278 Dr. van der Kolk on the Mechanical 



of affinity, such as day-light, for instance, suffices to produce. 

 The same applies to chloride of nitrogen, &c, and even in the 

 case of the decomposition of water there is nothing that need 

 surprise us. 



Water contains less energy than its components. If, there- 

 fore, these do not combine, this is due to the non-fulfilment of the 

 first condition, or want of a sufficiently strong affinity. If we 

 represent this affinity as =0 when no combination occurs in 

 presence of sufficient energy, in the case before us the affinity is 

 = at the ordinary temperature, it acquires a certain value at 

 higher temperatures, and it disappears again at a still higher 

 temperature — namely, that at which water-vapour is decomposed 

 by heat alone. It is not decomposed by itself at 1000°; but if 

 it is in contact with melted silver, the metal can, in the first 

 place, furnish the energy needed for the decomposition*; and in 

 the second place it can exert a different action on the two com- 

 ponents of the vapour, and thus weaken their mutual affinity. 

 The silver has therefore in this case a twofold action. 



Were simple decomposition to occur at ] 000°, without any- 

 thing further, it would certainly be surprising; but in presence 

 of melted silver this is not the case. 



The development of a higher temperature by the combustion 

 is quite another question. A product is formed containing less 

 energy than the substances from which it is produced ; conse- 

 quently energy is set free in the form of heat and raises the 

 temperature of the resulting water-vapour. This temperature 

 can be calculated from the known values of the heat of combi- 

 nation of water and of the specific heats of hydrogen, oxygen, 

 and water, and is found to be about 6800° C.f In this calcu- 

 lation, however, it is assumed that no heat is communicated to 

 the surrounding medium. But a temper atnre as high as this 

 must immediately fall ; and when we determine it, it is doubtless 

 already much lower. According to Deville's experiments, how- 

 ever, it may amount, under favourable conditions, to at least 

 2500° C. 



Now the melted platinum can, in the first place, supply the 

 requisite energy; and secondly, the metal very probably acts differ- 

 ently at this high temperature on the two constituents of aqueous 

 vapour. Otherwise this would become a problem of molecular 

 forces. 



The phenomena of dissociation cited by the author appear to 



[* How? Why should silver at 1000° impart to aqueous vapour more 

 energy than the latter would receive by contact merely with the sides of a 

 porcelain or platinum vessel at the same temperature ? — Transl.] 



t Legons de Chimie et de Physique, professees a la Societe Chimique de 

 Paris (1861), p. 65. 



