Mr. M. M. Pattison Muir o?^ Chemical Classification. 203 



was based upon tlie functions performed by these substances, 

 and those in which the system was based upon the composition 

 of the substances classified. In the six subsequent paragraphs 

 I endeavoured to estabKsh the truth of this statement upon 

 historical evidence, while at the same time I attempted briefly 

 to trace the rise of that system of classification which, founded 

 mainly upon the doctrine of valency, should pay regard alike 

 to function and composition in attempting to frame an exhaus- 

 tive classification of chemical substances. In paragraphs 11 

 to 23 I endeavoured to sketch the outlines of the theory of 

 valency, noticing the difficulties which have been overcome 

 and those which are yet to be conquered ; and I sought to 

 show that a connexion really exists between the properties 

 (chemical and physical) of a compound and the valency of its 

 constituent atoms. I have also endeavoured to prove that the 

 theory of valency, along with those subsidiary theories which 

 have sprung from it, enables us to learn something concern- 

 ing the composition or structure of chemical compounds. 

 The facts already stated are, I think, a justification of what I 

 said in paragraph 11, viz. that the system of classification 

 which is mainly based upon the valency of the combining 

 atoms pays regard alike to function and to composition of com- 

 pounds. But I think it will now be possible to indicate in a 

 more definite manner how this result is gained. 



26. In paragraph 23 I have adduced reasons for assigning to 

 common alcohol the constitutional formula CHg — CHg — OH, 

 and to acetic acid the formula CH3 — CO — OH. One of the 

 reasons given is the fact that one hydrogen atom, and one 

 only, can be replaced by metal in either of these compounds. 

 It is hardly correct to say, can be replaced by metal ; for 

 although the statement holds good, so far as we know, for 

 acetic acid, yet it is only the alkali-metals which are capable 

 of replacing one sixth of the hydrogen contained in alcohol ; 

 and, moreover, the alkali-metal must be presented in the isolated 

 state. Although, then, we assign to each of the compounds 

 under consideration a constitutional formula containing the 

 group OH, yet we must believe that the function of this group 

 is influenced in some way by the other atoms constituting the 

 molecule. 



From a consideration of the reactions of water, caustic pot- 

 ash, and hypochlorous acid, we assign to these bodies the for- 

 mulas H — OH, K — OH, and 01 — OH respectively ; that is, 

 we believe that the molecule of each contains the group OH : 

 yet the properties of these three bodies are very different. It 

 is impossible, on the theory of atom-linking, and accepting the 

 generally received valency of the constituent atoms, to assign 



