In the earlier part of this chapter, I consulted clearness by keeping separate the hypothesis (1) of joint and independent affection of B and C by A, and the hypothesis (2) of C's affection by B who alone is directly affected by A. Now looking back at these hypotheses in the light of the evidence, the objections (see § 2) to the assumption of independent psychical affection of B and C by A come back on us with only increased force. As long as telepathic hallucinations are rare, and lead by their rarity to the conclusion that they generally require not only an abnormal condition of the agent, but specific susceptibility in the percipient, nothing can make it seem otherwise than astonishing that two closely similar specimens of them, in connection with the same agent, should independently concern two percipients at the same moment. One might admit such an astonishing coincidence once or twice - I have suggested its application to a few cases in § 2 above;1 but it seems impossible to lay it down as a principle of explanation, by which any number of collective hallucinations may be accounted for.

No view which shrinks from assuming a local and physical presence of A, and at the same time rejects every sort of direct transference between B and C, can avoid this difficulty; and the consideration seems to me of such weight as to exclude hypothesis (1) in the form stated. I feel absolutely driven to suppose that where C's experience resembles B's, it is in some direct way connected with B's; this is the only alternative that I can see to admitting a physical basis to the percept. But this does not necessarily imply the adoption of hypothesis (2) in its crudest and most obvious form; the "direct way" need not, I conceive, be a transfer between B and C wholly unconnected with A - a transfer, that is to say, which must have equally taken place had B's hallucination been purely subjective. Though the evidence in § 5 above inclines me strongly to the opinion that sensory hallucinations, as such, are transferable things, I do not believe this to be the complete explanation of the later telepathic cases.

And I now venture to suggest that with slight modification the two hypotheses - of joint affection by A, and of direct transference between B and C - may be amalgamated; and that the amalgamation is really more probable than either hypothesis in its isolated form.

1 In all of these, however, where the two percipients were near together and had been sharing the same life, I think it probable that the experiences were not truly independent.

Where A, the distant agent, is in rapport both with B and C, it is possible to suppose that B and C are jointly and independently impressed by A, though the particular form - the hallucination - in which they simultaneously embody their impression is still an effect of B's mind on C's, or of C's on B's. The joint impression from A may be conceived as having in itself a tendency to facilitate this further effect - that is to say, psychical communication between B and C may find a readier and wider channel at the exceptional moments when they are attuned by a common telepathic influence than, e.g., when one of them is staring at a card and the other is endeavouring to guess it. But even for these cases, I think it so dangerous, in view of the apparent rarity of " psychical " affections, to assume any sort of independent psychical affection of different minds at the same moment, that I should prefer to regard A's influence on C as derived through B. And this certainly commends itself as the process where C is a stranger to A, or not a person whom it would have seemed natural that A's vicissitudes should in any way affect.1 In such cases I conceive that, while C's experience depends on B's presence or existence, and even probably on the form of B's experience when the two are similar, yet A's influence may really and truly extend to C; that in fact there is a rapport between A and C, established ad hoc by the rapport of both of them with B. B would be thus not the instigator, or not solely the instigator, but the channel, of C's percipience - the assumption being that a mind in which B holds a prominent place, such as C's, may be abnormally susceptible to an influence which abnormally impresses B. Especially would this conception relieve the difficulty of such extreme cases as Nos. 242 and 355, above; where B's part in the occurrence was to all appearance suppressed, and C, a stranger to A, was the sole percipient.2 We can scarcely doubt that the presence of B, the near relative of the supposed agent, was a condition of C's percipience; while at the same time it seems absurd to suppose that B infects C with a sensory hallucination which he himself does not experience.

We seem driven, then, to regard B as a mere channel of influence; and that is a part which there is no absurdity in supposing to be played unconsciously. For the better established facts of telepathy have familiarised us with both unconscious reception and unconscious propagation of telepathic impulses; and however unexpected, it is at least quite conceivable that the two events should take place as part of a single process - which is all that the transmission of an impulse from A to C through the unwitting B implies.

