LORD KELVIN 303 



means of pumps; but nevertheless, this process failed, even 

 with the highest obtainable pressures, in the case of a number 

 of gases, such as oxygen, nitrogen, and hydrogen. 



Faraday therefore took up the investigation again in 1 844; 

 now starting from observations published since the time of 

 his first experiment by Cagniard de la Tour, a French 

 engineer.^ In his experiments, liquids such as ethyl ether, 

 carbon bisulphide, and water, were heated in closed glass 

 tubes, and it was observed that they are completely trans- 

 formed into vapour above a certain temperature, although 

 the pressure, which was measured accurately in several 

 cases, becomes very high, and although the density of the 

 vapour as calculated from its volume cannot be much smaller 

 than that of the liquid. He determined for each liquid a 

 definite temperature, later called the critical temperature, 

 above which they did not remain liquid in spite of the 

 highest attainable increase of pressure; for example, this 

 temperature for ethyl ether was 187° C. 



Faraday realised that this must also be of importance as 

 regards the converse problem, namely the liquefaction of 

 gases, which would accordingly be more a question of suffi- 

 ciently low temperature than of the degree of pressure used. 

 He therefore made a large number of new experiments with 

 many gases, in which he made use of simple pumps, and a 

 freezing mixture of solid carbon dioxide and ether, which 

 at that time was already known, and the effect of which he 

 further increased by pumping off carbon dioxide, whereby a 

 temperature of about — 100° C was reached. ^ 



He thus succeeded in liquefying a further number of 

 gases; but oxygen, nitrogen, hydrogen, nitrous oxide, and 

 carbon monoxide still resisted liquefaction. Faraday also 

 investigated the vapour pressure of the various gases which he 



^Cagniard de la Tour (1777-1859) was also the inventor of the siren 

 which has become important for acoustical investigations, and is now also 

 used for signalling on ships, and so on. 



2 Described in the Philosophical Transactions of the Royal Society, 1845. 



