240 Mr. J. J. Thomson on the 



remain constant; or we may say that in such cases the dissocia- 

 tion produced is inversely proportional to the square root of 

 the density. 



It will be easy and convenient to express equation (3) as 

 an equation giving the vapour-density of the dissociated gas. 

 Let A be the vapour-density of the dissociated gas, D the 

 vapour-density of the gas when it is not dissociated ; then 



_N 

 A 2 N N 



D n + m 2n + 2m N + »* 



. n_ T)-A 

 "N A ' 

 and 



. ro 2A-D 

 '"'N 2 A ' 



Substituting these values for n and m, equation (2) becomes 



Now if p be the pressure of the gas, p = cN, where c is a con- 

 stant ; substituting for n we have 



v-*yU^Y « 



Equations (3) and (4) could be applied most easily to the 

 case when the dissociation is produced by external agency, 

 such as light or electricity, because in this case t and r would 

 be independent of N, and these equations would then give the 

 vapour-density of the gas at different pressures, the disturb- 

 ance producing the dissociation remaining the same. No 

 quantitative measurements seem, however, to have been made 

 for such a case. Some experiments have been made on the 

 dissociation of iodine-vapour by heat. If the dissociation 

 were due in this case to the collision of the particles, then the 

 paired time would vary inversely as the number of collisions, 

 and therefore inversely as N, so that Nt would be constant, 

 and, as equations (3) and (4) show, the dissociation would be 

 the same at all pressures ; the experimental results show that 

 this is not the case, so that from our point of view we cannot 

 assume that the collisions are the cause of the dissociation. 

 Another circumstance which points to the same conclusion is 

 the fact that dissociation is very much hindered by the pre- 

 sence of a neutral gas; if the dissociation were due to shocks, 



