on the Physical Properties of Gases. 183 



of dissociated atoms and of molecules of A A and B B become 

 equal. As the temperature still further rises, the numbers of 

 molecules of A A and B B become a maximum somewhere 

 about J Oqj after which they become less and less, whilst at a 

 temperature near ^ Oq the number of dissociated atoms of A 

 or B is about equal to the number of molecules of A B. From 

 this point the dissociation of A A, B B, and A B takes place 

 pretty uniformly, until at an infinitely great temperature they 

 cease to exist, whilst the number of A or B atoms increases 

 correspondingly up to N. 



32. The final conclusion we must draw is that an hypothesis 

 which depends for the explanation of chemical combination 

 on the dissociation of the component molecules, and combi- 

 nation of the dissociated atoms by any simple law of attrac- 

 tion, is not sufficient to account for all the phenomena; but it 

 seems as if the action between two atoms must be a compli- 

 cated one. It may easily be shown that, unless there be 

 dissociation at all temperatures, the dissociation at high tem- 

 peratures cannot be explained simply on the supposition that 

 the internal energy of the molecule is too great to allow it to 

 exist. For let Q^, Q^ be the heats of combination of two gases 

 (masses m^, 1712) at temperatures t and respectively, Ci, C2, c 

 the specific heats at constant volume of the components and 

 compound ; then by combining the gases at temperature ty 

 cooling the compound to zero, decomposing, and heating the 

 mixed gases up again to temperature t, we may show that 



Qi = Qo "" { (^1 + ^^2)^ — ^1^1 — ^2^ } i- 



Now suppose t to be the temperature at which dissociation of 

 the whole would suddenly take place, then clearly Q< = 0, and 

 we should have 



Qo 



t= 



(mi + m2)c — niiCi — W2C2 



In the case of a diatomic perfect gas, c = Ci = C2 and ^ = qo ; or 

 no dissociation could take place unless Q^=0, i. e. unless there 

 is never any combination. But gases are only nearly perfect, 

 and consequently the denominator in the above will be ex- 

 tremely small : e. g. if we take the case of H2 0, and suppose 

 that the specific heats remain the same at all temperatures, we 

 shall find that t has a value which must be greater than 

 64,000° C. ; and we know that steam is almost wholly decom- 

 posed at temperatures far below this. 



33. So much we gather as to the chemical properties of 

 gases. The physical properties of simple elementary gases 

 were discussed in the previous paper with a somewhat more 



