186 COLOR-BLINDNESS IN ITS RELATION TO 



Eule. — The trial must be continued until the one examined has placed 

 near the sample all or the greater part of the skeins of the same shade, 

 or else simultaneously or separately one or several skeins of "confusion." 

 He who selects either the light or deep shades of blue and violet (es- 

 pecially the deep) or the light or deep shades of one kind of green or 

 gray inclining to blue has committed an error. 



Diagnosis. — 1. He who is color-blind by the first test, and who, upon 

 the second test, selects only i^urple skeins, is incompletely color-blind. 



2. He who, in the second test, selects with purple only blue and violet, 

 or one of them, is comjyletcly redhlind. 



3. He who, in the second test, selects with purple only green and gray, 

 or one of them, is completely green-blind. 



Remarli. — The red-blind never ratifies the test of the green-blind, and 

 vice versa. However, it happens in certain cases that the green-blind 

 selects a violet or blue skein, but always the lightest shades. This 

 should not affect the diagnosis. The examination may end with this 

 test, and the diagnosis be considered as perfectly settled. It is not even 

 necessary, practically, to decide whether the color-blindness is red or 

 green. But to be more entirely convinced of the relation of complete 

 color-blindness with the signal colors, and especially to convince, if neces- 

 sary, the railway employes and others who are not specialists, the exami- 

 nation may be completed by one more trial. The one we are going to 

 mention is not necessary to the diagnosis, and only serves to corrobo- 

 rate the investigation. 



Test III. — The red skein is presented to the subject. It is necessary 

 to have a vivid red color like the red flag used as signals on railways. 



Rule. — This test, which is applied only to those completely color-blind, 

 should be continued until the person examined has placed beside the 

 specimen all the skeins belonging to this shade or the greater part or else 

 separately one or several "colors of confusion." The red-blind then 

 chooses, besides the red, green and brown shades which, to the normal 

 sense, seem darker than red. On the other hand, the green-blind selects 

 opposite shades which appear lighter than red. 



Bemarlc. — Every case of complete color-blindness discovered does not 

 always make the precise mistakes we have just mentioned in the preced- 

 ing examinations. These exceptions are either instances of persons 

 with a comparatively inferior degree of complete color-blindness, or of 

 color-blind persons who have been exercised in the colors of signals, and 

 who endeavor not to be discovered ; they therefore usually confound at 

 least green and brown, but even this does not always happen. 



Additional note. — We have not given rules for discovering total 

 color-blindness, because we have not found any cases of this kind. If 

 any such should be found, they will be recognized, according to the 

 theory, by a confusion of every shade having the same intensity of 

 light. Violet-blindness will be recognized by a genuine confusion of 

 l^urple, red, and orange in the second test. The diagnosis should be 



