320 
CHARLES A. SHULL AND S. P. SHULL 
part of the system to another under the conditions could take place only if 
the colloidal matter had a higher vapor pressure than the liquid water; 
and as the water evaporated from the colloidal mass, it would have to 
condense at the surface of the liquid water. Bancroft's explanation of the 
higher vapor pressure is based upon the shape and size of the water droplets 
in the gelatin. They are assumed to be round, and it is held that water 
evaporates more readily from a curved surface than from a flat one. The 
droplets of water are so minute that the curvature of the surface of each 
droplet is quite sharp; this results in a vapor pressure higher in the colloid, 
and a consequent distillation of water from it to the liquid phase of the 
system. The distillation should go on until in some way or other the vapor 
pressure throughout the system reaches equilibrium. 
Being engaged in the study of the relation of certain colloidal organic 
substances to water, we have had occasion to perform some experiments 
which led to a repetition of some of von Schroder's work. However, it 
was not possible to tell from von Schroder's discussion just how he set up 
his experiments, how he controlled the temperature, and how he secured 
and maintained saturation. In a number of ways the discussion leaves 
one in the dark, and we were compelled therefore to work out a method of 
investigation which no doubt differs in a number of ways from von 
Schroder's. The methods employed are briefly stated. 
Materials and Methods 
The gelatin used in the experiments to be recorded here was the Gold 
Label gelatin commonly handled by dealers. No attempts were made to 
purify it in any way; for, although pure substances are to be preferred in 
original work of any kind, we felt justified in using the gelatin in its com- 
mercial condition because the experiments we were trying to repeat had 
been carried on with unpurified gelatin. There is no doubt that von 
Schroder's gelatin as well as ours contained acids, and that in both cases 
salts were present. We could have neutralized our gelatin with some 
inorganic base, with subsequent dialysis until it was salt-free. This would 
have given us a purer gelatin, but we could not then have made a direct 
comparison between our own and von Schroder's results. It seemed to us 
best therefore to repeat the experiments with ordinary gelatin, as the 
differences between the two substances would likely be less than if we used 
any purification method to remove acids and salts. 
During the earlier tests the gelatin was prepared by dissolving it at 
gentle heat in distilled water, after which it was poured into petri dishes 
which had been slightly smeared with glycerol to prevent the gelatin from 
sticking to the glass; but as the glycerol is hygroscopic, this method was 
abandoned to avoid contamination of the gelatin with glycerol. Mercury 
was used as an agent to prevent the gelatin from sticking to the glass during 
drying. In this way sheets of gelatin of desired thickness could be secured, 
