214 



THE AMERICAN NATURALIST [Vol. XLII 



is of course the essence of "reflective" imitation. (4) Thorn- 

 dike concluded that cats have probably no "free ideas"; no 

 stock of images which are motives for acts. The association in 

 the cats was always between a motor impulse and a present 

 sense perception ; there was no association of ideas. This nega- 

 tive conclusion was based largely on the inability of the animals 

 to learn from being put through an act. In this latter matter, 

 Cole calls the reader's attention particularly to "the radical 

 difference at every point" between the cats of Thornidike's 

 experiments and the raccoons of his own. "If inability thus 

 to learn is evidence against the presence of ideas, then ability 

 to do so should be equally strong evidence for it." Further- 

 more, Cole gives much additional evidence for the presence of 

 ideas in the raccoons; and certain results of some extremely 

 ingenious experiments amount to a demonstration that the ani- 

 mals do hold mental images, so far as such a thing can be 

 demonstrated. 7 The animals seemed to remember definite ob- 

 jects for a time, then forget them; then suddenly, under certain 

 conditions, recall them. They fought against being put into 

 boxes with complex fastenings, from which they had some time 

 before had difficulty in escaping, though they willingly went into 

 similar boxes whose fastenings they had found simple. In cer- 

 tain experiments there were two alternative signs to be raised; 

 the green one meant food, the red one meant none. The raccoons 

 learned to raise these signs by clawing at the standards, but they 

 could not see beforehand which sign would come up by claw- 

 ing at a certain standard. "When the red one came up they 

 clawed it down again, then clawed up the green one, and made 

 ready to receive food. Clearly, the red sign did not correspond 

 to an image that the animal had in mind, while the green one did. 

 Other experiments were devised in which success depended on 

 the animal's holding in mind the images of certain things that 

 had gone before ; the raccoons stood these tests successfully. It 

 is difficult to see how there could be more conclusive proof of 

 the presence of ideas in animals that can not talk. 



T A subjective thing, such as an idea, can, of course, not be absolutely- 

 demonstrated by objective methods. It is always possible to substi- 



existence of ideas" in animals can never go further than to show that 

 they act as men do when men have ideas. 



