(p. 301).  What, not even if all hallucinations, or ninety-nine per cent., coincided with the death of the person seen?  In heaven’s name, why not?  Why, because the ‘weightiest’ cause of all has been omitted from our calculations, namely, our good old friend, the association of ideas (p. 302).  Our side cannot prove the absence (italics) of the association of ideas.  Certainly we cannot; but ideas in endless millions are being associated all day long.  A hundred thousand different, unnoticed associations may bring Jones to my mind, or Brown.  But I don’t therefore see Brown, or Jones, who is not there.  Still less do I see Dr. Parish, or Nebuchadnezzar, or a monkey, or a salmon, or a golf ball, or Arthur’s Seat (all of which may be brought to my mind by association of ideas), when they are not present.

Suppose, then, that once in my life I see the absent Jones, who dies in that hour (or within twelve hours).  I am puzzled.  Why did Association choose that day, of all days in my life, for her solitary freak?  And, if this choice of freaks by Association occurs among other people, say two hundred times more often than chance allows, the freak begins to suggest that it may have a cause.

Not even the circumstance cited by Herr Parish, that a drowsy tailor, ‘sewing on in a dream,’ poor fellow, saw a client in his shop while the client was dying, solves the problem.  The tailor is not said even once to have seen a customer who was not dying; yet he writes, ’I was accustomed to work all night frequently.’  The tailor thinks he was asleep, because he had been making irregular stitches, and perhaps he was.  But, out of all his vigils and all his customers, association only formed one hallucination, and that was of a dying client whom he supposed to be perfectly well.  Why on earth is association so fond of dying people—­ granting the statistics, which are ‘another story’?  The explanation explains nothing.  Herr Parish only moves the difficulty back a step, and, as we cannot live without association of ideas, they are taken for granted by our side.  Association of ideas does not cause hallucinations, as Mrs. Sidgwick remarks, though it may determine their contents.

The difficult theme of coincidental collective hallucinations, as when two or more people at once have, or profess to have, the same false perception of a person who is really absent and dying, is next disposed of by Herr Parish.  The same points de repere, the same sound, or flicker of light, or arrangement of shadow, may beget the same or a similar false perception in two or more people at once.  Thus two girls, in different rooms, are looking out on different parts of the hall in their house.  ’Both heard, at the same time, an [objective?] noise’ (p. 313).  Then, says Herr Parish, ’the one sister saw her father cross the hall after entering; the other saw the dog (the usual companion of his walks) run past her door.’  Father and dog