pt. 80). He, too, got negative results, and did not go on long enough. In fact, he failed so completely that he failed even to prove that telepathy did not exist, or that at any rate he and his colleague were completely devoid of telepathic ability. Only 605 experiments were made, and only 284 complete successes were obtained. Now this is very sensibly below the most probable number (302); but, as Dr. Troland observes, an abnormal de- ficiency is quite as significant of something other than chance as an abnormal excess. It may mean the presence of some factor that inhibits success, and if this can be established, it is just as supernormal as one that produces success. However, Dr. Troland does not hold that in his experiments the deficiency is sufficiently great. He has not observed how it arose. His total figures were arrived at by lumping together two sets of experiments. In one of these the stimulus shown to the " agent," to which the " percipient " was to react by pushing an instrument either to the right or to the left, was exposed for 30 seconds; in the other, for 15 seconds. Now in the former series there was no de- ficiency of right reactions; 129 successes out of 249 experiments are slightly above the probable number, 124. The whole of the deficiency was incurred in the is-second series, which yielded only 155 successes out of 354 experiments, instead of a most probable 177. As the only difference between the two series was in the duration of the exposure, the idea easily suggests itself that the i5-second exposure was too short to enable the percipient to react rightly. And not only that; it seems to have positively inhibited the right reaction, presumably by inducing an " anxiety-neurosis." In other words, if the " agent," or more probably the " percipient," got " flustered " by the shortness of the exposure, his very knowledge of the right reaction would lead him to make the wrong one. Thus a marked deficiency in correct responses over a long series might imply as much super- normal knowledge, and yield as good evidence of telepathy, as a marked excess; much as it is implied in the " negative hallu- cination" of a hypnotic subject that he both sees and does not see, the object of the hallucination, and indeed that he must see it (subconsciously) in order to avoid it. Again, however, the series of experiments was not long enough to make the appeal to the calculus of probabilities decisive. For the present, therefore, it is best to conclude that the reality of telepathy is not yet either proved or disproved: the evidence is just about enough to keep it alive as a hypothesis.
Trance. The phenomena of trance continue to be studied, and although Mrs. Piper, the most famous " medium " of this type, was pensioned by the S.P.R. and retired so long ago as 1910, she has no lack of successors. Indeed, the great majority of the customers of " psychics " frequent trance-mediums. Their manifestations continue to be much the same; entranced psychics become obsessed by one or other of their regular " controls " usually grotesque personages that cannot be identified, and may fairly be suspected of being creations, at least to a large extent, of the medium's subliminal imagination. There are poured forth (in the " good " sittings one hears about) masses of details about the sitters and their concerns, often hesitant, incon- clusive, vague, sometimes wrong, often non-significant, but sometimes so startlingly apposite as to shake all but the sturdiest scepticism. The evidence presented in Sir Oliver Lodge's Raymond was obtained in this way; Miss Radclyffe-Hall and Lady Troubridge have recorded similar evidential sittings with Sir Oliver's chief medium, Mrs. Leonard, in S.P.R. Pro- ceedings, pt. 78. Mrs. Sidgwick produced a final and monumental review of the Piper case in S.P.R. Proceedings, pt. 71 (1915).
As regards the theoretic interpretation of these trance- communications, the tendency, even among those most inclined to believe that they convey authentic messages from the de- parted, is to complicate the process of communication. It is recog- nized more and more that there have to be reckoned with, not only the medium, with his natural limitations of faculty and
- raining, but the medium's " subliminal " or subconscious, the
nedium's " controls," who are supposed to transmit the messages
om the communicator proper, and possibly the effects of ab-
ormal conditions, not only in the medium (trance-personality)
but also in the " control " and " communicator," owing to the effort to communicate. It is evident that these complications may account for many errors and obstructions; but they detract pro lanto from the authenticity of the actual communications.
Automatic Writing. Automatic writing continues to flourish and to furnish psychical researchers with large masses of raw material. But its quality is not equal to its quantity, and its interest is for the most part psychological rather than evidential. Nevertheless a few cases of automatism laying claim to scientific importance may be noted. Undismayed by the failure of Mrs. Verrall to gat, through automatic writing, at the contents of a sealed letter left, before his death, by Frederic Myers with Sir Oliver Lodg2 (cf. S.P.R. Journal, Jan. 1905), many of the lead- ers of the S.P.R. continued to work at cross-correspondences, and the results of their labours bulk large in the Proceedings of the society 1911-9. They discovered some curious cases among the writings of their automatists, the most remarkable perhaps being that entitled The Ear of Dionysius (1917), which was worked out by Mr. Gerald Balfour, and held to indicate the post-mortem agency of Prof. Verrall. But unfortunately the value of the coincidences on which the method relies is not capable of exact determination, and the whole method of proving spirit- identity by cross-correspondences is too literary and recondite to be appreciated without an intellectual effort, and so fails to impress the ordinary man. The automatic writing of a Dublin lady, Mrs. Travers Smith, excited some interest, both on account of the enormous speeds attained in its method of production (a planchette travelling over an alphabet under glass), and because of the claim that communications had been received from Sir Hugh Lane, before it was known that the Lusitania had been sunk, and that he had been one of the victims of this outrage. The case is narrated in Voices from the Void (1919).
Great interest was excited when Mr. Bligh Bond, in his Gate of Remembrance (Oxford, 1918), announced that he had been guided in his excavation of Glastonbury Abbey by the automatic writings of a friend who produced copious communications, largely in very debased Latin, from a number of the monks who had inhabited the Abbey from the nth to the i6th century, and had revealed the correct location and dimensions of the Edgar Chapel, though all the extant antiquarian evidence had made these statements seem quite improbable. Mr. Bligh Bond also had the courage to print in the first edition of his book similar predictions about the Loretto Chapel, of which the remains had not then been found: when, after the war, excavation was re- sumed, these also were found to be correct substantially i.e. allowing for the facts that the original script was in some points capable of more than one interpretation and that the excavators did not always hit upon the right one. Cases of practically valuable information received in a supernormal manner are extremely rare, and Mr. Bligh Bond's is one of the best of them.
Physical Phenomena. To pass from automatic writing to physical phenomena is to pass from the least to the most con- tentious of the subjects that concern the psychical researcher, from a region where the facts are" admitted and the interpretation alone is in dispute, to one where fraud has to be guarded against at every step and where all the facts are suspected by some to be due to it. Not that fraud is excluded in the former case: automatic writing can be simulated (like anything else) and with a little luck and ingenuity organized deception can be effectively practised with great success, as is amusingly shown in E. H. Jones's The Road to Endor (1920), describing how two British officers beguiled the tedium of their captivity and fooled both their comrades and the Turkish officials in charge of their prison camp. In fact, fraud is so easy that nothing depends on it w r it is recognized by all competent inquirers that the whole value of automatic writing depends not on the mode of production but on the evidential character of the contents. In dealing with physical phenomena, on the other hand, the elimination or discounting of fraud is the primary consideration; the more so that fraud is certainly abundant, and that the conditions seem designed to facilitate it. This should be recognized by both sides, and should be no reason for refusing absolutely to investi-
