32 ROBERT M. YERKES AND JOHN B. WATSON 



Professor E. C. Sanford at Clark University, indicate both the 

 serviceabiHty and disadvantages of colored papers in this kind 

 of work. 



In order of preference, we mention the following sets of papers : 



1. The Hegg colored papers. (Pfister und Streit, ]\Iath. 

 physikalische Werkstatte, Bern. S\Yitzerland). — These are mix- 

 tures of oils on paper yielding the hues red, yellow, green, and 

 blue. These hues are claimed to be equal in intensity and satura- 

 tion for the human eye. The set is useful as a means of ascer- 

 taining, in a preliminary survey, whether an animal readily 

 discriminates two hues which for us are of nearly the same 

 intensity and saturation. 



2. The Wundt colored papers. (E. Zimmermann, 21 Emilien- 

 strasse, Lei]:zig, Germany.) A series including the various 

 spectral hues and purple in a number of saturations (chromas). 



3. The Hering colored papers. (R. Rothe, 16 Liebigstrasse, 

 Leipzig, Germany. C. H. Stoelting Company, Chicago, 111., 

 American Agent.) A series of twelve strong colors, including 

 purple. 



Rothe supplies also a set of fifty neutral papers, ranging 

 from white to black, but it is unsatisfactory. 



4. The Bradley colored papers. (Milton Bradley Company, 

 Springfield, Mass.) A useful series, including a great variety 

 of colors and saturations. 



Colored cloths. — Cloths, as secondary sources of chromatic 

 stimuli have most of the defects and few, if any, more merits 

 than papers. They are useful, under certain circumstances, for 

 qualitative work. 



Especially valuable in this class of reflecting surfaces, because 

 it reflects only a small quantity of white light, is silk velvet. 

 In experiments which demand change in the intensity of the 

 chromatic stimulus without marked change in its hue or satura- 

 tion this material, properly dyed, is superior to colored papers. 

 A great variety of hues of silk velvet are on the market, but they 

 are quite likely to prove unsatisfactory for tests of color vision 

 because their dyes reflect, in varying amounts, light of difterent 

 w^ave-lengths. Could we obtain a set of these cloths which 

 reflected respectively only red, yellow, green, blue, and violet, 

 we should be able to use them to advantage in many of our 

 qualitative experiments on vision in animals. But they would 



