32 CLARKE 



ered in the light of Faraday's law. Equal quantities of energy, 

 in the form of electricity, are required to effect the decomposi- 

 tion of the three molecules. We now see that equal quantities 

 of thermal energy are needed to produce the same result, or 

 will be liberated in the formation of the compounds. The same 

 principle is operative in both cases, as should be expected, but 

 its influence has hitherto been masked. The apparent heats of 

 formation offer no suggestion of it, but the absolute values 

 speak more clearly. It is to be hoped that in the near future 

 the laws of thermal union and Faraday's law may be more com- 

 pletely correlated. 



The application and verification of this principle are evidently 

 complicated by the difficulty of eliminating all purely physical 

 disturbances, such as are due, for instance, to changes of state. 

 Many of the existing thermochemical data are in this respect 

 almost hopelessly complex. And yet, here and there, we find 

 observations of the most suggestive character, as for example, 

 in the determinations which have been made of the heat devel- 

 oped by the solution of an alkaline metal in water. These 

 measurements are as follows, for one gramme atom each of the 

 several metals : 



It is evident at a glance that these quantities are all of the same 

 order, and that the differences may possibly be ascribed to dif- 

 ferences in the liquefaction and ionization of the several metals, 

 and variations in the concentration of the solutions produced. 



^ Therm. Unt., vol. 3, pp. 225, 229, 234. 



*Compt. Rend., vol. 123, p. 694. 



'Ann. Chim. Phys. (6), vol. 12, 358. 



*Jahresbericht, 1891, pp. 4S1, 483. Data for caesium approximate. 



*Compt. Rend., vol. 128, p. 384. 



