Cole and Long, Visual Discrimination in Raccoons. 675 



We next varied the experiment by rnbbing apple on the inside of the 

 food-glass. Both the raccoons then very promptly learned to select it, 

 but now we could often make out distinct sniffing and the animals held 

 the nose very dose io the top of the glass as they had not done hefore. 

 This is doubtless the best evidence we have that odor was not a giiide 

 in the color tests. If the animal had to direct his nose to the top of 

 the glasses in order to detect the odor of apple in a glass whose inner 

 surface had been thoroughly rubbed with it, then any odor less strong 



TABLE 18. 



Gray 5 in a Group of Gray 5. 



Raccoon No. 2. 



than this could have had no effect or else it surely would have 

 elicited the same behavior. The same reaction was elicited by the 

 use of meat and cake as food. The only odor not excluded by these 

 experiments is that possibly due to the different pigments of the col- 

 ored papers. True, the papers had been kept for some months in 

 the same drawer, and the glasses were close together in the row, yet 

 this only makes discrimination by the pigment odor improbable, not 

 impossible. 



