( 734 ) 
quantity by MRT/v, and so his paper cannot give any elucidation on 
the point which requires it most. But that notwithstanding we owe 
to Lorentz’s labour a considerable widening of our views, will as I 
hope, appear from the continuation of this Communication. 
Also BonrzMaNn’s paper leaves us in the dark as to the question 
why the quantity 6, which in other cases plays such an important 
part for liquids, seems to have no influence on the value of the 
osmotic pressure. In the equations, which he draws up, he never 
takes the size of the molecules into account *) and it does not appear 
why he does not do so. Further he stops at the result, that the 
osmotic pressure is equal to the sum of the pressures exercised by 
the two kinds of molecules, without discussing the part played by 
the different kinds. For these reasons I cannot see a satisfactory 
solution of our problem in BOLTZMANN's paper either. 
§ 4. To arrive at a solution it seems in the first place necessary 
to give three definitions. 
dst. Given a fluid. Placed in it a body of perfect elastic impermeable 
substance, which does not exert any attraction on the molecules 
of the fluid. The thickness of this body (or this surface) be infinitely 
small; let us suppose it to have an area of 1 em’. The “kinetic 
pressure” in that fluid is then the quantity of motion in unity of 
time transferred by the molecules of the fluid to this body (or obtained 
in the elastic collisions from this body). 
2nd, In the second place I imagine a body’), which is distinguished 
1) See speciaily 1. c. 475 equation (4), which is evidently meorrect, when part of 
the cylindre is not open to the centres of the molecules, because it is occupied by 
distance spheres of other molecules. 
2) That I assume that the body does not attract the molecules of the fluid, is 
for simplicity’s sake, but it is not essential. If we imagine a wall, which does 
attract the fluid, more molecules will reach its surface (cf. the footnote p. 739) 
and hence will impart a greater quantity of motion to the wall. But on the other 
hand the particles of the surface will now be drawn into the fluid with an} equally 
greater force. The elastic displacement of the particles of the surface of the solid 
wall, and with it (with sufficient elasticity) that of the layers lying under it, in 
other words the pressure which propagates in the solid body, and which would 
be measured with a manometer of any kind, will be perfectly the same in the 
two cases. If we wish to take also negative external pressures into account, we 
shall even have to give the definition by means of an attracting body, because 
in this case a non-altracting body would not even be reached by the molecules 
of the fluid. (Cf. the well-known fact that for the observation of the negative 
pressure slrongly adhering walls are required). In this case the impulse of the 
attraction of the molecules is simply greater than the quantity of motion which 
they impart to the wall (and which may still be very great), the elastic displacement 
is therefore not from the fluid, but towards it. 
Also in the case that we wish to take capillary layers into account, our definition 
