Kohlenberg—Osmosis and Osmotic Pressure. 267 
would be if they were proportional to the absolute temperature 
that additional comment seems quite superfluous. The osmotic 
pressures here investigated, then, do not follow the gas laws at 
all. Again, the electrolyte LiCl yields a much lower pressure 
than the non-electrolyte, sugar, which is exactly the opposite 
of what the theory of electrolytic dissociation requires. 
OBSERVATIONS CONCERNING DIALYSIS. 
When both cane sugar and copper oleate are dissolved to¬ 
gether in pyridine and the solution is separated from pure pyr¬ 
idine by means of a vulcanized caoutchouc membrane, the cop¬ 
per oleate passes through the septum and the sugar remains be¬ 
hind. This is what one would expect from what has been said 
above. But here we have a case where a crystalline body, the 
sugar, is separated from a non-crystalline substance, the cop¬ 
per oleate, by dialysis in which process the non-crystalline or 
colloid body passes through the septum and the crystalloid re¬ 
mains behind. Again, when camphor and sugar are together 
dissolved in pyridine and the solution is separated from the solv¬ 
ent by means of a rubber septum, the camphor passes through 
and the sugar remains behind, which might have been antici¬ 
pated. Here we have a case in which crystalloids are separated 
from each other by dialysis. 
In fact it is clear that in general any substance which is sol¬ 
uble in both hydrocarbons and in pyridine may be separated 
from sugar by dialysis, when the pyridine solution of it and 
the sugar is separated from the pure solvent by a rubber sep¬ 
tum. In such cases the sugar will always remain behind in 
the dialyser and the other substance will pass through whether 
it be crystalline or non-crystalline. Any other substance sol¬ 
uble in pyridine but not soluble in hydrocarbons may, in gen¬ 
eral, be substituted for sugar. Thus the role of the nature 
of the membrane in the process of dialysis is demonstrated. 
The current view that crystalloids always pass through mem¬ 
branes more readily than colloids is evidently untenable, for 
it has been shown that just the opposite may occur, and that 
even crystalloids may be separated from each other by dialysis 
