EXPERIMENTS AND OBSERVATIONS 69 



number i. In the second half of the tests, all these 

 differences disappeared and errors were infrequent and 

 seldom exceeded + i or — i . These results of practice are 

 not to be accredited to the horse, but to the experimenter, 

 who was at first quite unskilled. This difference in results 

 does not appear in the case of Mr. von Osten, for his. 

 initial practice had been had many years previous. The 

 values obtained in his case were very constant throughout 

 our experimentation and generally showed something like 

 90% of correct responses. To be sure, in his case also, 

 the number i was somewhat unfavorable, (79% were 

 correct responses). But the percentages obtained in his 

 case showed no improvement whatever throughout our 

 experimentation. We need scarcely add that with the 

 voluntary control of the giving of the signs, in the case at 

 least of such small numbers as are here discussed, no 

 errors, whatever, occurred. 



We have discussed the influence of the experimenter, 

 i. e., the one who asked the horse to tap ; now let us con- 

 sider the influence of others present upon the horse. 



As a general rule, other persons had no effect upon the 

 horse's responses. This appears from the failure of nearly 

 all tests in which all of those present — with the exception 

 of the questioner himself — knew the number which the 

 horse was to tap. Even when the others concentrated 

 their whole attention upon the number, it profited little as 

 a close analysis of the 136 cases, which belong under this 

 head in our records, go to prove. Thus, in the presence 

 of a group of twenty interested persons — during the 

 absence of Mr. von Osten — ^twenty-one problems were 

 given to the horse, the solutions of which were known to 

 everyone but myself, the questioner. Result: only two 

 correct responses. Only when there was among the 



