20 K. S. LASHLEY 



Tests at these values were carried out with the results shown 

 in table 7. The ratio of 45 to 1 gives uncertain results owing 



TABLE 7 



Tests for yellow-green vision with varied brightness relations of the stimulus 

 lights. Figures in italics show where the yellow is brighter for the chick. 



Energies Chick A. Chick D. 



Yellow Green Trials Errors Trials Errors 



1.0 1.0 14 2 20 4 



.01 1.0 20 2 20 2 



.11 2.2 14 1 



.05 1.0 10 



2.0 i.o 29 6 12 1 



1.0 1.0 10 1 



to difficulty with the apparatus: a reduction of the yellow below 

 .1 of the standard brings it near the chick's threshold 3 and 

 a greater energy of the green than 2.2 could not be obtained 

 with the present arrangement. The tests are therefore not as 

 extensive as I desired. However, chicks long trained to bright- 

 ness discrimination failed to distinguish this red and green even 

 at the energies of 1 to 1 and the tests with variations in both 

 directions from this ratio seem adequate to eliminate a reaction 

 to brightness. As in the case of red and green the effective 

 stimulus is restricted to wave-length. 



The evidence for color vision thus far presented seems to 

 prove that the chick can distinguish between monochromatic 

 lights of any intensity between threshold and the Pfund standard, 

 irrespective of the brightness or saturation. The effective stim- 

 ulus is the wave-length. The question remains, is the discrim- 

 ination based upon retinal sensitivity to differences of wave- 

 length or upon some entoptic phenomenon. 



The difference in visual acuity in light of different wave- 

 length is well known, and, though it seems to be largely a matter 

 of relative brightness it may furnish a basis for discrimination. 

 Whether a like variation in acuity occurs in the color blind I 

 cannot discover from the literature. Allied to this phenomenon 

 is that of diffraction or dispersion within the eye. At low 

 intensities the red and yellow stimulus patches have for me 

 sharp outlines while the green of equal brightness lacks sharpness 



3 The absolute intensities of the lights in this experiment are not comparable 

 with those of the Purkinje tests. The introduction of the smoked wedges cut 

 down the energy of the spectrum so that the equation of the lights with the Pfund 

 standard had to be made before they were reflected from the plaster surfaces instead 

 of afterward as in the first experiment. 



