( 245 ) 
which was almost exclusively restricted to the critical neighbourhood, 
and the system water-natrium sulphate presents so many remarkable 
peculiarities also outside this critical region, it was resolved to take 
in hand an extensive quantitative investigation of this system, the 
results of which will briefly be communicated here. 
The first part of this investigation dealt with the determination 
of the vapour pressure lines. 
Fig. 1 represents what was found schematically, the temperature 
and the pressure being accurately indicated at the most important 
points. 
One part of this figure has been reproduced in Fig. 2 on a larger 
scale in the real proportions and contains among others the three- 
phase-curves for 
Na,S0 4 .10aq-f L + G , Na, S0 4 + L + G , Na,S0 4 .10aq 
+ Na,S0 4 + G , Na,S0 4 .7aq + L + G , Na 2 S0 4 . 7 aq 
+ Na, S0 4 -f- G,, which have all been measured accurately to 0,1 mm. 
by means of a very simple apparatus with mercury manometer. 
From the situation of these lines, which had already been indicated 
schematically by van’t Hoff, 1 ) the situation of two quadruple points 
and D follows with pretty great certainty, which is perhaps best 
proved by this that the temperature of the point E agrees perfectly 
with the exceedingly accurate determination of the transition tem¬ 
perature by Richards and Wells 2 ), who found 32.384° for the 
transition point of the transformation Na, S0 4 . JOaq ^ Na, S0 4 -f aq 
under a pressure of 1 atm., which differs very little from that under 
the vapour tension pressure E. (See table p. 247). 
It follows in the first place from these tables that the three-phase 
roive for Na, S0 4 -J- L -|- G lies between the two others, which the 
| e °rv requires, and further that for 32.4° the three three-phase pressures 
" are the same value, viz.: 30.8 mm. Hg., from which follows that 
at this temperature and pressure Na 2 S0 4 .10 HO, Na, S0 4 , L and G 
are in equilibrium with each other, which may be considered as in 
Perfect harmony with Richards and Wells’ observation. 
we now apply van ’t Hoff’s formula to the table for the 
curve Na, S0 4 .10 aq -\- Na, S0 4 -j- G in the same form 
* used by Frowein ’), we get: 
■ -:> 
h Vorlesungen 1, 58. 
) Zeitschr. f. physik Chem. 43, 471, (1903). 
) Z. Ph. Chem. I, 8 (1887). 
17 
roceedings R oya l Acad. Amsterdam. Vol. XII. 
