Temperature and Molecular Attraction. 107 



holds good. Some few people have been misled into thinking 

 that Clerk Maxwell's investigation of the distribution of 

 velocity in a system of moving particles applies to a system 

 under the influence of attractive forces, but such is not the 

 case. The mistake has arisen because the investigation is 

 sometimes quoted with no very clear statement of the funda- 

 mental suppositions of the mathematical investigation, and 

 because the investigation has been shown to apply to compo- 

 site molecules if there is no residual attraction beticeen the 

 molecules. But if there is molecular attraction, Clerk Maxwell's 

 investigation will not apply. 



It is worthy of note that if, after the addition of energy 

 amounting to E*, the new orbit assumed be spherical, the 

 new velocity will, as stated, be less than it was before by E#. 

 Now if, by any arrangement, the body could be held in its 

 new orbit, and its temperature (velocity squared) increased 

 to Ea greater than its original temperature (velocity squared), 

 then the total energy necessary for this purpose would be 

 3Ea:. I mention this because it might have some connexion 

 with the fact that the total energy required to bring most 

 metals into the molten condition at their melting point is 

 very approximately three times the energy that would be 

 necessary to raise a perfect monatomic gas of the same 

 molecular weight to the same temperature. 



19. The absurdity of trying to define temperature as the 

 kinetic energy of the particles in a system under the influence 

 of attractive forces can be seen if the chemical reaction 

 H 2 + = H 2 is considered. I have shown* that the 16 

 grams of oxygen and the 2*016 grams of hydrogen taking- 

 part in the above reaction, together at — 273° 0. occupy 

 34*78 c.c.j and when occupying this volume at that tempera- 

 ture they contain 67,300 calories more of energy than does 

 the 18*016 grams of ice which they form if united, and which 

 occupies a volume of 19*21 c.c. 



There is no supposition whatever connected with this fact, 

 except the slight uncertainty due to the measurements and 

 their extrapolation, and this uncertainty is probably not 

 greater than 100 calories. The total energy possessed by 

 the hydrogen and oxygen at —273° 0. is probably very much 

 more than 67,300 calories, and may be many times that 

 amount, for we have no evidence whatever as to the energy 

 which the ice possesses at that temperature. We only know 

 that the energy of the ice is 67,300 calories less than the 

 amount possessed by the equivalent hydrogen and oxygen 

 before the combination. The ice may, and in our opinion 

 * Trans. Amer. Electrochemical Society, xiv. p. 35 (1908). 



