Such unequivocal results, however, were only obtained after we had provided blinders of sufficient size (15 × 15 centimeters). Mr. von Osten believing that the horse would not suffer these to be applied, had at first proposed other measures. He held a slate before his face. Some of the horse's responses were right, others wrong. The tests were repeated and were successful as long as I, myself, held the slate before my face, but not a single one of the responses was correct when another would attempt to hold the slate before me. Mr. von Osten then brought forth a kind of bolster which he fastened on the right side of the horse's face,—the side which was turned toward the questioner. But this also gave uncertain results. Finally he agreed to apply blinders. But these were much too small and projected at a great angle from the head (Mr. von Osten had cut the straps, for he thought they worried the animal). The result was that only the posterior part of the horse's normal field of vision was obstructed. Therefore, one could never be quite sure whether Hans, who—it will be borne in mind—made every attempt to see the questioner, had not perhaps after all been able to peer over the edge of the blinder. The number of "undecided" tests, therefore, became very great. Of 108 tests, only 25 could be placed in the category of "not seen", 44 in the "seen", and 39, i. e., a third of the total, in the "undecided." The percentage of correct answers for these three categories were, respectively: 24%, 82% and 72%. Here we have once more approximately the same ratio between the categories of "seen" and "not seen" as in the case of the tests with the smaller blinders. If we were to count the cases which we had put under the head of "undecided," in the same category as those in which vision had been excluded—as Mr. von Osten had done—then one would have been led to the conclusion that the horse did not need visual signs. Several observers had thus been led astray: e. g., General Zobel writes in the "National-Zeitung" (Aug. 28, 1904), that upon request Mr. von Osten had covered Hans's right eye "by means of some sort of blinder, so that he was unable to see his instructor", and that Hans did not fail to respond correctly. We evidently have here to do with the unreliable bolster mentioned above. Furthermore, Mr. Schillings made a number of tests with the small blinders, in which 50% of the answers were correct, and probably in the same manner were obtained the results published in one of the daily papers (the "Berliner Tageblatt", Dec. 12, 1904), several days after the publication of the December report, and reading as follows: "Tests have been made upon Hans with blinders over his eyes and it is to be noted that, in spite of these, he still responds correctly." Mention is also made of the experiments noted in [Supplement III (page 257)], in which Mr. von Osten hid behind the questioner and merely encouraged the animal by occasional exhortations, but it is not possible to say with any degree of certainty in how far he was really hidden from the horse's view.

I would add that the horse—in so far as it was at all possible to decide—never looked at the persons or the objects which he was to count, or at the words which he was to read, yet he nevertheless gave the proper responses. But he would always make the most strenuous efforts to see the questioner. (See [page 43]). I would furthermore add that several experiments, in which Mr. von Osten and the horse were separated from each other by means of the canvas tent, failed completely, and that, on the other hand, all tests were successful in which the questioner was present in the feed-room and the door between this and the horse's stall was opened wide enough for him to be seen by the horse. I would also mention that toward evening the responses became less and less accurate. The conclusion that visual stimuli were here operative cannot be gainsaid.

It was possible, to be sure, that other senses might also be involved, but it was certain that auditory sensations did not enter it. This is shown by the fact that one might remain just as silent while the horse was tapping his answer as during the putting of the question and yet obtain a correct response. Hans, furthermore, could scarcely be distracted by auditory stimulations. If either the experimenter or anyone else present sought, at a given moment, to interrupt him by such calls as "Halt", "Wrong", etc., while he was going through the process of tapping, they very seldom succeeded in their attempt. Even though such interruption did succeed in seven out of the twenty-one cases in which it was tried, the assumption is well grounded that the success was due entirely or almost entirely to minimal movements involuntarily executed by those attempting the interruption. It is to such minimal movements that the horse, as we shall see later, promptly reacted. When the experimenter (Pfungst), himself, made the interjections, which certainly should have been more effective, we found that the horse was actually disturbed in only two of the fourteen cases; and finally in ten consecutive cases of attempted interruption not a single one was successful. There was almost a complete absence of any ear movements on the part of the horse, a fact in which I have been borne out by Mr. Henry Suermondt, the distinguished horseback rider. Indeed, I cannot recall that Hans ever turned his ears toward me, a fact which is strikingly curious in the case of a horse so attentive and so spirited in temper.

Finally, I might also mention that the breathing of the experimenter in no wise influenced the outcome of the experiment. Whether he held his breath or breathed on the leg or body of the horse, made no difference.

