400 

A study of these small spherical grains by R. B. 
Wilsey, however, using the methods of X-ray crystal 
analysis, shows that even the smallest grains are still 
definitely crystalline and have the same structure 
as the large grains, the crystalline form being a cubic 
lattice. ‘ 
So long ago as 1915 it was realised that the dis- 
tribution of the different sizes of grains in an emulsion 
might play a very important part in determining the 
characteristics of that emulsion. The problem was 
to measure the distribution of -the grains; that is, 
the number of grains of a given size which occurred 
in an emulsion and the variation of the number with 
the size of the grain. This problem has often arisen 
in scientific work. It has been studied in connexion 
with suspensions of all kinds. Various indirect 
methods of attacking the problem have been suggested. 
It is possible to get determinations by settling the 
emulsion, taking advantage of the fact that the larger 
particles will settle most rapidly according to Stokes’s 
law, but the direct method is, clearly, to spread out 


a thin layer of the emulsion and to count the different 
sizes of grains occurring in it. Trivelli photomicro- 
graphed a thinly coated emulsion at an enlargement 
of 2500 diameters, enlarged the negatives to 10,000 
diameters, outlined all the grains of these enlargements, 
and then planimetered the grains and obtained tables 
showing the areas of the different grains present, at 
least a thousand grains being counted for each emulsion. 
Sheppard and Wightman obtained the same results 
by the use of the camera lucida instead of photo- 
micrography. From these tables curves were obtained 
showing the relation between the size of grains and 
the number present for several standard emulsions. 
Fig. 2 shows the results for the portrait film and slow- 
lantern emulsions. It will be seen that the curve 
shows a distribution of sizes of grain which corresponds 
approximately to a probability curve, the maximum 
number of particles being of a diameter of approxi- 
mately o-5 #, the particles both smaller and larger 
than this being fewer, until we have very few par- 
ticles indeed of larger size than 2-7 » and also few of 
smaller size than 0-2. On this small side no par- 
ticles can be measured less than o-2 », because this is 
the limit of the resolving power of the microscope. 
It is probable, however, that curves showing 
diameters will not be of real value, because the con- 
NO. 2786, VOL. 111 | 
NaTURE 

[MaRcH 24, 1923 
trolling factor will not be the diameter of the particles, 
but their projective area, as shown in Fig. 3. 
Svedberg investigated systematically the relation 
between the size and sensitiveness of grains in photo- 
graphic emulsions. He prepared emulsions so thinly 
© SIZE — FREQUENCY CURVE 


AS STANDARD SLOW LANTERN SLIDE 
© = PAR SPEED PORTRAIT Fiin 
FREQUENCY PEA 1000 GRAINS 
SS oe 
Fic. 2. 
coated that the grains were all in single layers, and 
counted the grains of different sizes by classifying 
them into four classes. The emulsions were then 
exposed and developed, and the developed silver 
removed, the remaining grains, representing those 
which had not been made developable by the action 
of light, being counted. In this way curves could be 
obtained showing the sensitiveness of the grains of 
each class, and it was found, as might be expected, 
that the larger grains were much more sensitive than 
the smaller grains. 
Svedberg next assumed that the product of the 
light action in the halide grain—that is, the substance 
of the latent image—consists of small centres distributed 
SIZE—AREA CURVE 
ASSTANDARD SLOW LANTERM SLIDE 
B=PAR SPEED PORTRAIT FILM 
PER SO.cM, 
TOTAL AREA Im yf? x 10 
.’ ree 
Fic, 3. 
through the grain or through the light-affected part of 
the grain, and that these centres are distributed 
according to the laws of chance. Ifa plate be developed 
for a very short time the grains show these centres 
as small black spots upon them. This was shown by 
Hodgson as early as 1917 (Fig. 4). 
Not only did Svedberg demonstrate the existence 
of these centres, but he also made plain their relation to 
the silver bromide grains by photographing the grains 


