96 JOSEPH U. YARBROUGH 



that are variable from trial to trial otherwise they cannot be 

 selective in nature. 



No temperature controls were used. They were thought to 

 be unnecessary because of the following: 1. The lights were 

 turned on but for a short time. 2. They were outside of the 

 main apparatus. 3. The cats oriented immediately when the 

 lights were turned on and reacted precipitately when released. 

 And, 4, the behavior of the cats on light was the same as that 

 of those on sound where temperature could not be involved. 



B. "Delayed" experiments. — Since in the first four delays used 

 the entire reaction was not performed after the stimulus had been 

 removed, it is probable that they should not be termed delays 

 at all. The stimulus was always continued until the experi- 

 menter was convinced from all external evidence that the cat 

 had become aware of its presence. 



The cats tested on sound and those on light were all pre- 

 sented their problems by the method described above, but for 

 convenience the data will be discussed separately. 



(a) Set A (cats tested on light). — 



Delay I. — In delay I the light was turned off just as the cat 

 reached the correct compartment. Bobby was given 30, Jim, 

 20 trials; and for both of them each trial was successful. With 

 the association well established, the turning off of the stimulus 

 at this point in the reaction effects no change in their percentage 

 of correct response. 



Delay II. — In this delay the stimulus was cut off when the 

 cat was half way from the release box to the correct compart- 

 ment. 2 Jim was given 10 trials with all of them correct. Bobby 

 was given 60 trials with 56 correct. There appears to be no 

 difficulty in making the step from delay I to delay II, even 

 though the cats here made one-half of the distance of response 

 in the absence of the stimulus. After the cat is well set out, then, 

 on his reaction, the stimulus may be withdrawn without affect- 

 ing the response. 



Delay III. — The only difference in this delay and number II 



is that here the stimulus is withdrawn before the cat is well on 



2 In case the cat started from the release box in a different direction from that 

 of the stimulus, e.g., if he started toward c when the stimulus was at a, the stimulus 

 was not turned off until he did turn in the direction of the stimulus compartment, 

 and so in this case, was well on his way. 



