50 THE ROYAL SOCIETY OF CANADA 



— 140-2° and 141 —0° respectively. The values of the critical pressure 

 are 37 «8 and 40 «7 atmospheres respectively. The averages are 



— 140-6°C. and 39-2 atmospheres, which is nearer the correct values 

 than those given by the earlier observers. 



A-fter the position of best agreement between the air and carbon 

 dioxide isotherms was found, the curves were redrawn in the approxi- 

 mately correct position. Figure 1 shows the result. It will be seen 

 that, except in the critical region, the results are very satisfactory. 

 The curves for temperature below the critical temperature are not 

 drawn, as the agreement between a simple substance and a mixture 

 is bound to be poor. In the two phase state of a mixture, the pressure 

 is not independent of the volume and the distortion of the curves 

 persists into the single phase state of liquid or vapor at these tem- 

 peratures. 



The same method was attempted using the data for argon given 

 by Onnes and Crommelin,^ but the agreement between the two families 

 of isotherms is not good, owing, no doubt, to the fact that the critical 

 coefficient is much lower, viz. 3-28.^ 



1 Onnes and Crommelin, Communications, Leiden, 118 (b), 1910. 



2 Onnes and Crommelin, Communications, Leiden, 121 (b), 1911. 



