Yoakum, Behavior of Squirrels. 567 



to fit snugly over the front lower edge of the entrance. These were 

 placed over both boxes and thus shut out all temperature discrimina- 

 tion by contact, thus forcing the animal to depend upon the air in 

 the entrance to the tunnel. The effect is shown in the records of 

 the first set of tests, series 15, 16, 17 and 18, Table V. 



The squirrel was plainly confused, which he showed by snifiing 

 the air and testing each box once or even twice before entering. He 

 soon learned, however, to de]3end upon the air at the entrance of the 

 box in deciding which box to enter. It seems quite possible that 

 under the first set of conditions, the squirrel was discriminating 

 partly by actual contact with the metal of the box and partly by 

 means of the air at the entrance. 



(&) Discussion of results of tests upon white rats. — Table VI 

 shows the work of three white rats on temperature discrimination. 

 The method used was slightly different from that pursued with the 

 squirrels. The rats seemed so uncertain and irregular in their early 

 tests, that it was practically necessary to develop a position habit. 

 Such a habit was accordingly developed and then broken. The rat 

 learned to find the food in one box by position, then the position 

 was changed and he was compelled to learn a new position. To the 

 observer at least the earlier stages of the learning process seemed 

 to be much more easily detected by this method than by the method 

 used in the experiments upon the squirrels. 



IV. Conclusion. 



1. In the solution of the problems placed before the squirrel in 

 the series of experiments, it is shown that the squirrel learns by the 

 trial and error method. His learning curves are in the main similar 

 to the curves representing the learning processes in the white rat. 



2. The greater irregularities in the curves obtained from the squir- 

 rel may, perhaps, be explained on the basis of emotional factors that 

 are more prominent in these animals than in the rat. The greater 

 sensivity to emotional disturbances seems to be due to the fact that 

 the distance receptors play a larger role in the life of the squirrel 

 than in the rat. Experimental proof of this seems possible. 



3. Successful training in certain problems is highly prejudicial 



