286 FREDERICK S. BREED 



one or both of the following conclusions: (i) A difference exists 

 between the chromatic luminosity values for the chick and those 

 for the human; (2) chicks exhibit a qualitatively determined 

 preference, such as is apparent in some other animals. Since, 

 however, the preference of the chicks for red and blue varied 

 with the variations in the relative brightness of the two color 

 stimuli, brightness was clearly shown to be a controlling factor. 

 Whatever natural preference the chicks might have had for 

 red, this was overcome, in every case tried, by a sufficient rela- 

 tive increase in the intensity of the opposing blue. In view of 

 these facts it was thought best at this stage of the investigation 

 to seek such intensities of red and blue that initial preference 

 for neither would appear. This point of indifference once located, 

 it was planned to proceed with a process of training to deter- 

 mine the ability of the chicks to discriminate between two such 

 red and blue stimuli. In the final preliminary tests red, from 

 a source light 60 cm. distant from the screen, with a blue from 

 a source of about equal power, 10 cm. distant, gave the reaction 

 records of 6-4 and 7-3 for chicks no. 58 and no. 65. Red 70 

 with blue 10 resulted 4-5 and 3-7 for no. 61 and no. 65, re- 

 spectively. Hereupon, the apparatus was slightly modified in 

 preparation for the more careful preference tests that preceded 

 training on the indifferent combination such as red 70, blue 10 

 seemed to approach. Cardboards containing circular openings 

 8 cm. in diameter replaced those with rectangular openings 

 used in the preliminary tests. Circular water screens (Bausch 

 and Lomb projection -lantern cooling cells) 12.5 cm. in diameter 

 and 4.8 cm. thick were substituted for the rectangular cells in 

 use up to this time. ' Furthermore, during the preference tests 

 and throughout the following training and control tests, provi- 

 sion was made for the dark-adaptation of the animals by keep- 

 ing them in the dark-basket in the dark-room for a period of ten 

 minutes each day just prior to experimentation. Between trials 

 also each chick was kept in the dark-basket. Among the prefer- 

 ence tests, A-F, with red 70 opposed to blue 10, red was selected 

 by the four chicks 28 times, blue 52 times, out of a total of 80 

 trials. Red and blue at these intensities, although not ideally 

 balanced, were weighted on the proper side for a test of the 

 chicks' ability to form a habit of positive reaction to red, so- 



