﻿Quantum Theory of Photographic Exposure. 965 



of this kind can well be tested experimentally. In fact, if 

 that assumption be correct, then the light traversing two 

 or more equal photographic plates piled upon each other 

 should contain, successively, a larger percentage of the 

 coarser light darts, so that the formula? of type (12) repre- 

 senting the number of affected grains or clumps of various 

 sizes should have not only a decreasing n, but also a succes- 

 sively increasing average value a of the cross-sections of the 

 darts, a comparatively larger proportion of the more slender 

 darts being absorbed each time. In short, we should have a 

 kind of sifting effect. Such experiments which, to be at all 

 convincing, require obviously a much higher degree of 

 accuracy in counts and area measurements, are now in 

 preparation. Their results will be published in a subsequent 

 paper. In the meantime, the parameter a may and profit- 

 ably will be retained as a small but desirable correction of 

 the exponential formula without, however, being given either 

 of the alternative interpretations. 



It may be well to add here also a few remarks about n, the 

 chief parameter in the fundamental formula. This was 

 originally defined as the number of light-quanta or darts 

 thrown upon the photographic plate per unit of its area. 

 Now, apart from the generally small correction term con- 

 taining a, the parameter n appears in the formula only 

 through the product 



p=na, 



where a is the area of the grain. Thus, essentially only the 

 value of tliis product (a pure number) can be determined 

 from microphotographic experiments. Suppose now that 

 the sizes of all grains of the given emulsion were reduced in 

 the same ratio, converting every a into ea ; then, the same 

 experimental value of p would indicate a number of light 



darts - times larger. Now, such would exactly be the 



position if for every grain not the whole but only a fraction 

 e of the area were vulnerable, i. e. deprived of an electron on 

 being hit by a light dart. The grain may be sensitive only 

 in spots scattered over its area, and each perhaps of very 

 minute dimensions. Provided that all these spots occupy a 

 fixed fraction e of the total area of the grain, the micro- 

 photographic counts and measurements could not inform us 

 about the value of this fraction unless the exposure given to 

 the plate is known in absolute energy measure. Thus, for 

 instance, if, as was tacitly assumed, 6=1, the number of light 

 daris in the set (I.) of observations just described would lead 



