140 WALTER S. HUNTER 



• E " 



the path by which they climbed into the box: here the rats 



merely initiated the second complete cycle by necessarily elimi- 

 nating the first one. In all essentials the behavior of the rats 

 and the raccoons is in agreement. The reader may object, 

 however, that the two sets of results differ fundamentally in 

 that it is not known but what the rats would have behaved in 

 the manner described even though they had not been " put 

 through;" that, in other words, it was either chance or curiosity 

 which led the rats into the box. In answer to this I submit 

 the following: (i) The two rats that entered the box stayed 

 in it until taken out. Had they been exploring merely, they 

 would have left the box in a short time. (2) Chance is not 

 the explanation, because the behavior was repeated for a little 

 more than eight days. Again, (3) evidence against both chance 

 and curiosity as the explaining factors is to be found in the fact 

 that three rats failed to enter the box although the two others, 

 as just stated, did so for some days. All five rats were close 

 together on the table, so that this failure of three rats to follow 

 the other two is very noteworthy in light of the results obtained 

 by Berry." This investigator records many instances where 

 rats that had failed to leave a box, by doing which they would 

 have secured food, did leave the box after they had seen (?) 

 others give the reaction several times. 



Let us now attempt a correlation of the three sets of data 

 above given. Cole's results and those of the present writer 

 oppose Thorndike's in the sense that both find that some animals 

 (even those as low in the scale as the white rat), will learn certain 

 problems by being put through them. Two possible reasons 

 may be given to account for the fact that Thorndike's cats did 

 not learn the reaction in question: First, they were not given 

 enough trials. On a priori grounds, it is a valid assumption that 

 it may take a much larger number of trials for an animal to learn 

 a problem when being put through than when it experiences 

 its own initiation of the act. Yet on the middle of page 67/ 

 the inference is to be drawn that Thorndike thought no more 

 trials should be required. Even twice the number of trials 

 is not sufficient to justify the conclusion that an animal cannot 



8 Berry, C. S. The Imitative Tendency of White Rats. 1906. Jour. Comp. 

 Neur. and Psychol., vol. 16. 



See particularly experiments I and IV. 

 7 Op. cit. 



