An analysis of the impressions showed that most of them were reproductions of objects familiar to the percipient, in certain cases of hallucinations previously imposed upon them in the course of these or other experiments. With some of the successful percipients these spontaneous impressions showed a marked tendency to recur. Thus P. had a wrong impression--of an elephant--no less than four times in the course of the experiments; and T. of a woman and a perambulator three times. One of these coincided with the subject actually set, and the coincidence may perhaps therefore be attributed to chance. Speaking generally, however, this tendency to repetition amongst the percipient's native impressions constitutes an additional argument, if any such is needed, for attributing the frequent coincidences of the impression with the subject set to some other cause than the automatic association of ideas. An instance of a quasi-experimental character, which closely resembles the cases above described, is recorded by Dr. A. Gibotteau:--[41] No. 19.--By DR. GIBOTTEAU. "Madame P. complained of headache. I placed my hand upon her forehead, and in a few minutes she was in a light hypnotic sleep. Without deepening the trance I endeavoured to give her a sensation of calm and well-being, and to procure this sensation for myself in the first place, I called up a picture of the sea, in which air and water were full of sunlight. 'I feel a little better,' she said; 'how fresh the air is!' I then proceeded to imagine myself walking along the _Boulevard Saint Michel_, in a slight rain. I saw the hurrying people and the umbrellas. 'How strange it is!' said Madame P.; 'I seem to be at the corner of the _Boulevard Saint Michel_ and the _Rue des Écoles_, in front of the _Café Vachette_' (the exact spot I pictured); 'it is raining, there are a great many people, a hurrying crowd. They are all going up the street, and I with them. The air is very fresh. It gives me a pleasant, restful feeling.' With these words she opened her eyes and gave me further confirmation of her impressions. "I should add that this scene took place in the provinces; I had not been in Paris for some months, nor Madame P. for several years. "There had been no mention of the subject in the course of our conversation that day." It will be seen that Dr. Gibotteau attempted to transfer to the percipient only the general sensation of calm and rest induced in himself by the imagined scene, and that the success obtained was therefore of a kind by no means anticipated.