34 Chapters on Animals. 



most likely to be called for, such as chien, dog, horse, cat, 

 pferd, canis, etc., etc. The restriction to one set of letters 

 simpUfied the business considerably. But M. du Rouil 

 confessed quite frankly that she could not get through a 

 word unless he were present. On the other hand he could 

 not make her spell a word in public that she had not before 

 practised with him in private. So it was with Lyda and 

 the iigures. She really knew the figures when isolated, and 

 this had been satisfactorily demonstrated when he left the 

 room, and she gave me the number asked for, up to 9. But 

 he would not tell me the secret of the confederacy. I told 

 him what guegses had been made on the subject, but he 

 simply answered that I must have observed how impossible 

 it was for him to make signs with hands or feet when he 

 moved neither hand nor foot. 



Would he give me some account of the earlier stages 

 of training through which these dogs had passed .' Yes, 

 very willingly. The first thing was to teach a dog to fetch 

 an object, the next to make him discriminate between one 

 of two very different objects placed together, and bring 

 one or the other as it was mentioned by its name. In 

 beginning the alphabet he put two most dissimilar letters 

 side by side to begin with, such as an O and an I, avoid- 

 ing the confusion of similar ones, such as O and Q, or B 

 and R. Gradually, the dog became observant enough to 

 discriminate between letters in which the difference was 

 not so marked. M. du Rouil told me that he had found 

 the greatest difficulty in teaching Blanche to distinguish 

 between the knaves and kings in playing-cards, but that 

 she learned the aces very promptly. With regard to the 

 time required for educating a dog sufficiently to perform 

 in public, he said that an hour a day for eighteen months 

 was the time required, and he preferred a single hour to a 



