LORD KELVIN 305 



for every gas a special figure for this molecular volume and 

 for the molecular forces. The molecular forces are, in 

 agreement with Laplace's theory of capillarity, very small in 

 the case of gases, since the gas molecules when in motion are 

 at an average distance apart much greater than the range of 

 action of these forces; but they do not entirely vanish. 



The existence of small molecular forces in the case of 

 gases could already be concluded, long before this time, from 

 the remarkable investigation of Joule and Thomson in the 

 years 1 852-1 862, which we have already mentioned. The 

 investigation showed that cooling took place when a gas ex- 

 panded without doing external work. According to the 

 energy principle, such cooling could only take place when 

 work was done in some way within the gas itself by the in- 

 crease in volume, which work would then result in a cor- 

 responding disappearance of heat. This would be the case 

 if attractive forces between the molecules existed. The ex- 

 periments carried out in 1 806 by Gay-Lussac had not shown 

 any cooling; this cooling, and hence the effect of the mole- 

 cular forces, could thus in any case only be very small. 

 Joule and Thomson finally succeeded, after overcoming 

 many difficulties, in actually demonstrating a small decrease 

 in temperature (a few tenths of a degree), by pumping the 

 gas in a circle, and arranging that it should expand sud- 

 denly at a constricted part of the pipe line. External work 

 was not done by the gas, since its total volume in the circuit 

 remained unaltered; nevertheless, the small fall of tempera- 

 ture already mentioned was found to take place at the con- 

 striction. The presence of molecular forces, but also their 

 smallness in the case of gases where the molecules are far 

 apart, was thus proved, in full accordance with the small de- 

 partures from the laws of Boyle and Mariotte. 



When so much information had now been obtained con- 

 cerning gases, the ground was well prepared for an attempt 

 to liquefy oxygen, nitrogen, and also atmospheric air; this 



Ws 



