CANADA PORCUPINE 49 



and methods of choosing as will be seen in the four-wise tests 

 later. 



The six regular forms were then presented six-wise with the 

 circle still marking the food box. The forms were shifted ac- 

 cording to a regular schedule which placed the circle in every 

 position an equal number of times as nearly as possible, and 

 successively in a new environment after each reaction. No. 6 

 had considerable difficulty. His habit of finding the food in 

 the second form approached, came out very strongly. Many 

 times he would reject the first one and plunge into the second 

 without stopping to examine it. In this larger group there 

 was not merely a positive search for the circle but the method 

 employed involved a definite rejection of each of the forms 

 encountered before the circle was reached. The animal would 

 stop before each form separately and "choose" sometimes with 

 considerable hesitation either to enter or not to enter. When 

 the circle was at one end of the apparatus animals have been 

 observed to make as many as 19 "negatives" in succession and 

 finally end by entering the circle first and being credited with 

 a right selection. The animal's behavior in such cases could 

 hardly be misinterpreted. He usually rested his forefeet on the 

 apparatus below the forms, shifted side-wise along the front, 

 frequently looking from side to side as if in search of the circle 

 but never showing indications that he recognized it when he 

 was more than 12 inches from it. 



Immediately following these tests the circle, the ellipse, the 

 circle-triangle and the hexagon-circle were presented four-wise 

 (see tables VII and VIII) with the same apparatus used above 

 only removing the end forms and food-boxes. The first arrange- 

 ment was such that the base lines were all similar while the 

 top halves were parts of the circle, ellipse, triangle, and hex- 

 agon. It was not strange that No. 7 with his propensity for 

 using base lines should require three and one-half times as much 

 practice to learn to discriminate the circle from the group. Three 

 of the four errors made by No. 6 were the first three tests, two 

 of which were errors on the circle -triangle. Twenty-five of the 

 37 errors made by No. 7 were made on the circle -triangle which 

 it will be remembered both animals had previously failed to 

 discriminate when paired with the circle in this position but 

 which both mastered now under more difficult circumstances. 



