil, and Dr. Hodgson began with the automatic writing, saying that "here he was writing, though he had not intended to do it."

Now what took place probably in both these incidents was this. The subliminal saw, on the one occasion, the well known man mentioned, and on the other Dr. Hodgson, and simply turned her attention to them. This put her in rapport with them and their thoughts

and her automatic machinery began to write out their thoughts. If we are talking to a friend in social group, but turn the attention of the hearing to another, though we still avoid betraying our action to the friend with whom we are talking, we will hear the talk of the person we are listening to and not that of the person we are looking at. We have in the two incidents a psychological law exactly that which I have indicated in the conversation with a friend. Attention is the cause of rapport. Once that rapport is established, the automatic machinery of the medium will reproduce the thought which its attention has enabled it to receive.

Now it is the prevention of such anomalies that the control must cause. Or, if not the control, other personalities associated with him or her. The automatic machinery is such that it must respond like a telephone wire to the current. The whole process must be organized and protected in an intelligent way to make systematic communications possible. Then they must at the same time prevent the occurrence of hysteria and obsession. Their work must be done between the two extremes of getting through no message at all and causing insanity to the medium. Any one can indulge his imagination as he pleases on the complications of such a situation. But the process is not one controlled as easily as we control our own speech. It suffers from liabilities of all kinds and this is no place to analyze or develop them fully, either with or without the facts. I can only indicate that twenty-five years of records have produced the facts on which this outline is based.

When the pictographic process is added to this we have still greater complications. The control receives the communicator's thought in the form of phantasms or hallucinations and has to interpret them. The accuracy of the interpretation will depend on the extent to which the mental imagery of the communicator is reported to the control in correct form or in remote

symbols. If the symbols are remote, they will cause all sorts of error in the interpretation. I have witnessed instances in which the medium had great difficulty in finding out what the meaning was of very clear phantasms, and often the sitter or person for whom they were intended could not suspect their meaning until further imagery was transmitted and the message translated by the medium in various ways, often not altogether clear to him.

Now imagine how this would be complicated in the double control, or "driving tandem" as we have called it. A double distortion might take place before the phantasm came to the subconscious of the medium. The communicator's thought becomes a phantasm to Jennie P and she transmits this to G. P. who describes what he sees, though he must do this with the subconscious mechanism of the medium and have his ideas modified by the transmission. How do we ever get anything accurate at all? But this matter of accuracy aside, the main thing of interest is that the process of communicating is not like our own, but the transmission of symbolic phantasms, perhaps by a telepathic process, through two or more minds before it reaches the sitter, and perhaps often through half a dozen or more minds. No physical or neural machinery is employed until the message reaches the subliminal of the medium and we may assume that from that point on the process is like our own. But its initial stage has no resemblance to anything we know except the phantasms which sometimes occur in telepathic phenomena.

Let me briefly summarize the conditions affecting the process of communications between the dead and the living, and in connection with them the main elements of the process, so far as they are known. We, in fact, know very little of them, and such as we do know are barely general outlines of a process which is not especially familiar to normal life.

1. There is a state of dissociation in the medium, some interruption of the normal relation of his or her own consciousness with the organism.

2. Rapport with a transcendental world, whether that be of incarnate or discarnate consciousness. In hypnosis and secondary personality this rapport is usually with the physical world.

3. In some cases a trance on the part of the medium, shutting off the influence of normal consciousness upon the machinery of expression.

4. In some cases the retention of normal consciousness, but the establishment of rapport with the transcendental so that messages may be received and interpreted and then expressed normally.

5. In some cases the interpretation of symbolic messages and consequent liability to distortion and misinterpretation on the part of either medium or control.

6. The existence of a control or guide through whose intervention all messages have to be effected. This control may be single or plural.

7. The existence of pictographic imagery representing the transformation or transmission of the communicator's thoughts into phantasms in the mind of control or medium.

8. The description or interpretation of these phantasms by the control so as to make them intelligible, when they are not self-interpretable, to the sitter.

9. The action of the control on the automatic machinery of the medium either by virtue of echolalia or through the intelligence, conscious or subconscious, of the medium.

10. The inhibition of intruding agencies in order to make the communications systematic and rational.

All these facts show how different the process is from that which we imagine it to be. There are no superficial resemblances or analogies to the intercourse and expression of normal life. If we then add to this