46 L. W. SACKETT 



table VI) the base of the triangle was down throughout but 

 he was unable in these 400 tests to perfectly re-establish the 

 association. A compound form was then made consisting of a 

 circle in one-half and of the apex of the triangle in the other 

 half. For convenience this irregular form is called the circle- 

 triangle. This, with its curve edge down, was paired with the 

 circle with the result that in 20 series there were 202 right 

 choices and 198 wrong ones (D, table VI). The porcupine had 

 simply formed the habit of inspecting the one at the left first. 

 This was correct half of the time ; and when it was not he would 

 move on to the next one and get the food. The circle -triangle 

 was then adjusted so that the apex was down with the result 

 that it took 12 series to break the place habits. After that 

 the circle was chosen fairly regularly. Only 1 1 errors were made 

 in the next 160 trials (E, table VI). But when the circle-triangle 

 was again inverted making the base lines of the two forms 

 similar, there was no ability to select the proper one. The 

 animal made only 47% correct choices after feeding in the 

 circle 2000 times previously (F, table VI). At this point, the 

 experimenter became convinced that porcupine No. 3 could not 

 solve the problem assigned to him and the experiment was 

 discontinued. Later experiments about to be described indi- 

 cate that the failure was due to the method employed, that 

 the experimenter was disheartened too soon and that No. 3 

 would eventually have mastered the problem. The difficulty 

 in method was to assume a certain proficiency with a given 

 amount of work and promote the animal on the basis of that. 

 This seems to work as disastrously with .the porcupine's learn- 

 ing as the same method does occasionally with children in the 

 public school. No. 3's failure to learn the forms was a peda- 

 gogical blunder of the experimenter rather than a case of sen- 

 sory deficiency, or mental incapacity on the part of the porcupine. 

 Tables VII and VIII representing the results of tests on 

 porcupines Nos. 6 and 7, respectively, require no elaborate 

 interpretation. These animals had recently been captured from 

 the forest, having had, preceding these form tests, only a lim- 

 ited number of tests of righthandedness and lefthandedness. 

 The pair-wise method was employed because this was simpler 

 than presenting all six forms at once. The progress of the two 

 animals with these two forms was almost parallel. Both also 



