I say at certain times advisedly; for all clear evidence of the sort seems to connect the phenomenon with circumstances of rather special absorption or excitement, sometimes even with a state of semi-trance.2 I do not know of'* In or near some Italian town, Mazzini saw a group of people standing, apparently gazing upwards into the sky. Going up to it, he asked one of the gazers what he was looking at. 'The cross - do you not see?' was the answer; and the man pointed to the place where the cross was supposed to be. Mazzini, however, could discern no vestige of anything cruciform in the sky; and, much wondering, went up to another gazer, put a similar question, and received a similar answer. It was evident that the whole crowd had persuaded itself that it was contemplating a marvellous cross. 'So,' said Mazzini,' I was turning away, when my eye caught the countenance of a gazer who looked somewhat more intelligent than the rest, and also, I thought. had a faint air of perplexity and doubt in his gaze.

I went up to him, and asked any instance where the sane and healthy A, simply by saying at a casual moment to the sane and healthy B, "There is such and such an object " (not really present, and not capable of being imposed as an illusion on some object really present) has at once caused the object to be conjured up in space before B's eyes. In the most extreme case that has come to my knowledge, where something like this has proved possible, very strong insistence and repetition on A's part, of the sort that a mesmerist employs when seeking to dominate a "subject's" mind, are needed before the impression develops into sensory form. In cases, therefore, where A has himself had a hallucination of which he has spoken at the moment, and B has shared it, it is too much to assume at once that B's experience must have been exclusively due to the verbal suggestion; for if A's mere suggestion can produce such an effect on B at that particular moment, why not at other moments when he suggests the imaginary object without having himself seen it? None the less, of course, ought the hypothesis of verbal suggestion to be most carefully considered, in relation to the special circumstances of each case, before any other hypothesis is even provisionally admitted.

1A probable example is the recent remarkable delusion at Corano - starting from a peasant girl's alleged vision of the Virgin - in which a crowd of children and many adults shared. It is described in the Times for July 31, 1885.

2If (as intelligent English eye-witnesses believe) a semi-hypnotic condition, due to abnormally concentrated attention, is in great part answerable for the extraordinary illusions of Indian jugglery, the same condition might naturally be looked for in eases of collective hallucination. Very suggestive in this respect is the following record, by Professor Sidgwick, of a scene described to him by Mazzini: -

I have, perhaps, said enough to define the phenomena which are really of interest for us here. Fairly to allow of explanation by thought-transference, a collective case must present evident marks (I) of being a hallucination and not a mere illusion; (2) of having occurred, so to speak, in an isolated way, and not under the dominance of any special prepossession; and (3) of having been independently projected by the several percipients, and not merely conjured up by one on the suggestion of another. It is naturally not always easy to ascertain how far these conditions are met. In judging of the auditory cases, especially, great caution is necessary; for,as we have seen above (pp. 443-4), there is scarcely any sort of mere noise which may not have some undiscoverable external origin in the house or the neighbourhood. Intelligent speech, on the other hand, and certain musical sounds, such as bell-sounds or distinct melodic sequences, if externally caused, imply conditions the presence or absence of which it is usually possible to ascertain.

So again in the visual cases, the fact of dim or uncertain light may favour the hypothesis of illusion; but where the light is good, the presence or absence of an adequate external cause in the vicinity can often be determined with all but complete certainty.1 One what he was looking at. "The cross," he said, "there." I took hold of his arm, gave him a slight shake, and said, "There is not any cross at all." A sort of change came over his countenance, as though he was waking up from a kind of dream; and he responded, "No, as you say, there is no cross at all." So we two walked away, and left the crowd to their cross,' It is nearly 20 years since I heard this story; but it made a considerable impression on me, both from the maimer in which Mazzini told it, and from its importance in relation to the evidence for 'spiritualistic' phenomena".

1 I am including only cases of hallucinations which have occurred to more than one percipient simultaneously) or very nearly so. The extremely perplexing cases, few, but well attested, where the same phantasm has been independently described by different persons who have at different times encountered it in the same locality, may possibly be also connected with the infectious character of hallucinations; for we cannot pronounce it to be indispensable that the infectious influence should act at the moment. A certain amount of evidence for this explanation is afforded by cases where the experience (not apparently due to suggestion or illusion) has somepoint of uncertainty often remains, owing to the way in which the evidence reaches us: we cannot be sure how far the mere verbal description of one percipient, after the occurrence, may not have caused another to fill in or modify his own recollection with details which he did not himself observe. But if both clearly shared in the experience, it is not important that their percepts may not have been so precisely similar as is sometimes alleged.

So far, indeed, from telling against the theory of mental transfer, such want of identity is rather what we might have expected, both from the numerous approximate successes in experimental thought-transference - e.g., in reproducing drawings - and from the evidence that a telepathic impression is liable to be reacted on in various ways by the person whom it affects.