INTELLIGENCE IN ANIMALS. 183 



In the following case, the object of an animal's 

 action in such cases was tested by an experiment, but 

 the evidence is less satisfactory in one respect than 

 that afforded by the two previous cases, the animal 

 having been taught the action: "A small English 

 terrier belonging to a friend/' says the narrator of 

 the story, " has been taught to ring for the servant. 

 To try if the dog knew why it rang the bell, he was 

 told to do so while the girl was in the room. The 

 little fellow looked up in the most intelligent manner 

 at the person giving the order (his master or mistress, 

 I forget which), then at the servant, and refused to 

 obey, although the order was repeated more than 

 once. The servant left the room, and a few minutes 

 afterwards the dog rang the bell immediately on being 

 told to do so." Here it is to be noticed that the dog 

 did not ring the bell (as, in each of the preceding 

 stories, the cat knocked at the door) to get some end 

 of his own accomplished. He rang to save his master 

 or mistress trouble. And the fact that he had been 

 taught to ring for this purpose, although making the 

 act itself less obviously a sign of reasoning power 

 than the cat's action in knocking at the door, makes 

 his refusal to ring when told to do so a more manifest 

 evidence of reasoning than it would otherwise have 

 been. If the dog rang for the servant because of 

 some advantage he always gained from the servant's 

 coming, it would have been natural enough that he 

 should refrain from ringing when the servant was in 

 the room. But his refusing to do what he had been 



