518 



SCIENCE. 



[N. S. Vol. VII. No. 172. 



mum for the red is in the brightest part of 

 the spectrum affecting the eye as red ; it 

 would ordinarily be called scarlet. The 

 maximum for violet is likewise in the 

 brightest part perceived as violet ; it would 

 be called a bluish violet, or almost ultra- 

 marine. The maximum for green is in the 

 brightest part of the green. Maxwell's 

 curves, obtained experimentally by use of 

 his col or- box (Phil. Trans., 1860), differ 

 from those estimated by Helmholtz. The 

 maxima for red and green are each shifted 

 very decidedly toward the yellow, while 

 that for the violet is in the typical blue. 



into the violet and red regions, but by no 

 means to the ends of the spectrum; and 

 the red curve reaches but little beyond the 

 F line in the blue. The violet maximum is 



Kg », 



An elaborate investigation was under- 

 taken, under the direction of Helmholtz, 

 by Koenig and Diderici, and published in 

 1886. {Sitzungsberichte der Koeniglich Preus- 

 sisehen Akadeinie der Wissenschaften zu Berlin, 

 22 Juli, 1886.) This was based on the ex- 

 amination not only of those having normal 

 eyes, but of several persons with different 

 grades of color-blindness. No English 

 translation of this paper has thus far been 

 published, and it is to be regretted that 

 some parts of it are wanting in clearness, 

 if judged by American standards. The 

 collective color equations were deduced 

 from observations with a Helmholtz color- 

 mixing apparatus, but the reader is in- 

 formed that " the experimental details de- 

 mand for their representation so much 

 space that we cannot here go into them." 

 The final result is shown in the curves of 

 Fig. 2, which should be compared with 

 those of Fig. 1. It is seen that the violet 

 curve extends here from the H line only to 

 the D line ; that the green curve extends 



fi or 



at -I = 0.445 /I, corresponding to ultramarine. 

 The green maximum is at X = 0.545 /j., cor- 

 responding to a yellowish green. The red 

 maximum is at -'. = 0.564 /x, which corre- 

 sponds not to red, orange, or even pure 

 yellow, but to a slightly greenish yellow. 

 This is so wide a departure from the earlier 

 ideal curve of Helmholtz, and from Max- 

 well's result, that one is tempted to ask 

 whether there may not have been some 

 very arbitrary assumption involved in the 

 deduction of one or the other. The over- 

 lapping of the curves emphasizes the well- 

 known fact that even the most nearly pure 

 of spectral colors near the middle of the 

 spectrum are decidedly impure. Taking- 

 account of this fact, the authors reach the 

 conclusion that the pure fundamental sen- 

 sations would coiTcspond to hues about as 

 follows : for red, A = 0.671 p-, which is near 

 the red end, between the B and C lines, in- 

 stead of the maximum, of the red curve; 

 for green, X = 0.505 /-i, a slightly bluish 

 green, which is close to the intersection of 

 the green and violet curves, instead of the 

 maximum of the former ; and for violet, 

 X = 0.470 /J., a pure blue, which is decidedly 

 nearer the middle than is the maximum of 

 this curve. The reader is not favored with 

 the calculations by which these wide differ- 

 ences are found between the curve- maxima 

 and the spectral positions corresponding to 

 the fundamental sensation hues. 



