§ 2. First, then, as regards the theory of the simultaneous origination of two or more hallucinations by a distinct agent - we certainly know of no reason why a state of the agent which is telepathically effective at all, should be bound to confine its effects to a single percipient. That it generally does so confine them, may be easily explained by supposing a special susceptibility on the percipient's part, or a special rapport between him and the agent; but that occasionally the impression should extend to others, who have also been sympathetically related to the agent, may seem no very astounding fact. Now if the impression were a merely inward experience, an impression of a merely ideal or emotional kind, and did not give rise to actual hallucination, this account of the matter might be plausible enough: it would apply for instance to Mr. H. S. Thompson's case, p. 78. But it will be remembered that we have seen reason to regard the hallucination as distinctively the percipient's work - as some-thing projected by him under a telepathic stimulus; and we have found these sensory projections to take various forms according to the projector's; idiosyncrasies.

We have found, moreover, that the time during which such hallucinations may take place extends over several hours - that we cannot name an exact moment at which the telepathic message will reach consciousness, or externalise itself to the sense. It becomes, then, extremely improbable that two or more persons should independently invest their respective telepathic impressions, at the same moment, with the same sensory form; that they should all at once see the same figure, or hear the same sound, in apparently the same place. We should expect to find one of them embodying it in sound, and another, perhaps half an hour later, in visible shape; or one of them embodying it in sound or shape, and another only conscious of it as an inward idea; and so on. And for divergences of this sort, the evidence, though it exists, is small in amount.

But this is not all. On the theory that joint telepathic hallucinations are all exclusively and directly due to a distant agent, there is one thing that we should not expect to happen, and one thing that we should expect to happen. (1) We should not expect the group of percipients to include anyone who was a stranger to the agent; or who was not personally in such relations with the agent as would have rendered it natural for him, had he chanced to be alone at the time, to suffer the same telepathic experience. Nevertheless, cases exist where such an outside person has shared in the perception. And (2) we should expect that in a fair proportion of cases two or more percipients would share the perception, though they were not in each other's company at the time. For on the theory that is being considered, there would be no virtue in the mere local proximity of the percipients to one another; the agent is supposed to affect them by dint of his respective relations to each of them, which have nothing to do with their being together or apart.

Now, in point of fact, we have a group of cases where the persons jointly affected have been apart, but they are disproportionately rare in comparison with the experiences shared by percipients who have been together; and in several of them, moreover, B and C, the two percipients, were near each other, and had been to some extent sharing the same life - conditions which may have had their share in the effect (see pp. 514-5). However, the existence of this type might no doubt be regarded as an argument for the occasional production ab extra of several similar and simultaneous hallucinations; and our few specimens may conveniently be cited at once.