Page:The Naturalisation of the Supernatural.pdf/368

348 that in all five cases the number dreamt of was a high, i.e., a favourable one, and the dream no doubt would win more credence because of its good augury. But it is not quite so easy to be satisfied that the dreams last noted—of which three are reported as occurring in the same village, Eecloo, in the course of less than ten years—were due wholly to chance. It would certainly appear that there is a case for further enquiry here.

We have a few cases of correct predictions made by professional mediums. But here again, in view of the large number of predictions made under similar circumstances which are not fulfilled and are forgotten, it would be unsafe at present to count too highly the few shots which hit the mark.

We have numerous cases reported to us of unusual sights or sounds—animals, corpse-lights, Banshees, the death-watch—preceding a death. But the evidence in these cases is in its present state quite insufficient to establish any connection. One obvious defect in the symbolic dream or omen is that there is no intrinsic relation between the event and its symbol. Our own ancestors saw a connection between comets and disasters; and