The foregoing instances may perhaps suffice to show that a purely psychical account of these joint experiences - as due either partly or wholly to a thought-transference between the percipients - is at all events possible; and that acceptance of the phenomena as genuine, i.e., as percepts truly described, does not imply any materialistic theory of phantasmal beings who travel about through space (sometimes in their carriages) on their own account. And possibly a certain number of my readers may further agree with me in supposing some, at any rate, of these cases to have been in their inception purely subjective, and will not feel the need of invoking for them an unknown or post-mortem "agency," however little disposed to rule the possibility of such agency out of court. I cannot, indeed, deny a certain force to an objection which Mr. Myers urges, that we know of no instances where a hallucination which can be connected with insanity or other distinctly morbid conditions in the person impressed, and which is thus quite clearly proved to be purely subjective, has become collective in the way supposed. But then neither do we know of instances where a person in one of these morbid conditions has exercised any other form of telepathic influence.

We have no instances of telepathic impressions of the deaths of dying lunatics. The ultimate conditions of telepathic agency are as little known to us as the ultimate conditions of telepathic percipience; and transient hallucinations of the sane, such as those of the preceding examples, differ so greatly in their nature and ostensible conditions from the types of hallucination to which Mr. Myers points as never transferred, that it seems rash to assume that they may not differ also in the particular point of transferability. At any rate, whatever the difficulties of that view, it is one that may be provisionally entertained by those who see equal difficulties in any other; and whatever my own surmises as to future discovery may be, in the present state of the evidence I feel as much bound here to press the theory of thought-transference, before admitting causes of an obscurer kind, as in a former chapter to press the theory of unconscious physical indications before admitting the reality of thought-transference.

The degree in which the infectious character may exist is very hard indeed to determine; for the majority of hallucinations (purely subjective and telepathic alike) occur to persons who are alone - silence and recueillement being apparently favourable conditions; and we have no means of knowing how many of these hallucinations might have been shared by some one else, if some one else had happened to be present at the time. All that can be said is that, taking the whole class of transient hallucinations of the sane, the cases where the experience has been shared by a second person appear to be more numerous than those where a second person has been present, awake, and rightly situated, and has not shared the experience. Nor, again, can I at all adequately explain why these phenomena should be a form of mental impression specially liable to spread to neighbouring minds. That those of them which are telepathically produced in the first instance should have a tendency to spread in this way may appear, perhaps, less remarkable, if we remember that a telepathic impulse, as such, seems sometimes to have very distinct and peculiar physiological effects; witness Mrs. Newnham's exhaustion (p. 51) in experiments where the ideas conveyed were in themselves of a quite unexciting sort.

But as regards the transference in purely subjective cases, all I can suggest is that sensory hallucinations, and especially the occasional hallucinations of sane and healthy people, are to begin with and in themselves very peculiar things; and that a fresh peculiarity, meeting us in something that we do not completely see round or understand, is less staggering than if it met us in something of which we have held our knowledge to be complete. At any rate the fact, if admitted, that purely subjective hallucinations may spontaneously become collective, greatly simplifies the consideration of the collective cases whose origin is traceable to an external "agent." The appearance of an absent person's figure to several spectators at once has had in it something specially startling; and when associated with the idea of death, it has almost inevitably suggested a material or "etherial" spirit - an independent travelling ghost. But as soon as the experience is analysed, it is found to involve nothing new or antagonistic to scientific conceptions.

In being connected with the absent person, it is merely on a par with other specimens of telepathy - e.g., many of those cited in the preceding chapters: in being collective, it is merely on a par with other specimens of hallucination - e.g., some of those already cited in this chapter. Still, though a telepathic impulse from an absent person may not be an essential condition, it may be, and I believe is, an exceptionally favourable condition for a collective hallucination. And I now proceed to the final group of examples of which that condition is the distinguishing mark.