8 Art. 10. - S. Kinoshita and H. Ikeuti : 



proportional to the mass of silver contained in a unit area of the 

 plate. Therefore, if it he assumed, as a first approximation, that 

 all the grains are of equal mass, the above equation may he 

 written in terms of .the number s of the grains per unit area in the 

 plate. 



.9 = .<?„(l-e— ), (2) 



where Sq is the value of s for the plate for which 7)=7)„ and would 

 be the total numl)er of the halide grains initially pre-ent in tlie 

 emulsion film per unit area. 



Consequently, the number of the falling « particles required 

 to increase one more developable grain will be 



^ = 1 . (3) 



ds c (s^—s) 



If we take the corresponding values of s and n in a plate of 

 very small density, .s is negligibly small compared with s,y In this 

 case, the above equation reduces to 



dn _ 1 

 ds csq 



Giving values experimentally found: 

 c= 1-18. 10"' and So=l-]6. 10' for a Wratten Instantaneous Plate, 

 and c='93.1U-'' and s„=-97.10' for a Wratten Ordinary Plate, 



——= '73 for the Instantaneous Plate, and 

 ds 



= I'l for the Ordinary Plate. 



On examining microscopically, it was found that the Instan- 

 taneous Plate of very high density was entirely covered with silver 

 grains, while the Ordinary Plate of maximum density was almost 

 covered with grains, but about one-tenth of the area was estimated 

 as allowing « particles to pass through without contact with the 

 grains, and that in an Instantaneous Plate of small density al)Out 

 one-third of the grains were overlapping others, while in an 

 Ordinary Plate the overlapping was practically negligible, so that 

 one a particle did not strike more than one halide grain. 



Taking these facts into consideration, it can be concluded 

 with certainty that the number of « particles required to change 

 one halide grain into the developable state is one, whenever it 



