Polarization Flicker Photometer, 369 



is to be noted, however, that this high sensibility, located 

 as it is in the ultimate receiving apparatus, can never be 

 utilized to its full value for the detection of small fluctua- 

 tions in the stimulus. Because of the damping effect of the 

 channel through which the impression of the fluctuating 

 stimulus is transmitted, only very considerable fluctuations 

 retain any amplitude large enough to be detected. In other 

 words, the very small value of the brightness discrimination 

 fraction of the ultimate receiver for successive impressions 

 is all that makes possible our appreciation of ordinary 

 fluctuations in the stimulus. If this had the value of the 

 ordinary Fechner fraction we could tolerate much lower 

 frequencies of light fluctuation than we now do. On the 

 other hand, if we could stimulate the ultimate receiver 

 directly, without the interposition of the present trans- 

 mitting channel, it would be necessary to use rather high 

 speeds of alternation to make flicker vanish. Upon substi- 

 tuting different values of 8 in equation (4) of the second 

 paper on the theory, the effect on g> is given numerically. 

 A value of "01 for 8 in place of '00018 about halves the 

 critical speed. 



6. Critical Speed Relations with a Difference of Colour 

 between the Two Fields under Comparison. 



It follows from the treatment, in the two theoretical 

 papers, of colour flicker and of brightness flicker with 

 unequally bright fields, that critical speeds should follow 

 the same law in the case of two coloured fields as with two 

 unequally bright fields, up to the point of the appearance of 

 colour flicker. The latter interposes a bar to the progressive 

 decrease of speed to zero which occurs at the equal brightness 

 point for fields of the same colour. Detailed discussion of 

 the statement just made may be avoided by reference to the 

 diagrammatic insert to fig. 7. Here the two fields under 

 comparison are supposed to be of different colour. The left- 

 hand end of the figure represents the condition with one colour 

 field alone exposed, and, as indicated below the larger figure, 

 the brightness of this field is progressively decreased 

 while the other field is increased in such manner that their 

 sum remains constant. All points below the two critical 

 frequency curves are in the region of brightness flicker. The 

 curve which is drawn convex upward, with its maximum at 

 the equal brightness point, is taken from the first theoretical 

 paper and is the upper limit to the colour flicker region, or 

 region in which the hue difference for successive phases of 

 the flicker photometer' is greater than the least perceptible. 



