As regards the possibility of unconscious indications of the number thought of being given by the agent, it seems certain that no such clue could have been perceived through the sense of sight or touch, contact between agent and percipient having been absolutely excluded throughout the experiments. It remains to consider whether any indication could have been given by means of sounds. In the presence of two or more attentive and vigilant witnesses any indications by sounds--_e.g._, an unconscious whispering of the number by Mr. Smith--could only have been perceived by persons of abnormal susceptibility. We know, indeed, of no precise limit which can be set to the hyperæsthesia of hypnotised subjects. But, on the other hand, hyperæsthesia of any sense in such subjects is generally the result of suggestion, direct or indirect, on the part of the operator; and in these experiments the only suggestion given--a suggestion apparently acted on throughout--was that they should _see_ the result. Since, indeed, hypnotised persons are apparently not necessarily aware of the channel by which information reaches them, this circumstance is not in itself conclusive; but taken with the fact that no direct suggestion to hear was given, it tends to make auditory hyperæsthesia less probable. It is perhaps more important to note that the experimenters, including Mr. Smith himself, were fully aware of this source of error, and on their guard against it; that no movements of Mr. Smith's lips, such as must have occurred if he had whispered the number, were observed; and that a careful analysis of the failures shows no tendency to mistake one number for another similar in sound--_e.g._, _four_ for _five_, _six_ for _seven_, or _five_ for _nine_. _Experiments with Agent and Percipient in different Rooms._ [Illustration]