12 K. S. LASHLEY 



have determined are very rough but they indicate that the 

 relative stimulating value for the light adapted eye of mono- 

 chromatic bands of high intensity is proportional to the threshold 

 values of the same wave-lengths with light adaptation, as de- 

 termined by Watson ('15). 



SENSITIVITY TO WAVE-LENGTH 



As the primary object of these experiments was to determine 

 whether or not the chick is sensitive to differences of wave- 

 length and as six months had been consumed already in the 

 preliminary tests it was thought best to abandon the latter and 

 test for color discrimination. Red 650 w. and green 520 /*/*. 

 at equal energies were used as stimuli. Preference tests were 

 of course impossible with chicks A and C. 2 Learning was 

 rapid and after 200 trials the chicks were judged ready for 

 control tests. The control tests described on page 5 limited 

 the reaction to the illumination of the stimulus plates. Tests 

 to distinguish between reaction to energy and to wave-length 

 were made by altering the brightness relations of the saturated 

 lights, and by diluting them. As determined in the Purkinje 

 tests the brightness equation point for 650 pp. and 520 ^/t is some- 

 where near the energy relations of 8 to 1. The energy relation 

 of the stimulus lights was varied in both directions from this 

 ratio as described in table 1. Tests with the colors diluted 

 were made as indicated in table 2. The results of these tests, 

 summarized in table 3, show that a reversal of the brightness 

 values of the red and green did not seriously disturb the accuracy 

 of the discrimination. The errors made appeared when the 

 energy of the positive light was much reduced, irrespective of 

 whether this was red or green. The dilution tests give fairly 

 clear evidence of discrimination except in the case of Chick C 

 with dilute red. 



2 The behavior of some other birds in tests for color preference seems worth 

 recording. Training for red-green was begun with five hens but not completed. 

 Four of these showed no color preference for the lights at the brightness equation 

 point for light adaptation. The fifth under the same conditions chose the green 

 in preference to the red in 20 successive trials. This behavior suggests either an 

 acquired color preference or color-blindness. Unfortunately, this bird was killed 

 by rats before her vision could be tested. 



