Principles underlying the Flicker Photometer. 65 



by this figure was invariably recognized. The effect of 

 altering the distance of the eye was always much more 

 marked in the case of the equality of brightness instrument. 

 The discussion of this result will be postponed till a later 

 stage in this paper. 



It may be noted that Wild * has also stated that in some 

 experiments undertaken by him, the effect of varying the 

 region of the retina upon the apparent brightness of two 

 beterochromatic surfaces was confirmed, at least qualitatively; 

 but no such effect could be traced to exist in the case of the 

 flicker photometer. 



Flicker Photometers and the Purkinje Effect. 



In the paper read before the Physical Society in 1906, the 

 author referred to the question whether or no flicker in- 

 struments were, like those of the equality of brightness type, 

 subject to the Purkinje effect. The results of a method 

 employed in some experiments described in that paper seemed 

 to suggest that this was the case, but this method (which 

 involved reliance on a colour change instead of flicker) was 

 admittedly open to objection. In these experiments, the 

 photometer was intentionally run at such a speed that all 

 trace of flicker disappeared, and the results must therefore be 

 supposed to refer to sensation of fused colours rather than 

 flicker. 



The chief difficulty in studying the real question at issue is, 

 as was then pointed out, that the flicker photometer cannot in 

 general be successfully used at the low order of illumination 

 necessary in order to exhibit the Purkinje effect in a marked 

 manner, especially when the retinal area employed is small 

 and central. Subsequently, however, the author has made a 

 series of experiments with the arrangement shown in fig. 2 

 which allows of a somewhat larger retinal area being used, and 

 therefore seems to give indications of the Purkinje effect at 

 a somewhat higher order of illumination. Even so the 

 indications available are not very distinct, and the results 

 obtained must be regarded as very approximate. 



It may again be pointed out, that the arrangement utilized 

 enables us to compare the course of the photometrical readings 

 with increasing illumination by means of both flicker and 

 equality of brightness sensations in a particularly satisfactory 

 manner, owing to the fact that the retinal areas employed in 



keach case can be kept practically the same : all that is 

 necessary is to take a series of readings by the equality of 

 * Electrician, Aug. 16, 1907, letter. 



