nutting: absorption of light 



325 



sions is a well known experimental fact. P is of the order of 

 10 mg. of silver per square decimeter of film, per unit density. 

 The relation between P and size of grain has not yet received 

 experimental confirmation, tho P is known to vary in different 

 kinds of plate. 



In the following table are numerical values calculated for dif- 

 ferent photographic densities. 



TABLE r 



T is the fraction of the Ight transmitted by the entire thickness 

 of film, 1 — J. is the corresponding transmission of any single 

 layer 1 grain deep while A is the fraction absorbed, on an aver- 

 age, in any single layer. The final column, 10 A, is what the 

 absorption of a film 10 grains deep (an ordinary film) would be 

 if each grain were fully effective. Thus, for unit density, the 

 transparency is 10 per cent, the absorption of a single layer 20 

 per cent and there are present 2.1 times as many grains as would 

 completely cover a surface. 



Taking the particular case of unit density and Ai =20 per 

 cent (0.20) and computing the absorption of the first n layers 

 by the formula 



5i....n = 1 - n = 1 - (1-^)", 

 B^ = A = 0.20, 5:,2 = 2A - A.2 = 0.4 - 0.04 = 0.36 



that is, in the first two layers 36 per cent is covered and 4 per 

 cent of this covered twice. In the first three layers 



Bx..z = 3A - 3^2 + ^3 = 0.60 - 0.12 + 0.008 = 0.488 



