226 E. B. TITCHENER— ETHNOLOGICAL TESTS OF SENSATION. 



wind, plenty blow";*^ the inference is plain. I cannot here take the 

 space to analyze these other vocabularies ; I have chosen that of ]Mur- 

 ray Island as the most favorable to Rivers' hypothesis of defective 

 blue-vision. But I point out that the case against that hypothesis 

 does not stand and fall with my interpretation of golegolc. 



The subject of color nomenclature may be approached from an- 

 other angle. No one can read the Murray Island terms without 

 wondering what a group of civilized persons would make of colored 

 papers, if they were required to give them the names of concrete 

 objects. I thought it worth while to make an experiment on the 

 matter. I cut two-inch squares from three of the Milton Bradley 

 series, the spectral colors, the tints No. i, and the shades No. i; 

 54 colors in all. These were mixed in haphazard order, and were 

 shown to 5 observers, two women and three men. The instruction 

 ran : " You are to name these colors by first impression as soon as 

 shown. Use no abstract color names, but use always the name of 

 some concrete object that the color suggests to you. You may also 

 use the words. Dull, Dark, Bright, Light, if the stimuli impress you 

 in that way, without seeking any specific color name." I included 

 these four terms because the Murray Islanders used words of the 

 same significance. The experiment went smoothly in a period of 

 some 20 minutes ; the only modification of procedure was that, if an 

 observer gave " sky " for blue, I called for a second concrete name. 

 " Sky," as I have shown, was foreign to the Murray Island vocabu- 

 lary. I subjoin the results for all colors in which blue was involved. 



I have italicized the terms Dull, Dark, Bright, Light. In the 

 whole series of 270 namings, these words were used 64 times (I ex- 

 clude the cases in which Sky was changed to Light, though I regard 

 these as significant). In the 120 cases of what we may call the 

 blue-range, quoted above, the words occur 43 times. The general 

 percentage is thus 23.7, the percentage for the blue-range is 35.8. 

 If we take simply the three blues (GB, B, VB), the percentage is 

 35-5 (or, if the "Sky-Lights" are included, 42.2). Were then my 



*i R, 59. " These instances," says Rivers, " illustrate very well the liking 

 of these people for similes." They seem rather to show the direction of 

 practical interest. 



