264 
DIFFUSION, OSMOSIS, AND FILTRATION. 
r is the ratio of the diffusion coefficients of the two salts, with 
separate diffusions. 
r is the ratio with simultaneous diffusions. 
it, the ratio of the amounts diffused of the same salt in separate 
and in simultaneous diffusion, i.e. the alteration of the coefficient of 
diffusion produced hy the presence of the other salt. 
Separate. 
Simultaneous. 
r. 
r . 
r' :r. 
Ji. 
fNaCl . 
1 N:l,,S0 4 
•5833 
•6054 
1 
1 
1-038 
•3770 
•2497 
•590 
■352 
•596 
•662 
(KC1 . 
•8560 
•9276 
1 
1 
1 -083 
1 Bad., . 
•8433 
•4424 
•572 
■401 
•701 
•814 
fNaCl . 
•7142 
•7883 
1 
1 
1-019 
IBaCl. . 
•5673 
■5225 
•757 
•66S 
•882 
•921 
/K o S0 4 . 
1 MgSO, 
•4745 
•202S 
■4378 
•1684 
1 
•382 
1 
•345 
•903 
•901 
•830 
/Xa.,S0 4 
\ MgS0 4 . 
•3757 
•2097 
•3420 
•1823 
1 
•523 
1 
•502 
•960 
•910 
•S69 
As a rule, as seen in R, the more diffusible salt is accelerated, the 
less diffusible delayed. In the two last pairs both members are delayed, 
but the less diffusible more markedly. 
In the body it is rare to find the conditions present for a free 
diffusion between the constituents of two solutions; a membrane, 
whether composed of cells or the surface layer of the protoplasm of 
a cell, as a rule intervenes, and obviously the permeability of the 
membrane affects the result. If pig's bladder separates methyl alcohol 
and ether, the methyl alcohol diffuses into the ether, but if a caoutchouc 
membrane separates the two liquids, the ether diffuses into the alcohol. 1 
< >SMOSIS. 
The term osmosis is applied to diffusion taking place between two 
liquids separated by a membrane. 
The simplest case of this is that in which a solution of a substance 
is separated from the pure solvent by a membrane permeable by the 
solvent but impermeable by the dissolved substance. Such membranes 
were first prepared by Traube, 2 in the form of colloidal precipitates, 
such as tannate of gelatin and ferrocyanide of copper, but Pfeffer 3 
was the first to thoroughly study the process of osmosis under such 
conditions. The name " semipermeable " has been given to such mem- 
branes, but it must be noted at once that this expression is seldom 
strictly accurate and must always be used relatively to some particular 
substance. Tamman 4 has pointed out that such membranes are by 
no means the "molecule sieves" that Traube imagined, 5 and in experi- 
mental work the membrane must be chosen to suit the substance, or 
vice versd. Copper ferrocyanide forms one of the best of such mem- 
branes, and is nearly impermeable to cane sugar. 
1 Kaoult, Ztschr. f. physihd. Chem., Leipzig, 1885, Bd. xvii. S. 
2 Arch.f. Anat. u. Physiol., Leipzig. 1867, S. 87 and 129. 
3 "Osmotische Untersuch.," Leipzig, 1877. 
4 Ztschr. /. pkysikal. Chem., Leipzig,' 1892, Bd. x. S. 25.".. 
5 See also Walden, ibid., 1892, Bd. x. S. 699. 
735. 
