In various models designed to imitate living cells the surface of the protoplasm is represented by guaiacol which acts in some respects like certain protoplasmic surfaces. The behavior of water in these models presents interesting features and if these occur in vivo, as appears possible, they may help to explain some of the puzzling aspects of water relations in the living organism.
When sufficient trichloroacetic acid is added to a two-phase system of water and guaiacol the two phases fuse into one. The effect of the acid is due to its attraction for water and for guaiacol. This is shown by the following facts.
During the addition of the acid the mole fraction of water in the guaiacol phase increases but the activity of water in the guaiacol phase falls off. The activity coefficient of water may fall to less than one twelfth the value it had before acid was added.
The behavior of guaiacol presents a similar picture. During the addition of acid the mole fraction of guaiacol in the aqueous phase increases but the activity of the guaiacol in the aqueous phase presumably decreases. Its activity coefficient calculated on this basis may fall to about one ninth of the value it had before the acid was added.
Somewhat similar results are obtained when acetone is substituted for trichloroacetic acid or when ethanol is substituted for trichloroacetic acid and ethylene chloride for guaiacol.
As trichloroacetic acid increases the mutual solubility of guaiacol and water we find that guaiacol saturated with water and having a high vapor pressure of water can take up water from an aqueous solution of trichloroacetic acid with a low vapor pressure of water: acid passes from the aqueous to the guaiacol phase, thus raising the vapor pressure of water in the aqueous phase and lowering it in the guaiacol phase.
Diffusion experiments present some interesting features. When an aqueous solution, A, of trichloroacetic acid is separated by a layer of guaiacol, B, from distilled water, C, under certain conditions water moves from A to C. This depends on the fact that acid moves in the same direction and appears to carry water with it. Similar but less striking results were obtained with acetone diffusing through guaiacol and with ethanol diffusing through ethylene chloride.
These phenomena differ from "anomalous osmosis" through solid membranes if it depends, as many suppose, on the diffusion of electrolytes through pores. We therefore suggest the term "anaphoresis" for the phenomena described here.
Measurements of the mutual solubilities of water, guaiacol, and trichloroacetic acid and of water, guaiacol, and acetone are given and are discussed in relation to the diffusion experiments. To give a complete picture of the process of diffusion we need to know the activities and concentrations in all parts of the system. The difficulties of achieving this are obvious.
The solubility relations are such that a concentration gradient of trichloroacetic acid in guaiacol produces a concentration gradient of water in the same direction, but the activity gradient of water is in the opposite direction.
Since in certain respects guaiacol acts like some protoplasmic surfaces it seems possible that similar phenomena may occur in living cells. If so these results have an obvious bearing on the movement of water in the organism and on methods of studying permeability. It becomes necessary to know to what extent a substance entering or leaving the cell appears to carry water with it in the manner here indicated.
In certain of the diffusion experiments the water takes a circular path, passing out of the dilute solution at one point and back into it (as vapor) at another. This circular path recalls the situation in the kidney where the water continually passes out of the blood into the glomerulus and tubule and then back into the blood from the tubule (where the solution is more concentrated). In both cases the circular path of the water is an essential feature.