386 Drs. M. W. Travers and C. J. Fox. [June 26, 



" The Vapour Pressures of Liquid Oxygen on the Scale of the 

 Constant-volume Oxygen Thermometer filled at different 

 Initial Pressures." By MORRIS W. TRAVERS, D.Sc., Fellow 

 of University College, London, and CHARLES J. Fox, B.Sc., 

 Ph.D. Communicated by Sir WILLIAM RAMSAY, K.C.B., 

 F.E.S. Received June 26, 1903. 



The vapour pressures of liquid oxygen and liquid hydrogen on the 

 scales of the constant-volume hydrogen and helium thermometers has 

 recently been determined by one of us in conjunction with Dr. A. 

 Jaquerod and Mr. G. Senter, and it has been found that two scales 

 of temperature differ by amounts which increase as the temperature 

 falls. 



Vapour Hydrogen Helium 

 pressure. scale. scale. 



Liquid oxygen 760mm. 90'10 90'20 



Liquid hydrogen 760 20'22 20'41 



These results are in accordance with Callendar's calculations of the 

 deviation from the thermo-dynamic scale of measurements with ther- 

 mometers filled at an initial pressure of 1000 mm. of mercury at the 

 melting point of ice. Whether the deviation becomes smaller when 

 the thermometers are filled at a lower pressure has not been deter- 

 mined, and, indeed, with the means at our disposal it would be 

 practically impossible to do so. In order, therefore, to investigate 

 the variation of the readings of the gas thermometer with change of 

 initial pressure, we decided to measure the vapour pressures of liquid 

 oxygen on the scales of the constant-volume oxygen and nitrogen 

 thermometers, for which the deviations from the thermo-dynamic scale 

 are considerably greater. 



Previous investigations. The vapour pressure of liquid oxygen has 

 been determined by Dewar,* at temperatures near the boiling point, 

 by means of a constant-volume oxygen thermometer. In calculating 

 his results he employed the value 0*0036525 for the pressure coefficient 

 of oxygen. 



Jhe pressure coefficient of oxygen. Makower and Noble, t using the 

 apparatus and method of Travers and Jaquerod, have determined the 

 pressure coefficient of oxygen at pressures corresponding to 375 and 

 700 mm. of mercury. Their results lead to the conclusion that the 

 coefficient increases with rise of pressure, and if the variation can be 



* ' Roy. Soc. Proc.,' 1901, vol. 68, p. 44. 

 f ' Roy. Soc. Proc.,' p. 379 ante. 



