[103] As opposed to "dream-illusions," which depend on various organic sensations, or on the stimulation of the external organs of sense. The distinction is made by Professor Sully, loc. cit., p. 139.
[104] These figures do not include second-hand cases. There are besides 29 undated cases, most of which probably belong to the remote period. See column 1 of table on p. 218 (visual cases).
[105] The calculation is based upon an analysis of the whole number of visual cases reported during the most recent month, which would indicate an annual rate of about 140. The figures for the most recent quarter indicate an annual rate of about 120.
[106] A hallucination which coincides with a death is defined, for the purposes of this inquiry, as a hallucination which occurs within twelve hours of the death.
[107] There is another possible explanation—viz., that some of the recent death-coincidences have been withheld from us, on account of the painful associations connected with them. That some cases—and recent would be more affected than remote examples—have been withheld on this account seems certain; but the explanation given in the text must, it is thought, be held primarily responsible for the discrepancy in the figures.
[108] Cases, that is, in which the collector is known or suspected to have asked the question of the narrator, because he knew that he was to receive an affirmative answer.
[109] The gross total of visual phantasms recorded at first-hand as representing a living human being, or part of a human being (e.g., a hand or a face), is 381. This total includes cases given in columns 1, 4, and 5 of the table on p. 218. From this total we have deducted 31 cases where the percipient has had other experiences but has not enumerated them, and 28 cases which are estimated to have occurred before the age of ten, leaving the total given in the text, 322.
Of the gross total of 381, 80 are alleged to have coincided with the death of the person represented. Deducting in like manner 7 cases where the percipient has had other unspecified experiences, and 8 where the experience is believed to have occurred before the age of ten, we reach the total of 65 given above.
As, however, more care was no doubt taken to procure first-hand evidence in the case of apparitions coinciding with a death than in other cases, it would perhaps lead to more accurate results if in the larger total were included the second-hand non-coincidental cases, 38 in number. The reader can, if he prefer, work out the result for himself on this basis. But it will, of course, be understood that it is not practicable to sum up in a few pages the results of a long investigation; and those readers who are interested in the nature and distribution of casual hallucinations, and their relations to telepathic apparitions, are referred to the forthcoming Report, from which the figures in the text are quoted.
[110] These details are taken from notes made by the writer immediately after the interview.