Watson, Color Vision i?i Monkeys. II 



left open, the other half is covered with a sufficient number of 

 strips of plate glass to compensate for the absorption of the reversing 

 mirrors. To one end of this bar a spring is attached. When this 

 spring operates alone, the bar is held in snch position that the 

 beams of light have to pass through the open halves of the windows. 

 A cord, C8B, runs from the opposite end of the bar around to the 

 rod at X, which controls the miri'ors. Pulling upon this cord 

 brings the sliding bar forward to such a position that the beams 

 have to fall upon the halves of the openings which are covered 

 by the plate glass. This bar is made to work synchronously with 

 the mirrors in such a way that when the reversing mirrors are "out" 

 the plate glass windows are "in" and vice versa. A simple forward 

 pull or a release of the rod at X adjusts both mirrors and bar. This 

 compensatory device was used in certain of the control tests described 

 below (all of the red-green), but since many trials showed that the 

 reactions of the animals were not altered by its insertion or removal, 

 its use was discontinued.^ 



Behind the opaque screen, there is a vertically placed 12" x 24" 

 pane of acid ground glass (acid ground is less granular than common 

 ground glass ; milk glass would have been used, could it have been 

 obtained). The moment the screen is raised by pulling upon H, 

 the two bright-colored bands (surfaces) appear. Immediately behind 

 each of these bands (they are 8" apart) ^^^ a hole is cut in the plat- 

 form to admit the food-boxes shown in Fig. 4. A glass partition, 

 GP, set in a low wooden base serves to keep the animal from opening 

 both food-boxes at once. It also serves to force the animal to go 

 clearly to the right or to the left and to keep a position habit from 

 forming. ^^ 



•However, if one were workinc: with two colors approximately equal in 

 intensity to the animal, it mi.slit very well happen that this change in absolute 

 intensity would, owing to possible onset of the Purkinje phenomenon, alter the 

 intensity relation for the animal. 



"This distance depends upon the distance of M^ from M^. 



"Before tliis partition was at hand, one monkey, whose records are not 

 given, went always to the right, then down the screen of ground glass to the 

 left until he came to the box which contained the food. 



