ACCIDENTS BY RAIL AND SEA. 185 



alike; that the only question is the resemblance of tbe color; and that 

 consequently lie must endeavor to find something similar, of the same 

 shade, something lighter and darker of the same color, etc. If the 

 person examined cannot succeed in understanding this by a verbal 

 explanation, we must resort to action. We must ourselves make the 

 trial by searching with our own hands for the skeins, thereby showing 

 in a practical manner what is meant by a shade, and then restoring the 

 whole to the pile except the sample skein. As it would require much 

 time to examine each individual in this way, it is advisable, when ex- 

 amining a large number at the same time, to instruct all at once, and 

 moreover to ask them to attentively observe the examination of those 

 preceding them, so as to become more familiar themselves with the 

 process. By this, time is saved, without loss of security; for no one 

 with a defective chromatic sense finds the correct skeins in the pile 

 the more easily from the fact of having a moment before seen others 

 looking for and arranging them. He makes the same characteristic 

 mistakes; but the normal observer, on the other hand, generally ac- 

 complishes his task much better and more quickly after having seen 

 how it must be done, and this is the advantage of our method. 



The colors mentioned in this chapter are divided into two classes : 



1st. The colors for samples {test colors'), that is, those presented to the 

 j)ersons examined ; and 



2d. The " colors of confusion,''^ that is to say, those which the color-blind 

 selects from the heap, because he confuses them with that of the 

 sample. 



Test I. — The green sample is presented. This sample should be the 

 palest shade (the lightest) of very pure green, which is neither a yellow- 

 green nor a blue-green to the normal eye, but fairly intermediate be- 

 tween the two, or at least not verging upon yellowish green. 



Hule. — The examination must continue until the one examined has 

 placed near the sample all the other skeins of the same shade, or else, 

 with these or separately, one or several skeins of the class correspond- 

 ing to the "colors of confusion," until he has sufficiently proved by his 

 manner of doing it that he can easily and unerringly distinguish the con- 

 fused colors or until he has given proof of unmistakable difficulty in 

 accomplishing this task. 



Diagnosis. — He who places beside the sample one of the "colors of 

 confusion," that is to say, finds that it resembles the " test color," is 

 color blind. He who, without being quite guilty of this confusion, evinces 

 a manifest disposition to do so, has & feeble chromatic sense, 



BemarJc. — We must remember that we cannot allow more than five 

 colors for " confusion." But we have here in view, not every kind of defec- 

 tive color-sense, but only those important in the business of railways. 

 The number of colors allowed is therefore sufiicient, as these are the 

 most important and most common. 



Test H. — A purple skein is presented. The color chosen must be 

 between the deepest and lightest shades of the scale. 



