BIOLOGY: L. J. HENDERSON 
657 
consist in the diffusion through phases and across interfaces from points 
of higher to points of lower potential. In the internal phenomena no 
less than in the exchanges with the environment, and in the internal 
phenomena of a single phase no less than in the heterogeneous process, 
water is quite as much in question as the dissolved substances. 
This fact has been generally overlooked, by the physicists even more 
than by the physiologists. We may consider a simple diffusion experi- 
ment in which a layer of water is placed above a layer of sugar solution. 
It is true that the sugar must diffuse up into the water, but it is also 
true that the water must diffuse down into the solution. Yet the 
phenomena of diffusion have always been discussed with the help of 
the mathematical theory of heat conduction as exactly defined by Fou- 
rier. Thus Fick's theory of diffusion overlooks the role of the solvent. 
This has been possible because the process takes place as if the solvent 
were inert and the dissolved substance possessed a higher diffusibility. 
But the diffusion of the solvent is probably greater than that of the 
solute, in that water is, with a few exceptions, the most diffusible of 
substances. In their neglect of the relativity of motion, contemporary 
accounts of diffusion involve an old fallacy that occurs even in the 
astronomy of Ptolemy. 
It may perhaps be objected that throughout the organism water 
exists at a uniform potential. This, however, is both untrue and 
beside the point. For, if two phases are free to exchange material, an 
exchange involving all their components will take place unless, at the 
outset, all their components possess the same potential in both phases. 
Thus, as Hober^ has shown, isotonic solutions of sodium chloride and 
magnesium sulphate change in volume when brought into contact. 
For the chloride diffuses faster than the sulphate. Thus the sulphate 
solution becomes more concentrated and, as a result, water diffuses into 
it from the chloride solution. In like manner differences of pressure 
and of temperature influence the potential of water and of all dissolved 
substances. Finally, the processes of metabolism are continually 
altering the concentrations of solutions, and therefore the potential of 
water. 
When these facts are taken into account, it becomes clear that the 
chief physical factor in the internal regulation of volume is water, 
through its distribution between the infinite assemblage of phases which 
make up the organism. The general concepts of the phase rule reveal 
the several elements of this process — except those which depend upon 
so-called selective activities — but the physical theory of diffusion, and 
therefore the whole kinetic description, has been developed from a 