The above view, of rapport through community of mental occupation, may likewise afford some explanation of the otherwise puzzling cases where the telepathic influence exercised by A seems itself to have depended rather on local than on personal reasons; as in case 29 in Chapter V (Specimens Of The Various Types Of Spontaneous Telepathy), where the agent's form was seen by a person only slightly connected with her, in a spot in which she was known to have been considerably interested;3 or in cases where the actual percipient had little or no connection with the agent, but was situated in a place where the agent might naturally conceive some other and nearly-connected person to be; 1 or in cases where a dying person's form is alleged to have been seen by strangers in that person's old home; or in a converse case in the Supplement, a lady's veridical dream of the death of a comparative stranger in her own old home. It is not necessary that two persons should know one another, for certain daily scenes and local impressions to be deeply stamped in common on their two minds; and in this way locality might constitute an ideal bond between A and B who are apart, as we conceived that it might do between B and C who are together.

1 e.g., cases 339, 348, 350, 357 [the Wynyard case].

2 See also case 307, where A's bond, such as it was, was with B and not with C; and compare case 311.

3 It is probable that a local explanation would apply to cases 239, 343, 350.

An even further extension could be given to this idea, if we admit the supposition that A's own susceptibility may be quickened, in the way that was so strongly suggested by some of the reciprocal cases in the preceding chapter. I there pointed out (pp. 460-1 and 462) the indications afforded of a special sort of clairvoyance; telepathic, in the sense that it depends on B's living presence in the scene which A perceives; but independent in the sense that B and his surroundings are perceived while B's own state is not critical but normal - the abnormality of state being confined to A, whose extension of faculty in trance or at death makes him percipient of B, as well as the agent of B's percipience. A view akin to this has been developed by Mr.Myers [in the Note omitted in the present edition]; and the temptation to apply it to the collective cases is considerable, since it enables us to conceive the scene, and the sense of being present there, as common to the minds of A, B, and C alike; and so far as such community is a favourable condition for telepathic affection, it would explain A's power to affect the other two.

To some joint hallucinations, however (e.g., case 329, and perhaps 348), where A, the original of the phantasm, has been in a normal waking state at the time, such an explanation seems quite irrelevant; and its admissibility elsewhere must, I think, depend on our obtaining more proof than we yet have of A's reciprocal percipience, in collective cases which are clearly due to his agency. The reciprocal type having seemed, on the evidence, to be a rare if not a doubtful one, we ought to be doubly cautious of making it the ground of explanation for further and more perplexing phenomena.

And indeed any conjectural explanations of these more outlying telepathic phenomena have, I am well aware, an air of rashness and unsoundness. This may very likely be due to their being really rash and unsound; but it may also possibly be due to the fact that our view of the field before us is still very partial and dim. The duty of caution in all evidential matters does not exclude the duty of keeping the mind open to new conceptions on this threshold of new knowledge, and not allowing any hypothesis that has provisionally commended itself to become a rigid barrier, within which further facts must be forced or else disallowed. And if our central thesis stands - if "psychical" transferences from mind to mind be admitted as in verum natura - the rashness, I think, would be in attempting to set a limit to the possible implications of this admission. Its tendency, at any rate, is to give a tangible meaning to that solidarity of life which Idealism proclaims; to lead us to regard individual minds, not as isolated units, but as all in potential unity - as entering into a scheme whose relation to the telergic influence somewhat resembles that of the physical world to electricity.

And in such a scheme we need not be surprised if the manifestations of action and affinity between the parts are as sudden and shifting, and to the superficial view as isolated, as in the physical world those of electrical relations between different pieces of matter. But a far larger basis of well-attested cases is, no doubt, needed before reflections of this sort can be profitably pursued; and I will not further run the risk of inverting the relation of speculation to evidence which it has been throughout my endeavour to maintain.

1 e.g., No. 242 just mentioned; and compare No. 307.