BRIDGMAN. — THERMODYNMIIC PROPERTIES OF LIQUIDS. 109 



Accordingly, in the previous discussion no use has been made of this 

 "internal pressure." 



The usual kinetic conception of temperature must undergo modifi- 

 cation at high pressures just as our kinetic conceptions of pressure. 

 We think of the temperature of a gas as proportional to the average 

 kinetic energy of translation of its molecules, the translational energy 

 being the energy of the center of mass during free flight. Now it is 

 inconceivable that at infinite pressure there should be any free flight, 

 that is, there can be no kinetic energy of the molecule as a whole, 

 but it is also inconceivable that at infinite pressures a substance 

 should not possess temperature and be capable of temperature equi- 

 librium with surrounding objects. Our model of the molecule may 

 be helpful to us again. When the volume decreases beyond a certain 

 limit, we saw that the boundaries of the molecule become fixed in 

 position, and that the temperature is represented by the energy of 

 internal agitation. This suggests that temperature changes in 

 character from an affair of the molecule as a whole at low pressures 

 to an affair of agitation within the molecule at high pressures. The 

 behavior of the specific heats of gases and solids also strongly suggests 

 the same thing. We compute the specific heat at constant volume 

 of a gas by supposing that the kinetic energy of translation of each 

 molecule must be increased by a fixed amount to produce a rise of 

 temperature of one degree. But for most solid elements the law of 

 Dulong and Petit holds, which is equivalent to the statement that to 

 increase the temperature of a solid by one degree we must increase 

 the energy of each atom by a fixed amount. Now a microscopic 

 analysis of a solid like iron discloses a crystalline structure of great 

 complexity; we find it hard to think that there are not groups of 

 associated molecules and that the molecules are monatomic. It 

 appears, then, that temperature has become connected somehow with 

 what is going on in the atom. It view of this it would seem that 

 another conception of temperature is desirable. It must be such a 

 conception as not to be at variance with what we suppose to happen 

 in a gas or a solid. 



The material for such a conception is at land. It is a property 

 common to the temperature energy of a gas molecule and to the 

 internal energy of our suppositious molecule, that it is constantly 

 in a state of flux, changing during collision from kinetic to potential 

 and back to kinetic again. A natural generalization, then, is as 

 follows : Temperature is proportional to that part of the energy associ- 



