12 ROBERT M. YERKES 



means of testing for various types of behavior. Thorndike him- 

 self devised various forms of apparatus and problem, while at 

 the same time making them contribute most stirringly to our 

 knowledge of the psychology of the chick, cat, dog, and monkey. 

 Kinnaman, Small, Porter, Watson, and a host of other Ameri- 

 can and European experimentalists followed Thorndike's lead 

 in the application of experimental devices to the analysis of 

 problem-solving behavior. 



It may not be amiss to point out that the puzzle-box method, 

 although an important advance scientifically over the casually 

 or inexactly arranged situations of the earlier period — not to 

 mention the anecdote — does not adequately fulfill the require- 

 ments of comparative and statistical method. True, it has 

 possibilities of adaptation or improvement in these respects 

 which have never been realized, but the fact is that mostly the 

 data of response to a puzzle-box problem or situation are so 

 meager and inexact as to be of scant value for purposes of com- 

 parison or statistical treatment. Comparative and genetic 

 psychology alike demand methods which shall yield precise, 

 varied, and comparable data of reaction from measurements of 

 various stages, types, and conditions of organization. 



L. W. Cole departed from the well-worn path which Thorn- 

 dike had earlier broken, in originating the serial stimulus method 

 of testing for imaginal or ideational behavior. This method, 

 also, was ill-adapted to statistical needs, and like the earlier 

 procedures, yielded only roughly comparable data. As thus far 

 used, it is an indicator of problems rather than a scientifically 

 exact instrument for solving them or of obtaining detailed de- 

 scriptions of behavior. It has already served an important end 

 in breaking up the monotonous succession of problem-box 

 studies. 



Simultaneously with Cole's work on raccoons, which really 

 revived interest in animal ideation, Hamilton, from a very 

 different direction, attacked the general problem of reactive 

 tendencies. As a psychiatrist, he had become deeply interested 

 in applying the comparative method to the problems of psychiatry 

 and in bringing the facts of animal psychology and genetic psy- 

 chology to bear upon the practical problems of mental disease 

 and defect. His first experimental attempt was a study of reac- 

 tive tendencies in the dog. Over a period of ten years, he has 



