30 SCIENCE PROGRESS 



would hold, whereas with real gases the states at which zero 

 values for the Joule-Kelvin effect are found only occur under 

 certain definite relations between pressure and temperature 

 for each gas (see curve d, fig. i, p. 38). 



Some difficulty is found in the application of equation (3) 

 to experimental results, as it is strictly only derived for 

 infinitesimal changes of temperature, and the total energy (E) 

 is supposed to remain constant. This makes its employment 

 to reduce experiments, in which the changes of temperature and 

 pressure are not very small, a somewhat difficult task, which is 

 also increased by the difficulty of excluding other effects which 

 tend to mask the one sought, and which sometimes allow a 

 considerable fall in pressure to take place with no change of 

 temperature. 



A gas which at the same time obeys the equation of state (2) 

 and which exhibits no Joule-Kelvin effect may strictly be called 

 a perfect gas, but, as pointed out, a gas may obey one without 

 necessarily obeying the other, at least over a certain range. 



Experimental investigation at even moderate accuracies soon 

 showed that gases obeyed these laws to a greater or less degree, 

 and it was noted that the greatest deviations were found with 

 gases such as carbon dioxide, sulphur dioxide, and ethylene, 

 which are comparatively easily liquefied. With the class which 

 Faraday called the " permanent gases " because he was unable 

 to liquefy them, such as nitrogen, oxygen, and hydrogen, the 

 deviations are much smaller. Still greater deviations are found 

 with vapours of liquids such as water, etc., just above their 

 boiling points. The extended kinetic theory as applied to real 

 substances takes cognisance of both the size of the molecules 

 and their attraction to one another, but has not been made 

 to include as yet the internal energy of the molecule and the 

 way in which this changes with temperature and pressure. It is 

 clear that the molecules must have something of the nature of real 

 extension, as shown by the increasing difficulty of compression, 

 as certain limits are approached, and by such phenomena as 

 effusion, and, on the other hand, a real molecular attraction as 

 shown in such phenomena as capillarity. Also these character- 

 istics are even more marked in the solid state. Molecules 

 are known from observations of the density of gases to consist 

 in most cases of two or more separate and distinct atoms, 

 among which there must be a certain amount of internal energy 



