﻿1166 Mr. J. W. T. Walsh on the 



In the American definitions * the first photometric magni- 

 tude defined is luminous flux, followed by luminous intensity, 

 illumination and brightness. 



In the British definitions j" the order is luminous intensity, 

 luminous flux, illumination and brightness. 



These two systems follow the respective customs of the 

 two countries as to the rating of illuminants. In Great 

 Britain a luminous source is rated in candles, while in 

 America the lumen is almost universally adopted. 



The existing official definitions, then, adopt as their 

 starting point the magnitude in which the unit is realized in 

 practice, for both luminous flux and luminous intensity are 

 characteristic of the ^source alone and are independent of 

 any other material body. 



It is true that both luminous flux and luminous intensity 

 are impossible of measurement per se, and it is not until the 

 source illuminates some surface that either the flux or the 

 intensity can be measured. It would thus appear that, in 

 'deciding on our first defined magnitude, we have to choose 

 between (a) the magnitude most intimately connected with 

 the material standard by which our unit is maintained, and 

 (A) the magnitude which is directly measurable. 



The authors of the paper referred to above apparently aim 

 at the second choice, but it is difficult to understand why, 

 after the irrefutable statement that 'Might measurements are 

 based on judgments of equality of brightness of photometric 

 surfaces," they go on without explanation to say that "the 

 fundamental photometric magnitude is illumination." 



In fact illumination appears to be the most abstruse of all 

 the photometric magnitudes and the one of which it is most 

 difficult to form a mental concept. It is not easy, for 

 example, to decide whether illumination exists in the absence 

 of a surface. 



Physical photometry being disregarded, it is clear that the 

 fundamental photometric magnitude from the point of view 

 of measurement is u brightness." Illumination follows as 

 that which causes surfaces to have brightness. Luminous 

 intensity and luminous flux then follow as before. 



In this connexion the proviso made by Dr. Campbell and 

 Mr. Dudding as to the avoidance of the Purkinje effect is 

 most interesting. On page 582 they say "But the second 

 law of addition is not true in all circumstances ; it is not 

 true, for example, when the Purkinje effect is apparent. 

 For, if U 1 and li 2 are red sources, B x and B 2 blue sources, 

 and if the illumination from Rj is equal to that from B 1? and 



* Ilium. Eno'. Soc. N.Y. Trans, xiii. (1918). 

 f Illuminating Engineer, London, xv. (1922). 



