52 



FREDERICK S. BREED 



1. 

 2. 

 3. 



4. 



5. 



6. 



7. 



8. 



9. 

 10. 

 11. 

 12. 

 13. 

 14. 



Preference tests B and D, table 12, indicate a natural pref- 

 erence for the brighter of the two cards. The orange-blue and 

 yellow-blue preference tests brought to light interesting data. 

 The yellow here was very light, much brighter than the com- 

 paratively dark blue, not only as judged directly by the human 

 eye, but as tested by the flicker method. Particular attention 

 is called to these first reactions to blue for, in so far as this 

 paper has anything to contribute to the study of color vision, 

 the interest will largely center in and results will turn upon the 

 reactions of the chicks to blue. 



The training of no. 8 on the black-blue was now carried through 

 with the result detailed in table 12. Within a reasonable 

 nvimber of trials, 140, including the final three perfect series, 

 the chick nicely developed the habit of selecting the black. 

 It was thereupon decided to adopt the black-blue combination 

 for a more searching test, with a view to its use in the determina- 

 tion of the rate of modification, the permanence of habit, and 

 some of the effects of training. 



Chicks nos. 16, 17, 18, 19, and 20 were selected for this work. 

 From all appearances they were normal chicks, and were chosen 

 at random from the healthy specimens in a flock of seventeen. 

 They were twelve days old when given their preference tests. 

 In all tests prior to those reported in table 13 but one set 

 of cards was used, the cards at D, figure 8. In this set of ex- 

 periments, after series 2, an additional ]3air of cards was placed 

 at F, whereby shock and color stimuli were presented simul- 

 taneously and not successively only, as in the previous experi- 

 ments. In table 13 the rate of decrease of error or the rapidity 

 of modification is shown in the column, " average number of 



