( 703 ) 
become = 6'), we have here a number of terms which approach 
infinity, but independent of each other. Only when we indicate 
exactly in what way we make » approach to 4 with decrease of 
temperature, we can find a definite value for »,-—1,, and only then 
the question put makes sense. 
The way in which NerNsr determines the maximum work, makes 
it impossible for him to see this. He makes all the reacting substances 
evaporate reversibly, then the reaction take place, and at last the 
reaction products condense reversibly. Everywhere he neglects the 
volumes of the condensed phase by the side of that of the gas phase. 
But as has already been observed by Dr. Scnrrrur, this is not 
allowed here, because the great terms cancel each other, and 
accordingly the difference of entropy will depend on the relation of 
the volumes in the condensed system. 
Therefore it does not appear, of course, from Nernst’s papers 
what way of approach is exactly meant in the “theorem of heat”, 
but as is demonstrated more fully in the footnote, on p. 702, it is 
impossible to assume anything else than that the approach to the 
absolute zero is meant in such a way that the substances before and 
after the reaction are under the pressure of their saturate vapour, or 
in other words that we reach the point 7==0, v= along the border 
line both for the reacting substances and for the reaction products. 
4. Then the result will depend on the way in which the volumes 
along the border line depend on the temperature. If we assume that 
v, —b,=a, f(T) and v, —b, =e, f(1), the sum of the first two 
terms becomes finite, and that of the last two infinite, because 
in general the specific heats of the reaction produets and of the 
reacting substances are certainly not the same for gas mixtures. 
0(4,—m,) ; 
i eee sl takes a finite value, viz. the difference of the specific 
” 
heats, and so the expression (2) does not become zero, but infinite. 
We can only evade this conclusion by making such a hypothesis 
about the dependence of v, — 6, and v, — 6, on the temperature, 
that this infinity just disappears, i.e. by putting: 
Le}, Eve, 
za yout 
f(T) and v, — b, = «,7 Ae 
2 2 
Ab) 
Apart from the fact that this brings us in collision with the law 
of the corresponding states‘), which must certainly hold in this case, 
1) For in corresponding states the c‚ do not become equal. Comp. VAN DER 
Waats—Kounstamm Lelirbuch der Thermodynamik p. 66. 
