682 Journal of Comparative Neurology and Psychology. 



make color discriminations, not that he does so without training. ^^ 

 Our raccoons also required many trials before they made fairly good 

 records with the first group of papers which we used. 



We state the above conclusion tentatively only because the use 

 of reflected light is possibly inadequate for the solution of this prob- 

 lem. We regard our work as preliminary^ and we hope to complete 

 the investigation by the use of methods^'^, which had not been 

 described until our experiments were almost completed. 



Summary. (1) Open feeding vessels were unsatisfactory in test- 

 ins- visual discrimination in raccoons because of an instinctive tend- 



o 



ency they have to explore such vessels by touch. 



(2) In the cases of discrimination described in this paper the 

 raccoons seemed not to make a selection with the eyes at a distance 

 of more than four or five inches from the glasses. 



(3) So far as evidence can be gained by our use of the method of 

 reflected light, it indicates that the raccoon is able to discriminate 

 by differences of color. 



(4) Many trials were required with the first group of colors used 

 before a high per cent of right choices was made. 



(5) The animals learned most quickly to select the food-glass 

 when dark groups of colored papers were used. 



Memory for Visual Differerices. Thirty-four days after the rac- 

 coons had learned to select EVT 1 as a food-color, and without train- 

 ing during this period, they were again tried with food in this glass. 

 Eaccoon E'o. 2 gave no evidence of having been trained on this group 

 of papers and^ selected KVT 1 9, 16, 23, and 29 times respectively 

 in each of four series of thirty trials. His learning series for 

 RVT 1 was 14, 29, 24, and 30 correct choices in each series of thirty 

 trials. Raccoon 'No. 3 selected the RVT 1 27 times out of the first 



^Samojloff, a., uud Pheophilaktowa, A. Ueber die Farbeu^A•alll•nehmuug 

 l)eim Hunde. Zent. f. Physiol., Bd. 21, S. 13.3. 



-^Yerkes, R. M. The dancing mouse, p. 152 ff. 1907. Metliods of studying 

 color vision in animals. Science, N. S., vol. 29, p. 432. 1909. Also Watson, 

 .T. B. Some experiments beai-iug upon color vision in monkeys. Jour. Comp. 

 Ncur. and Psych., vol. 19, p. 1, 1909. 



