276 DIFFUSION, OSMOSIS, AND FIITRATION. 
exert surface action, so that the possibilities for purely physical absorption are 
quite unknown, and so-called vital elective action may be the result of specific 
adsorptive affinity. Hof meister : has shown that gelatin has an " elective " action, 
for common salt, the concentration of the solution imbibed exceeding that of 
the surrounding solution ; and, further, that the combination of sodic chloride 
with the gelatin favours the uptake of water. Again, gelatin takes lip more 
water from - 5 to 2 per cent, solution of ethyl-alcohol in water than from pure 
water. 
With salts that undergo electrolytic dissociation in solution, permeability 
must be a function of ions. Thus, according to Ostwald, 2 copper ferrocyanide 
is permeable to potassium chloride, because both chlorine and potassium ions can 
pass ; it is impermeable to barium chloride, because the barium ion is stopped ; and 
impermeable to potassium sulphate, because the sulphuric acid ion cannot pass ; 
and, under ordinary circumstances, on account of opposite electrical charges, if 
one ion is stopped, so must be the otber. There are, however, conditions 
under which an ion, stopped on account of the impermeability of the membrane 
to its fellow in a salt, may pass the membrane. 
If the negative ion of a salt is prevented from passing through the 
membrane, only because it is impermeable to its positive fellow, the addition of 
another salt, whose positive ion can pass the membrane, will allow the negative 
ion of the first salt to pass in company with it. Or a salt whose negative ion 
can pass the membrane may be placed on the opposite side, the two negatives 
exchanging with their positive fellows across the membrane, and equal numbers 
of the two negative ions passing in opposite directions in a given time. This is 
of interest to the physiologist, since it opens a possible physical explanation of 
the fact that a cell may hold back a substance under certain conditions, while 
under others, when surrounded by a differently constituted fluid, the same 
substance may be given up. 
Koeppe 3 has attempted to apply this to the formation of hydrochloric acid 
in the stomach from sodium chloride, maintaining that the stomach wall is 
impermeable to chlorine ions, but that the sodium ions are exchanged for 
hydrogen ions from the blood. That free hydrogen ions are present in the 
alkaline blood is, however, hardly possible. 
Whether permeability be a function of physical or chemical nature, it 
is obvious that in the case of a living membrane the complex to which 
the term " physiological condition " is ap] (lied must affect the property, 
so that one and the same membrane in the body may, under different 
circumstances, be more or less permeable by the same substance. 
The simplest living membrane with which experiments can be made 
is probably the differentiated outer layer of the protoplast of the vege- 
table cell (Plasmahaut). There is no doubt that the permeability of this 
membrane for different chemical substances is very variable. It is pene- 
trated by some dye-stuff's but not by others, very impermeable to 
many simple salts, though easily permeable by certain complex organic 
substances. 4 Since this membrane is in its living condition so 
slightly permeable to salts, the osmotic pressure within vegetable cells is 
high (3 to 4 atmospheres). This special relative impermeability to salts 
is obviously regulated in some manner by the " physiological condition " 
of the membrane. Jansen 5 found that the cell sap of the alga, Chato- 
1 Arch. f. exper. Path, n. Pharmacol., Leipzig, 1891, Bd. xxviii. S. 210. 
2 Ztsehr. f. physikal. '''Jinn., Leipzig 1890, Bd. vi. S. 71. 
3 Arch, f. d. ges. Physiol., Bonn, 1896, Bd. lxii. S. 567. 
4 Pfeffer, Abhandl. d, math.-phys. Gl. d. k. sacks. Gesellsch. d. Wissensch., 1890, Bd. 
xvi. S. 149. 
5 Vcrhandl. d. k. Akad. v. Wctcnsch., Amsterdam, 1888, vol. iv. p. 345. 
