[2] Vertot, an historian of the eighteenth century, failing to receive, when he was ready for them, the documents upon which he counted in order to write his _Siege of Rhodes_, finished his work for all that; and when the documents were handed to him, he contented himself with saying: ‘I am very sorry, but I have finished my siege.’ He preferred leaving his work imperfect to beginning it over again. I am sure the individuals observed by M. Janet have been very carefully studied by him; but I regret that my learned colleague has not encountered a genuine medium. I share his opinion concerning most spiritistic mediums; I have only found two interesting ones among them; the hundred others which I have observed have only given me automatic phenomena, more or less conscious; nearly all were the puppets of their imagination. It is outside spiritistic circles that I have discovered the best mediums. M. Janet’s criticisms are only erroneous because they are too sweeping. His conception of psychological disintegration is applicable to the greater number of cases; but it does not apply to all. It is a very different thing to study a crystal-vision, or an automatic writing revealing nothing beyond the tenor of the sensitive’s memory, or to observe a premonitory vision such as has been given me to do. The indication of a future event cannot be explained by Janet’s hypothesis. It reveals especial faculties that I can scarcely consider pathological, unless I consider them as such in the same way as one considers genius to be a sign of degeneration. It is more reasonable to think that our nervous sensibility will become more and more refined. It is rash to believe that the present human type is the definite end of evolution. Our species is only one link in the series of beings; the causes, which have led up to the improvement of the human race, are still in activity, and it is logical to think there are some natures above as well as below the average. The latter represent ancestral types—a return to cast-off forms; the former are perhaps precursors, possessing faculties which are abnormal to-day, but which may become normal to-morrow. I must pause, for I see I am forsaking the domain of facts for that of hypotheses; I hasten to return thither. I have pointed out the signs which permit us to suppose that a certain given person is a medium; although these signs are not certain, they seem to me probable. In reality, there is only one sure way of testing the faculties of a medium: that is to experiment with him.