TEMPERATURE AND THE PROPERTIES OF GASES 41 



under 5 when the density is taken as vanishingly small, and in 

 this case it is not necessary to make any assumptions about the 

 value of X, so it is probably not very far from the truth. It is 

 hoped that a more detailed consideration of this relation will 

 be published shortly elsewhere ; but the subject is painfully 

 lacking in data, those of Thompson and Joule made in 1854 

 being almost the only series available, although there are a few 

 other measurements by Olszewski, as mentioned above, and 

 others which are more or less capable of mathematical treat- 

 ment over a small range. 



It would be exceedingly important for the whole gas theory 

 to have a series of accurate measurements on one or more 

 gases for considerable ranges of temperature and determining 

 not only the sign but the value of the Joule-Kelvin effect, as a 

 function of initial temperature and of initial density. 



One of the most important applications of the study of the 

 isothermals of gases is in the corrections to be applied to the 

 gas thermometer to give temperatures on the absolute scale. 

 This involves two problems — the evaluation of the difference 

 between the centigrade and Kelvin scales, which depends partly 

 on strictly thermodynamic reasoning and partly on the deduc- 

 tions to be drawn from the properties of various gases. For 

 ordinary thermometric purposes, however, it is more important 

 to know the point-to-point differences between the scales of 

 any given gas used for thermometric purposes and the absolute 

 scale, that is, the correction which must be applied to the 

 temperature as read by the thermometer to get the real temper- 

 ature at any point of the scale. 



Until helium became known and reasonably obtainable, 

 standard thermometry may be said to have been confined to 

 the use of two gases, as no one gas is practically available over 

 the whole range of temperatures measurable by the gas 

 thermometer. 



For temperatures from ioo° C. upwards to the highest point 

 which the reservoir will stand, nitrogen is still the most suitable 

 gas, as the corrections are comparatively small ; it does not 

 penetrate the walls of the reservoir like hydrogen, or still more 

 helium, nor attack mercury like oxygen at high temperatures. 

 There is every reason to suppose that argon will be a still 

 more suitable gas when its thermodynamic properties are 

 sufficiently well known. For temperatures below ioq° C- 



