38 E. L. Nichols — Sensitiveness of the Eye to Colors. 



give it a perceptible tint. The pigments selected were red 

 lead, cbromate of lead, chromic oxide and ultramarine blue. 

 These were in the form of powder, the red and blue being 

 the red lead and artificial ultra-marine of commerce, whereas 

 the chromium compounds were freshly prepared by precipita- 

 tion. Each of these pigments was mixed with white in the 

 following manner. About ten cubic centimeters of the powder 

 was mixed in the dry state with an equal volume of magne- 

 sium carbonate, the mixture was divided into two equal parts, 

 half of it was again mixed with its own volume of the white 

 powder, the product was again subdivided and the process of 

 mixing with white by equal parts was repeated until all traces 

 of color had disappeared. Since at each stage of the process 

 only half the material was used for further dilution, there 

 remained a series of colored powders of which the pure pig- 

 ment formed the first, while the succeeding numbers were of 

 less and less saturated hue, and finally could not be distin- 

 guished from white. These mixtures were put into small vials 

 of white glass and labelled in such a manner as to ensure their 

 recognition by persons acquainted with the code and at the 

 same time to preclude the detection of the nature of their con- 

 tents from the label, without such knowledge. 



For the purpose of ascertaining the degree of saturation at 

 which the presence of these pigments becomes perceptible to 

 the eye, the four sets of bottles, containing mixtures of red and 

 white, yellow and white, green and white, and blue and white, 

 were mingled indiscriminately, and the observer whose eye 

 was to be tested was requested to arrange those in which he 

 could detect any trace of color, according to hue and degree of 

 saturation. The bottles were afterwards inspected by some 

 one acquainted with the code of labels, who threw out those 

 not in the proper set and recorded the number of bottles 

 remaining in each set and the number of each color which had 

 been properly placed as to shade. From the former record 

 the sensitiveness of the eye to colors of low saturation was 

 determined; the latter data served to indicate the ability of 

 the observer to detect small differences of shade. 



Fifty four persons, all of them with two or three exceptions 

 between the ages of fifteen and thirty, were examined in this 

 way. The Holmgren worsteds had shown one of them to be 

 completely green blind, three partially so and one partially 

 red blind. Color-blindness was not found to affect in any 

 marked way their ability to classify the colors. 



This method of measuring the sensitiveness of the eye is not 

 in all respects satisfactory. A method in which pure spectral 

 tints mixed with white light could be compared with a field 

 illuminated by white light alone and the amount of monochro- 



