188 Mr. M. M. Pattison Muir on Chemical Classification. 



capable of modifying the general action of the elementary 

 atom, or whether they are capable of bringing about changes 

 in the inner mechanism of the atom itself, in which case we 

 should have an explanation of " varying valencies," must at 

 present remain undecided. Summing up the evidence, Meyer 

 asserts (and, it seems to me, truly asserts) that we are justified 

 at present in regarding the " elementary atoms as fixed mag- 

 nitudes, and the composition of compounds as deducible 

 (1) from the unchanging valency of these atoms, and (2) 

 from the changing circumstances under which this valency is 

 exercised, or by which it is influenced in its working." 



Besides most fairly, as I think, summing up the evidence, 

 this statement of Meyer presents us with an admirable work- 

 ing hypothesis ; it is especially valuable as a guide in the ap- 

 plication of the principle of valency to chemical classification. 



19. Granting, then, that each elementary atom possesses 

 a certain capacity of saturation of its own, it would appear 

 that the general nature of compound bodies must be in some 

 way correlated with the valency of their constituent elements. 

 Omitting for the present the consideration of the infiuence 

 exerted upon the general nature of compounds by the chan- 

 ging circumstances under which their elementary atoms enter 

 into combination, let us see whether it is possible to express 

 the composition of some of these substances as a function of 

 the valency of the combining atoms. 



''Any quantity which depends upon and varies with another 

 quantity may be called a function of it, and either may be 

 considered a function of the other."*. 



The problem may be stated in other words : — Can it be 

 shown that the general chemical composition of a group of 

 compounds varies with the valency of the elementary atoms 

 composing these compounds ? When the valency varies, does 

 the composition vary with it ? 



The problem is an exceedingly difiicult one to solve. In 

 " valency " we have a measurable quantity ; but where shall 

 we find a unit in terms of which " general chemical compo- 

 sition " may be measured? 



If problems of the kind under consideration are to be rigo- 

 rously dealt with, it is necessary that the variable quantity 

 should be under the control of the experimenter, and that he 

 should be possessed of means for carrying out similar expe- 

 riments under similar conditions. In the present instance 

 valency is probably a fixed quantity ; chemical composition is 

 variable, but it is so ill-defined that we cannot be certain that 



* W. Stanley Jevons, ' Principles of Science/ vol. ii. p. 113. 



