CANADA PORCUPINE 29 



depressed. On the other hand, when the lever was down a 

 spring would catch it so that it could not be raised; when the 

 plug had been extracted releasing the next part of the appa- 

 ratus it could not be reset; and when the button had been 

 turned upward it could not be restored to a horizontal position. 

 These last checks were not all used at any one time but were 

 ready for use whenever the porcupine manifested any disposi- 

 tion to reset the devices once properly operated. 



Porcupine No. 4 was given this problem first, but as the 

 apparatus was new, the animal little understood and the ex- 

 perimenters inexperienced in such work, no attempt was made 

 to obtain quantitative results. The chief value was that it 

 trained the experimenters for taking more accurate records and 

 observations in succeeding experiments. The procedure fol- 

 lowed was to feed the animal at the entrance of the box occa- 

 sionally closing the door until the animal began to look for the 

 food, then opening it and attempting to build up an association 

 between the food and the inside of the box even \^-hen the door 

 was closed. After one day devoted to this preliminary drill the 

 door was closed and fastened by the lever at the beginning of 

 the feeding period. No. 4 tried to get at the exposed food 

 directly, but soon went around to the left end and on the top 

 of the box as if looking for another opening. On this excursion 

 it was probably his constitutional sluggishness that prompted 

 him to climb over the lever from the back rather than to walk 

 around it. In doing so he depressed it, the door opened and 

 he secured food 53 seconds after beginning. On the next trial 

 he again scratched on the door for a time, then turned to the 

 left around the box clawing and sniffing as before. The third 

 trip was very similar but the fourth showed a distinct variation. 

 This time he started toward the right, passed around the end 

 of the lever and climbed over it from the back side, opening 

 the box in 5 seconds. This procedure he continued until the 

 loth trial when, on his trip around the lever, he stopped, ex- 

 amined it from the front, probably attracted by some odor, 

 then used both hands in pushing it down while he was yet in 

 front of it. Probably, even in this case, he was preparing to 

 gnaw the lever but on hearing the door open his attention was 

 diverted. This was his ultimate solution. It was an excellent 

 opportunity to explain by saying that the animal "knows where 



