E. B. TITCHENER— ETHNOLOGICAL TESTS OF SENSATION. 217 



may have thought it fairer to leave them in. The Murray Islanders 

 and the Englishmen now agree as regards red and yellow ; both the 

 liminal values and their mean variations are of the same order; but 

 they differ markedly as regards blue. 



On this point I have two remarks to offer, (i) In experiments 

 which I have made with Hering's colored glasses, my observers 

 have reported that it is difficult to determine the chromatic limen of 

 R and B (I have had no such report for Y), because the faintly 

 colored glass shows, at the moment of exposure, a flush or wave of 

 color which immediately disappears through adaptation. It is pos- 

 sible that the four English observers who were " exceptionally in- 

 sensitive to red " failed to notice this momentary flush in the neigh- 

 borhood of the limen, while the other Englishmen and the Murray 

 Islanders (whom we know to be keenly interested in red) noticed it 

 and placed the limen accordingly. The range of results for the 

 Murray Islanders is 40 to 5 ; the range for the 14 Englishmen is 40 

 to 10 : a remarkably close agreement. In the case of blue, it is pos- 

 sible that the 16 English observers noticed the flush and reported it 

 as color, while the Murray Islanders (who are definitely uninterested 

 in blue, since they have in their surroundings no blue object of rev- 

 erence or utility) disregarded it. The range for the Murray Is- 

 landers is 100 to 30, and that for the 16 Englishmen 80 to 15; the 

 ranges are thus of the same order of magnitude, are very large, and 

 show a wide overlap. My distinction, therefore, must not be pressed ; 

 the possibility is rather that the Murray men tended to overlook the 

 flush, the Englishmen to report it as blue. In the case of yellow 

 (for which, as it happens, I have no report of the flush) the ranges 

 are 50 to 10 for the Murray Islanders and 60 to 4 for the 18 English- 

 men. The agreement is less striking than for red, but is still fairly 

 close. 



There is, then, a possibility that the difference in the case of blue 

 may be due, in part at least, to regard or disregard of the supralim- 

 inal flush of color to which I have called attention. I have myself 

 had student-observers who disregarded the flush, both of R and of 

 B, because they thought the field should look colored during the 