Investigations of the other senses became needless, for I had, in the meantime, succeeded in discovering the essential and effective signs in the course of my observations of Mr. von Osten. These signs are minimal movements of the head on the part of the experimenter. As soon as the experimenter had given a problem to the horse, he, involuntarily, bent his head and trunk slightly forward and the horse would then put the right foot forward and begin to tap, without, however, returning it each time to its original position. As soon as the desired number of taps was given, the questioner would make a slight upward jerk of the head. Thereupon the horse would immediately swing his foot in a wide circle, bringing it back to its original position. (This movement, which in the following exposition we shall designate as "the back step", was never included in the count.) Now after Hans had ceased tapping, the questioner would raise his head and trunk to their normal position. This second, far coarser movement was not the signal for the back-step, but always followed it. But whenever this second movement was omitted, Hans, who had already brought back his foot to the original position and had thereby put it out of commission, as it were, would give one more tap with his left foot.

If it was true that these movements of the questioner guided the horse in his tapping, then the following must be shown: First, that the same movements were observed in Mr. von Osten in every case of successful response; secondly, that they recurred in the same order or with only slight individual changes in the case of all who were able to obtain successful responses from the horse, and that they were absent or occurred at the wrong time in all cases of unsuccessful response. Furthermore, it was observed that it was possible to bring about unsuccessful reactions on the part of the horse as soon as the movements were voluntarily suppressed, and conversely, that by voluntarily giving the necessary signs the horse might be made to respond at pleasure; so that anyone who possessed the knowledge of the proper signs could thereby gain control over the process of response on the part of the horse. These requirements have all been fulfilled, as we shall see in the following pages.

With regard to the regular recurrence of the movements noticed in the case of Mr. von Osten, I was, after some practice, able to note carefully their peculiar characteristics. This was rather difficult, not only on account of their extreme minuteness, but also because that very vivacious gentleman made sundry accompanying movements and was constantly moving back and forth. To abstract from these the essential and really effective movements was truly difficult. It was much easier to observe these movements in the case of Mr. Schillings, probably on account of the fewer accompanying movements and perhaps on account of their greater distinctness. Usually he would raise the entire trunk a trifle, so that the movements could be noticed from behind. Besides these, I had an opportunity to observe the Count zu Castell, Mr. Hahn and the Count Matuschka. All three made the same movements, though somewhat more minutely than Mr. Schillings, yet none was as slight as those of Mr. von Osten.[I] I further noticed that Count Matuschka and Mr. Schillings often showed a tendency to accompany every tap of the horse with a slight nod of the head, the last being accompanied by a more pronounced nod and then followed by the upward jerk of the head, in other words, they beat time with the horse. In the case of the last three mentioned, for whom the horse responded far less effectively than for Mr. von Osten or Mr. Schillings, belated or precipitate jerks would frequently occur. This was found to be true in the case of all other persons who had failed to elicit adequate responses from the horse. Often, in both cases, a complete absence of any kind of minimal movement had been noted. The accuracy of these observations in the case of Mr. von Osten is attested by Mr. Stumpf and Mr. von Hornbostel, and by these same gentlemen and Prof. F. Schumann in the case of Mr. Schillings and myself. They also found these movements to be most minute in the case of Mr. von Osten. In my case also they pronounced them "minimal, and often quite imperceptible". All other persons who have seen me work with the horse, but who were not familiar with the nature of these movements, never perceived them, no matter how closely they observed me.

Since the doubt was expressed that these movements did not precede but followed closely upon the back-step of the horse (i. e., that an error with regard to the time-element was involved), it became important that time measurements be taken. This was done in the following manner: The questioner asked the horse to tap numbers from 5 to 20, seldom higher. He purposely refrained from pronouncing the number, but recorded it after each test had been completed. This was a matter of indifference to the horse (see [page 42]), and had the advantage that the measurement was not influenced by knowledge on the part of the time-keeper. Two observers were required, one watching the horse, the other the questioner. Both observers had fifth-second stop-watches. The larger face of this watch shows the fifth-second and a hand on the smaller face indicates the minute. By pressing upon the stem the watch may be set in motion at any moment desired, and by pressing it once more it may be instantly stopped, and the time

elapsing between the setting in motion and the stopping may be read on the face. By pressing upon the stem a third time the hands are brought back to zero, and the watch is ready for another test. At a moment agreed upon beforehand—usually the third tap of the horse—both observers started their watches. Practice tests had shown that this could be done with all the accuracy necessary in this case. As soon as the observer of the questioner noticed the latter's head movement he stopped his watch, and as soon as the observer of the horse noticed the latter's back-step he stopped his watch. Since the movement of the horse's foot does not occur as a jerk, but is of greater extent than a jerk would be, it was agreed that the observer was to stop the watch as soon as he recognized the back-step as such, not when the foot was being raised from the ground, because it was not then evident whether the horse would bring it back to the original position or whether he was preparing to give another tap, nor when he had brought his foot completely back, but at the moment in which it was evident that the horse intended to make the back-step. Experimentation had shown that an agreement as to this moment was possible. A tap with the left foot, which might possibly follow upon the back-step, could be left out of account. The difference in time between the two watches would show the time between the head-jerk of the questioner and the back-step of the horse,[J] and if the back-step was indeed a reaction upon the head-jerk, then the watches would have to show a later time for the back-step than for the head-jerk.