1228 
NH(C,H,), pr = 28 atm. for 31 atm, all deviations vanish. And suca 
diminutions of the critical pressures in later, more accurate experi- 
ments with purer substances are by no means rare. History even 
knows more considerable diminutions. Thus e.g. for CH,Cl the old 
and the new values are resp. = 73.0 and 65.9 atm.; a difference 
of almost 11 °/. In view of such facts it would be rash to attach 
too much weight to the deviations found for some amines. The 
more so as the agreement for the many other substances in the 
foregoing tables is almost perfect, or the deviations do not amount 
to more than 1 or 2°/, in the extreme cases. 
$ 4. The additive properties of the values of //a. 
As we have seen, the values of 6 can be built up additively from 
a small number of fundamental values — those of the elements —; 
such an additivity is not found, however, with regard to a, though 
it is with regard to Va. 
This is quite natural. For « is always composed of the product 
(or the sum of some products) of two values, one of which always 
refers to the first of two molecules that attract each other, the 
other to the second molecule. 
Thus e.g. when the “attracting mass” of a helium molecule is u, 
the total cohesion will be a = Cu’, when Cis a certain factor 
of proportionality, in which also the summation with respect to all 
the molecules is included. (This summation is the same for all 
substances, because a refers to the same volume v,, and the 
molecules lie equally far apart therefore). If the attracting mass of 
a Clatom = u, the cohesion of Cl, will be represented by a= C. 4u’. 
Lastly, if wu, is the attracting mass of H, u, of Cl, the quantity a 
for HCI will be represented by a= C (u,?+2u,u,+u,’) = C(u,-+u,)’. 
Accordingly, the cohesion is not supposed “specifie” (chemical) 
— so that e.g. the attraction of an atom H in another molecule 
being given by u, u, =u,°, the attraction between H and an atom 
Cl (likewise in another molecule) is determined by uw, X u,; ie. 
the attracting mass of H will remain the same, viz. 1, independent 
of the fact whether H attracts a second H in another molecule, or 
whether it attracts an atom Cl. 
Hence we assume — and this assumption is perfectly confirmed 
and justified by the found additivity of a — that the cohesion is 
of entirely physical nature, only depending on certain not yet 
sufficiently known circumstances concerning the number, mass, 
velocity, path of the different electrons, of which the atoms are built up. 
We may add that the above considerations come to this, that the 
