46 MILDRED A. HOGE AM) RUTH J. STOCKING 



TABLE I 



were made in the discrimination tests to choose the darker, 

 and vice versa. In forming two of the groups the rats which 

 sh< >\ved no preference were groupe 1 with those which had 

 shown a preference. Three groups of two individuals each 

 were formed. Each group of two contained one albino and 

 one black-and-white rat with pigmented eyes. The following 

 table shows the composition of the groups : 



Group i. Punished after choosing d and rewarded after 

 choosing b. Contained— 



Rat i, Albino. No preference. 



5, Pigmented eyes. Preference for d. 

 Group 2. Punished after choosing b. Contained— 

 Rat 3, Albino. No preference. 



7, Pigmented eyes. Preference for b. 

 Group 3. Rewarded after choosing d. Contained— 

 Rat 9, Pigmented eyes. Preference for b. 

 11, Albino. " b. 



Group 2 was never rewarded. However, since the rats in 

 this group showed great hesitation in entering B after having 

 been punished a few times, they were occasionally fed in A, 

 sometimes after a correct choice and sometimes after an incor- 

 rect choice. 2 



RESULTS 



Table II shows the records made by the six rats. Each deter- 

 mination represents the percentage of error for a series of 10 

 trials. Ten trials a day were usually given. If, however, a rat 

 made 10 correct choices during one day, he was given five more 



1 Rats 7 and 9 were given a greater number of trials because it was found after 

 the first 50 tests had been made that the two lights had not occupied the two 

 positions an equal number of times. The extra trials were given in order to realize 

 thi-. condition. 



2 Several check tests wore made in the course of all these experiments which 

 made it evident that the discrimination was effected by vision alone (by difference 

 in intensity and possibly in color.) 



