MancJiester Memoirs, Vol. Hi. (1908), No. 10. 47 



decomposition ; that in fact any transformation of more 

 complex type than this must be expected to occur in 

 successive stages. Thus the simplest case of a triple 

 dissociation, say of a molecule ABC into A and B and C, 

 may be held to occur in two stages, the first stage being 

 such as a change into AB and C : the reason may be 

 repeated, that without an intermediate diad stage the 

 velocity of association of such molecules would be 

 extremely slow compared with the velocity of dissociation 

 of those already formed, so that in equilibrium the triad 

 compound would practically not exist.* 



It appears indeed from the facts that chemical 

 equilibria involving processes more complex than double 

 decomposition occur but rarely. Where they occur 

 at all, it is here suggested that there must be an inter- 

 mediate stage, perhaps very transient ; and the question 

 arises whether its existence may modify the usually 

 accepted deductions from the Guldberg-Waage statistical 

 representation of the chemical equilibrium, or the 

 application of thermodynamics of which that theory 

 forms one aspect. 



The matter will, however, assume a sufficiently com- 

 plicated form, at any rate for initial consideration, in the 

 very simplest example. Let us then examine the chemical 

 equilibrium of a dissociating gaseous substance ABC, 

 mixed with its components. Let this symbol ABC 

 denote quantity of the substance ABC, measured in 

 chemical equivalents, say by number of molecules per 



* Considerations such as these must often have occurred to molecular 

 theorists. It is only recently (a year after this note was composed) that, 

 in an obituary notice of Mendeleef by Dr. E. C. Edgar [Manchester 

 Memoirs 51, 1907) a remark about "his persistent devotion to the 

 Mendeleef-Gerhardt law, that gases combine only in equal volumes " has 

 prompted a reference to the section on Atoms and Molecules in the 

 " Principles of Chemistry," where views essentially equivalent to the above 

 are powerfully supported on purely chemical grounds. [See footnote, p. 29 

 siipra.'\ 



