In my opinion the answer is that the authoress was not only not aware of, but _could not_ in the least have been aware of, the action that may have passed from herself to the dog so as to bring about the rapping of the answers; and that on the other hand it is not a question at all of thinking of a simple mechanical operation of the kind mentioned above, because in the presumed action of the authoress on the dog there is no need to have recourse to such a crude hypothesis (as surely there was no similar action of Krall's on his horses, especially when they were separated from him). I maintain, in fact, that in principle, even without any contact by hand, we may still presume that all the "wonders" obtained by Miss Kindermann are obtainable, taking, of course, into account the peculiar endowments of the animal we are dealing with. For if there be any automatism (and there is surely a good dose of it), it is certainly not a question of a mechanical automatism (of the type of Neumann's), but quite certainly of a true and proper _psychic automatism_; a very different thing, and without doubt much more complex. In all probability the first condition for the occurrence of genuine phenomena similar to those attributed to "thinking" animals must be a very particular psychic relationship between the animal and his master. And such a relation, although with reluctance, I am compelled to call of the mediumistic type. My reluctance is due in part to the very unhappy etymology of the term, derived from the famous word "medium," so unscientific both in its origin and in the meaning which some even now wish to associate with it. But even after having freed it from any "spiritistic" meaning, the term still leaves me reluctant; for I cannot hide from myself the weakness of a hypothesis which, in order to explain (only in part) one enigmatical fact (in this case, that of "thinking animals"), must have recourse to another unsolved enigma (in this case that of the "mediumistic phenomena"). However, it will already be something if the two problems are eventually merged together and so become a single problem; but it is not my object to explain any psychical facts themselves, whatever they may be, under which the phenomena of Lola and others of a similar nature may be eventually classified. It will be sufficient for me at present to group the performances of the animals, if possible, with something better known. And "mediumistic" facts, extrinsically at least, are certainly better known. I refer therefore to them as I find them described in the psychology called supernormal; because, from force of circumstances I am compelled to recognize that it is within this psychology that I must now continue the discussion. IV. MEDIUMISTIC "RAPPORT" AND TELEPATHY