666 'Journal of Comparative Neurology and Psychology. 



one carefully by touch and sight, then the next, and so on until the 

 vessel with food in it was found. It would then go on in the same 

 way to the end of the row, and often back again, rarely skipping a 

 single vessel. When, in the later trials, they made 24 per cent of 

 right choices, they of course passed by some of the glasses which did 

 not contain food. 



Because of this instinctive propensity of the raccoon, it seemed 

 plain to us that open feeding vessels would not serve satisfactorily to 

 test the visual sense of this animal. We did not, however, leave the 

 question to be decided by observation alone. For after having tested 

 our animals by means of closed vessels for some time we returned, 

 in the case of Raccoon ISTo. 3, to the use of the Kinnaman apparatus 

 to see whether the former type of numerical record would reappear. 



Thus we have in the following tables, the record of this raccoon 

 while learning to discriminate RVT 1 and, after his choices were 

 nearly perfect with closed glasses, his record on the same colors with 

 open glasses. The figures represent series of thirty trials each. 



TABLE 4." TABLE 5. 



Closed Glasses. Open Glasses. 



It is evident from these records that open feeding vessels may 

 actually obscure a discrimination habit which is fairly well estab- 

 lished. Yerkes has shown, in the case of the crab, that a test which 

 runs counter to a strong instinctive impulse is unsatisfactory.^^ As 



"Food was placed in all the glasses after the first series of thirty trials 

 and the food-glass was frequently exchanged for another of the same color 

 in order that it might not become soiled by the animal's paws. 



"Yekkes, R. M. Habit formation in the green crab, Carcinus granulatus. 

 BloL Bull., vol. 3, p. 241. 1902. 



