232 THE HORSE OF MR. VON OSTEN 



of Justinian, in the middle of the sixth century.'"' AH 

 of these animals were kept for purpose of speculation and 

 were exhibited for pecuniary reasons only. Nor does one 

 read that any person could work with these animals off- 

 hand, which was the characteristic feature of the Osten 

 horse.* In many cases we find mention made of the 



* There is only one, and I believe it is only a seeming exception 

 to be found in the literature on the subject. We are told that 

 about the year 1840 a French revenue official named Leonard had two 

 hunting dogs that, besides other things, were able to play at dominoes, 

 and this not only with their master, but with anyone and without the 

 master's assistance. The owner had educated them simply for the fun 

 of it, and not for pecuniary gain. This statement is made by both 

 writers who, apparently independently of one another, have discussed 

 the case, Youatt '"^ and de Tarade."" De Tarade himself played 

 with them, and gives directions how to teach dogs to play the game. 

 But his exposition is so naive, and even ridiculous, for those who know 

 anything about the subject, that we do not believe it necessary to at- 

 tempt a detailed refutation. Youatt never saw the animals. But he 

 tells us that not only the dog's partner, but also the master, sat at the 

 game. Youatt's assertion, however, that "not the slightest intimation 

 could have been given by Mr. Leonard to the dog," but that the animal 

 carried on the game by means of its own observation and calculation, 

 appears to me a rather bold statement. After my own experience with 

 dogs, I firmly believe this to have been impossible. Hachet-Souplet,"" 

 who shares my conviction, explains the matter as follows : the dog 

 would simply place a domino having the number of eyes named by his 

 partner, thus the 6 adjacent to the 6, the 3 to the 3, etc. But even so 

 a great deal would have to be attributed to the dog, (although in that 

 case real counting would by no means be absolutely necessary, for an 

 association between the number term and the total picture of the cor- 

 responding group of eyes would suffice.) But we must note that neither 

 of the writers mentions that the numbers were always called aloud by 

 the partner. After the failure of the experiments of Sir John Lub- 

 bock,'" we must doubt very much if a dog is able to match one domino 

 with another having the same number of eyes. We are therefore in- 

 clined to believe that this dog continually received signs from its master. 

 These signs probably were visual, perhaps also auditory, and they were 

 by no means involuntary. For in a book on the training of animals, 



