( 609 ) 
§ 3. Deduction of the values of d, and d,. Now that d,—d, is 
known, the values of d, and d, may be reached by assuming that 
the Cattueter and Matitas') law of the rectilinear diameter holds 
for argon. Considering that the accuracy of the experiments with 
which we are at present concerned is not very great we may quite 
well base our calculations on this assumption, the more so as 
Marnras and KAMERLINGH Onnes’) have just shewn from very accurate 
measuremen!s that oxygen obeys this law. The applicability of this 
law to argon, therefore, is not necessarily called into question by 
the fact that the critical temperature of argon is much lower than 
is the case with substances for which the validity of the law of 
the rectilinear diameter has been experimentally established. 
To be able to apply this law, however, to our present purpose, 
we must know the data for the diameter with considerable accuracy. 
They can be dedused only very inaccurately from the liquid densities 
given by Baty and Donnan*) which were used along with the 
diameter in Comm. 115 for determining the critical density. Their 
measurements, moreover, cover a range of only 6°, viz. from —189° C. 
to —183°? C. far removed from the critical temperature, and a 
difference of 0.2°/, between the errors in their extreme observations 
causes an error of about 3°/, in the critical density. And further, a 
small percentage error in the critical density is greatly magnified in 
the vapour volumes e.g. at —- 130° C. about 4 times, at — 134°C. 
about 7 times and at — 140° C. about 13.5 times. When the values 
of d, and d, calculated from the estimate of d, (0,496) based on 
Baty and Donnan’s results, which was used in Comm. 115, and the 
points on the boundary curve deduced from them were accordingly 
plotted in the pv-diagram of the argon isotherms and along with them 
experimental values of the vapour volumes obtained by slight extra- 
polation of accurate isotherms, it was seen that they deviated consi- 
derably from each other. ji 
I tried therefore to obtain d; independent of Bary and Donnan 
with a view to finding the constants of the diameter from their liquid 
densities regarded as a determination of d, + d, for a single tempe- 
rature far from the critical. This can be done with the help of the 
vapour pressures published in Comm. 115 and of the argon isotherms 
that are still to be published mentioned in § 1 of the present paper. 
At the eritical point the equation: 
!) L. CAILLETET and E. Maturas Journ. d. Phys. (2) 5, 549, 1886. 
2) Zitt.versl. Juni 1910. C. R. Ac. d. Se. Paris 151, 213 and 474, 1910. 
8) E. C. C. Baty and F. G. Donnan, Journ. Chem. Soc. 81, 911. 1902. 
