writing being illegible just to that degree which at first baffles the reader, but which afterward leaves no more doubt as to its having been correctly deciphered than if it were print. And there the line indubitably stood: 'A little more than kin and less than kind.' Now, as neither of us had been thinking of this line, or of any line (for we had been wholly occupied with the strag- gling movements of the instrument), the result, though not de- monstrative, is at any rate strongly suggestive of a true under- ground psychosis."
At other times the information conveyed is at once true and quite unknown to the subject. Some of these cases are undoubt- edly due to the automatic reproduction of memories which can not at the time be recalled—a common phenomenon in all forms
of automatism. Thus, in the case of B , to which I shall refer
at greater length hereafter, it was stated that a man named Parker Howard had lived at a certain number on South Sixteenth Street, Philadelphia, Upon going to the house, I found that a man named Howard—not Parker Howard, however—had lived there some time, but had moved away about two months before. More- over, the whole Howard incident proved to be mythical; no such
person as Parker Howard ever existed. But B told me that
after his hand had mentioned the name, and before the address was given, he stepped into a shop and looked through a directory for the name. Probably, as he glanced over the list of Howards, his eye had fallen upon the address which his hand afterward wrote, but he had no recollection of it.
Many other cases are certainly due to accidental coincidence.
B , for example, wrote long accounts of events happening at a
distance from him, which were afterward found to be in the main correct; but that this was a mere matter of chance was abundantly
proved to B 's own satisfaction. The chances of coincidence
are much increased by the extremely illegible character of much of the script, which leaves wide room for "interpretation." I can not but suspect that the "anagrams" sometimes written auto- matically often owe their existence to this kind of "interpreta- tion." Yet, after making all allowances for coincidence and for- gotten memories, nearly all investigators admit that there remains a residuum which can not plausibly be explained by any accepted theory. I can not discuss this residuum here; it is enough to point to its existence, with the caution that no theory can be regarded as final unless it can explain all the facts.
The importance of this material from a psychological point of view can not be overestimated. If the man's hand can write messages without the co-operation of the man's consciousness, we are forced upon the one horn or the other of a very perplexing dilemma. Either these utterances stand for no consciousness at