1 8 'Journal of Comparative Neurology and Psychology. 



Red given three times on right. 



Red given three times on left. 

 Red given three times on right 

 Red given three times on left. 



On April 22, in the nine trials given, the position of the red and 

 green was alternated. 



The Results of the Experiments. 



The tables given below show, separately for each animal, the 

 daily set of conditions to which he was reacting, and the results 

 of the test. The date of the experiments is given first in the tables, 

 then, in order, the number of choices of the positive color (i. e., 

 color with food), the number of choices of the negative color, the 

 percentage of correct choices, the source from which the spectrum 

 was obtained, the relative and absolute intensity of the two bands 

 (refer for photometric statements to table of constants, p. 15), 

 and the remarks. 



A reference to the tallies shows that they are divided into two 

 parts: A, with stimulus constant, and B, with stimulus variable. 

 The first part of division A, shows in all cases the gradual rise of 

 discrimination, Avhile the second part gives the animal's maximum 

 of steadiness under an unchanging set of conditions. When a max- 

 imum of steadiness had apparently been obtained, changes in the 

 relative brightness of the two bands and changes in their form and 

 surfaces were introduced. Division B of each table shows clearly 

 just what changes were made and the effect of such changes upon 

 the i^ercentage of correct choices. 



Discussion of Results. 



1. The surprising result of the early part of the test on the red- 

 green is seen to be the failure of the animals to read to the red. 

 This was noticeable in all of the animals tested, see especially Tables 

 I and II. The early records of monkey B. are not given in Table 

 III. His failure to react to red on the early trials was as pronounced 



