184 COLOR-BLINDNESS IN ITS RELATION TO 



est or darkest shades. The well-defined kinds and degrees of a defect- 

 ive chromatic sense confound only colors of mean intensity. I have 

 selected, to determine whether the chromatic sense is or is not defective, 

 a light green (dark green may be also used), because green, according 

 to the theory, is the whitest of the colors of the spectrum, and conse- 

 quently is most easily confused with gray. For the diagnosis of the 

 especial kinds of partial color-blindness, I have selected purple (pink), 

 that is, the whole group of colors in which red (orange) and violet (blue) 

 are combined in nearly equal i^roportions, at least in such proportions 

 that no one sufficiently preponderates over the others, to the normal 

 sense, so as to give its name to the combination. This is the reason for 

 this choice. Purple occupies a singular position amongst colors; although 

 it is a combination, it is, we know, a color, as well " saturated " as the 

 colors of the spectrum, and might be, from this point of view, classed 

 with them, although it is not found in the spectrum. In fact, it has been 

 regarded as the eighth color of the spectrum, closing the circle of satu- 

 rated colors. Purple is of especial importance in the examination of the 

 color-blind, for the reason that it forms a combination of two fundamental 

 colors — the two extreme colors — which are never confounded with each 

 other. In fact, from a color-blind point of view, one of two things must 

 liappen, according to the theory: either it excites but one kind of per- 

 ceptive organ or it excites them all. It appears then either like a sim- 

 ple color, that is to say, like one of the two colors of the combination, 

 or like white (gray). Experiment has confirmed this hypothesis. Our 

 sample colors, therefore, are the two complementary colors of each other, 

 green and purple. In the examination of the chromatic sense of a large 

 number of individuals, it is, of course, of importance to decide, first, 

 whether the chromatic sense of the individual is or is not normal. It is 

 only after establishing the existence of a defect that its nature or degree 

 must be determined. The sample colors are, therefore, employed with 

 more advantage in a certain order, as the test must be accomplished as 

 a whole, according to a plan that experience has proved the surest, most 

 rapid, and, finally, most suitable for the purpose. 



3. — TnE EXAMINATION AND DIAGNOSIS. 



The Berlin wool is placed in a pile on a large plane surface and in 

 broad daylight; a skein of the test color is taken from the pile and 

 laid aside far enough from the others not to be confounded with them 

 during the trial ; and the person examined requested to select the other 

 skeins most resembling this in color, and place them by the side of the 

 sample. In the first place, it is necessary that he should thoroughly 

 understand what is required of him ; that is, that he should search the 

 pile for the skeins making an impression on his chromatic sense, in- 

 dependent of any name he may give the color, similar to that made 

 by the sample. The examiner should explain that resemblance in 

 every respect is not necessary; that there are no two specimens exactly' 



