ix LEARNING AMONG HIGHER ANIMALS 179 



observations which some writers might have taken as a 

 proof of high intelligence, as proving precisely the 

 contrary. 



3. a. Habitation and number of instances. 



Explicit memory, as we have seen, serves as a basis for 

 inference from a single experience. If we could convert 

 the proposition and be assured that inferences from a 

 single experience imply explicit memory, then we could 

 without further difficulty assign memory and the type of 

 inference based on it to many animals. The elephant or 

 the dog that long cherish vindictiveness for a single 

 injury, the horse which knows better than its rider the road 

 they have once traversed, and that perhaps in a contrary 

 direction, could be said without more ado to draw inferences 

 of this type. We must return to this point later. It 

 is enough for the present to remark that while habits are, 

 as every one knows, generally formed gradually by many 

 repetitions, explicit memory is due to a single experience. 



Mr. Thorndike's first argument is based upon this 

 contrast. 



" The gradual slope of the time-curve, then, shows the absence 

 of reasoning. They represent the wearing smooth of a path in 

 the brain, not the decisions of a rational consciousness." x 



But is the slope gradual ? Turning to the actual curves 

 (p. 1 8, ff.) there seem to be many cases in which it is not 

 so by any means. Thus in box D, which was opened by 

 pulling a string at the top of the box, one cat appears to 

 have learnt the way of escape after four successful trials 

 (the failures clearly would have no direct influence), 

 another after two trials, another after one only. That is 

 to say, after the trials mentioned, the time taken, though 

 slightly fluctuating, did not exceed half a minute. Another 

 cat appears to have learnt the trick in one trial though 

 its time afterwards rose in two cases to about a minute, 

 and after many experiments it became so well versed in 

 the matter that a lapse of seventy-two days only raised its 

 time to a little over a minute again. 2 



1 P. 45- 



2 I speak with some diffidence as to the exact numbers, as the curves 

 are not so drawn as to be particularly easy to measure. On this point, 



N 2 



