(735 ) 
from the just mentioned body only by its being very thick compared 
to the sphere of action of the molecules. The quantity of motion 
transferred by this body per unity of time to the molecules, is called 
the ‘external pressure” in that fluid. 
34. In the third place I place in the fluid (which I now suppose 
to be a mixture) a body, which is distinguished from that mentioned 
under 2 only by the fact that the molecules of one component (solvent) 
pass through it without any change in their velocity. 1 shall leave undis- 
cussed here whether such a body can actually occur. The pressure 
to which this body is now subjected, and which might be measured 
e. g. by the elastic displacement of the particles of its surface, I 
call the “osmotic” pressure in that solution. 
From these definitions it is already clear that in dilute solutions 
the osmotie pressure defined here must be of the order of the kinetic 
pressure exerted by the dissolved substance, and not of that of the 
a 
external pressure. For these two differ, in that — has disappeared 
7 
for the kinetic pressure, and this will also be the case for the osmotic 
pressure defined here, as appears from the reasoning given above 
($ 3). I shall further show, that in dilute solutions this osmotic 
pressure has the value indicated by the law of Van ’r Horr, and that 
in any ease it is as great as the well known experimentally intro- 
duced and measurable osmotic pressure, i. e. the difference in external 
pressure of solution and pure solvent under the pressure of its own 
vapour. in equilibrium through a semipermeable wall. 
calls for fuller discussion. First of all this applies to what we have just now 
said, for just as for negative pressures so also in the capillary layer, as 
Van per Waats has shown in his theory of capillarity, the attraction of the 
surrounding layers is a necessary condition for stable equilibrium. But further, 
as Hursnorr has shown (These Proc. 8, 432 and Diss. Amsterdam 1900), the 
above defined quantity does not obey the law of Pascan any more, because mea- 
sured in the direction of the layer and perpendicular to it, it has a different value. 
In this case we might perhaps speak of a total external pressure, which might 
be split into an external fluid pressure and an external elastic pressure. The 
consideration of capillary layers round a free floating sphere, teaches us further, 
that the “external” in the name “external pressure” must not be understood 
in such a way, as might easily be done, viz. that the reactive force of this 
pressure, as it prevails in a certain point, would act in points outside the system 
in question, which would always be more or less arbitrary, as we may choose 
the limits of our system arbitrarily. The assertion: the external pressure is in a 
point of the fluid so great, comes simply to this, that wen I should place a 
strange body at that place, without altering the condition more than necessary 
for this, this body would experience a pressure of such a value, and would 
suffer an elastic modification in form which corresponds to it, so differing in the 
capillary layer in different directions. 
