635. The hypothesis which I suggested in Phantasms of the Living itself, in my "Note on a possible mode of psychical interaction," seems to me to have been rendered increasingly plausible by evidence of many kinds since received; evidence of which the larger part falls outside the limits of this present work. I still believe - and more confidently than in 1886 - that a "psychical invasion " does take place; that a "phantasmogenetic centre" is actually established in the percipient's surroundings; that some movement bearing some relation to space as we know it is actually accomplished; and some presence is transferred, and may or may not be discerned by the invaded person; some perception of the distant scene in itself is acquired, and may or may not be remembered by the invader.

But the words which I am here beginning to use carry with them associations from which not the scientific reader alone may well shrink in disgust. I am falling into the language of a "palaeolithic psychology" - into the habits of thought of the savage who believes that you can travel in dreams and infest your enemy as a haunting spirit. Fully realising the offence which such expressions must give, - the apparent levity of a return to conceptions so enormously out of date, - I see no better line of excuse than simply to retrace to my reader the way in which the gradual accretion of evidence has obliged me, for the mere sake of covering all the phenomena, to use phrases and assumptions which go far beyond those which Edmund Gurney and I employed in our first papers on this inquiry in 1883.

636. The facts of the case, then, are briefly as follows. When in 1882 our small group began the collection of evidence bearing upon " veridical hallucinations " - or apparitions which coincided with other events in such a way as to suggest a casual connection - we found that the subject had hardly as yet been seriously attacked. Cases, indeed, of various kinds had been vaguely recorded; but scarcely any of these reached the evidential level on which we wished our narratives to stand. Our own collection was miserably scanty as compared with the magnitude of the harvest waiting to be reaped; but at the same time it was copious enough to indicate those types of coincidental apparition which were at once commonest and most capable of evidential treatment. These were apparitions of living persons, coinciding with some crisis which those persons were undergoing at a distance; and especially the apparitions of persons who might indeed be regarded as still living, but who were undergoing the crisis of death.

These cases, I say, were the first to attain to a number and a weight which carried conviction to our own minds, and in various papers in the S.P.R. Proceedings, and then in Phantasms of the Living, they were set forth in evidential form, and were connected with experimental telepathy, being themselves regarded as spontaneous examples, upon a more impressive scale, of these transferences of impression from one to another mind.

But at the same time there were scattered among these cases from the first certain types which were with difficulty reducible under the conception of telepathy pure and simple - even if such a conception could be distinctly formed. Sometimes the apparition was seen by more than one percipient at once - a result which we could hardly have expected if all that had passed were the transference of an impression from the agent's mind to another mind, which then bodied forth that impression in externalised shape according to laws of its own structure. There were instances, too, where the percipient seemed to be the agent also - in so far that it was he who had an impression of having somehow visited and noted a distant scene, whose occupant was not necessarily conscious of any immediate relation with him. Or sometimes this "telepathic clairvoyance" developed into "reciprocity," and each of the two persons concerned was conscious of the other; - the scene of their encounter being the same in the vision of each, or at least the experience being in some way common to both.

These and cognate difficulties were present to my mind from the first; and in the above-mentioned "Note on a suggested mode of psychical interaction," included in vol. ii. of Phantasms of the Living, I indicated briefly the extension of the telepathic theory to which they seemed to me to point.

637. Meantime cases of certain other definite types continued to come steadily to hand, although in lesser numbers than the cases of apparition at death. To mention two important types only - there were apparitions of the so-called dead and there were cases of precognition. With regard to each of these classes, it seemed reasonable to defer belief until time should have shown whether the influx of first-hand cases was likely to be permanent; whether independent witnesses continued to testify to incidents which could be better explained on these hypotheses than on any other. Before Edmund Gurney's death in 1888 our cases of apparitions and other manifestations of the dead had reached a degree of weight and consistency which, as his last paper showed, was beginning to convince him of their veridical character; and since that date these have been much further increased; and especially have drawn from Mrs. Piper's and other trance-phenomena an unexpected enlargement and corroboration. The evidence for communication from the departed is now in my personal estimate quite as strong as that for telepathic communication between the living; and it is moreover evidence which inevitably alters and widens our conception of telepathy between living men.

The evidence for precognition, again, was from the first scantier, and has advanced at a slower rate. It has increased steadily enough to lead me to feel confident that it will have to be seriously reckoned with; but I cannot yet say - as I do say with reference to the evidence for messages from the departed - that almost every one who accepts our evidence for telepathy at all, must ultimately accept this evidence also. It must run on at any rate for some years longer before it shall have accreted a convincing weight.

But at whatever point one or another inquirer may happen at present to stand, I urge that this is the reasonable course for conviction to follow. First analyse the miscellaneous stream of evidence into definite types; then observe the frequency with which these types recur, and let your sense of their importance gradually grow, if the evidence grows also.

Now this mode of procedure evidently excludes all definite à priori views, and compels one's conceptions to be little more than the mere grouping to which the facts thus far known have to be subjected in order that they may be realised in their ensemble.

No more ambitious than this is my "palæolithic psychology." I merely endeavour, like my cannibal precursors, to find a formula which will somehow cover the observed facts - though with this difference, that where they find their formula easily credible, and do not care what white men say, I find my formula credible with difficulty - credible mainly just because I have heard what white men say, and because I cannot think that they have "saved the phenomena " - or have even shown much more grasp than the Stone Age possessed of the limit of cosmic possibilities.

"What definite reason do I know why this should not be true?" - this is the question which needs to be pushed home again and again if one is to realise - and not in the ordinary paths of scientific speculation alone - how profound our ignorance of the Universe really is.

My own ignorance, at any rate, I recognise to be such that my notions of the probable or improbable in the Universe are not of weight enough to lead me to set aside any facts which seem to me well attested, and which are not shown by experts actually to conflict with any better-established facts or generalisations. Wide though the range of established science may be, it represents, as its most far-sighted prophets are the first to admit, a narrow glance only into the unknown and infinite realm of law.