Polarization Flicker Photometer. 365 



flicker between two fields of unequal brightness : 



»= J --4i h±hL, .... (2 ) 



where &> = critical speed, 



$ = brightness discrimination fraction, 

 I ± and I 2 = brightness of the two fields, 

 K = diffusivity, 

 X = depth of penetration. 



If we assume that K is a function of the mean brightness, 

 we can, by holding the latter constant as I 2 and I 2 are varied 

 relatively to each other, make a test of the theory, in so far as 

 it assumes a difFusivity apart from any question as to how this 

 varies with intensity. Fig. 1 of the paper referred to shows 

 critical frequencies for various values of I x and I 2 under 

 these conditions, which were not attainable experimentally 

 with the apparatus then available. 



With the new apparatus just described the exact conditions 

 for this test are met. In fig. 5 are shown the results of two 

 extended sets of measurements of critical speeds for varying 

 ratios of two fields, with the mean brightness held constant. 

 The upper points were obtained at a relatively high, the 

 others at a relatively low, illumination. The full lines are 

 graphs of the equation (2), with a value & of *001, using 



X 2 



the values of -^ solved for from the end points. It is 



clear that the theory is very beautifully borne out. 



(b) Mean Brightness Varied. 



The second theoretical paper gives in equation (6) a 

 relation for the critical speeds at different illuminations, 

 which upon inserting the complete expressions for the 

 constants becomes 



n 2 ii-i 2 -] 2 



^= |_og g + ^y 1 ^ (alogL + ft), • • (3) 



N) 



where I A is the mean brightness, and a and b are the constants 

 (as determined by experiment) applying to the limiting case 

 of light alternated against darkness. It follows from this 

 equation that, critical, speeds for the limiting case being 



