350 H. M. JOHNSON 



ceeded in circumventing all these methods at the time of his 

 death. Monkey 2 developed a similar tendency. I never used a 

 charge heavy enough to make him tend to avoid the alley where 

 he was last shocked. When I introduced punishment into the 

 work on him I had already changed the wiring of the box as des- 

 cribed above. He has taken one of the grills by the free end 

 and depressed it ten or twelve times in rapid succession uttering 

 a slight exclamation at each shock. And yet he avoided touching 

 them while making a choice. The long breaks in his record 

 prior to March 11 are due to the necessity of modifying the box 

 so as to limit his movements. The change which finally proved 

 effectual was installing a removable wooden partition extending 

 across the box at the entrance to alleys A 1 and A 2 from floor to 

 ceiling. It contained two windows each 4 inches square, so 

 placed that its axis coincided with that of the test-object at the 

 other end of the alley. The animal crawled slowly through the 

 window at each trial and dropped on to the grill. The position 

 of the windows permitted the animal to view the test-objects 

 normally but prevented him from viewing them simultaneously 

 as the dog and the chicks could do. This was undesirable, and 

 at the suggestion of Professor Watson I substituted for the wooden 

 partition one made of plate glass 1-4 inch thick containing 

 circular windows 4 inches in diameter corresponding to those 

 in the wooden partition. The change caused no disturbance. 

 As the test-bands were reduced in width to the point where dis- 

 crimination became difficult, the monkey would come to the 

 partition and bring his head before first one opening and then 

 before the other in rapid succession. In the latest stages of 

 work on this problem he would sometimes peer through a given 

 opening two or three seconds. In the earlier stages he apparently 

 regarded the stimuli through the plate glass, and his movements 

 were very rapid. 



It has probably occurred to the reader that it was possible 

 for this animal to extend his head several inches through these 

 windows without dropping on to the grill. Thus he could bring 

 his eye nearer to the test-object than the 60 cm. distance which 

 the partition fixed. This would have introduced a variable 

 factor hard to control. Until May 7 I did not watch the animal 

 in the act of choosing, but his movements were so rapid that I 

 could follow them by the noises he made with fair accuracy. 



