ULTIMATE CONSTITUTION OF MATTER 75 



nises that, when he is working with a substance in solution, 

 his homogeneous or unit substance is not the salt with the 

 properties of which he is conversant in its dry condition. 

 The salt tends to divide into its ions. It is the reactions 

 of the separated ions that he is now investigating, not the 

 reactions of the salt as a whole. But at the end of the 

 reaction a new product comes back into the light, and he 

 speaks of this as the product of the interaction of the salts 

 which he dissolved and the other reagents which he used, 

 whatever they may have been. He has therefore to resolve 

 the mixed constituents of the globe into their elements, 

 and to ascertain the properties of every combination of 

 elements which can exist, whether these combinations are 

 separable as forms of matter which can be isolated and 

 set aside in the drawers and bottles of the laboratory, or 

 whether they can exist as separate bodies only under con- 

 ditions which render their isolation impossible. 



But in chemistry, as in all other branches of Natural 

 Science, the observation of phenomena provokes reflec- 

 tions as to their cause. Why do the elements combine? 

 Why, when a compound has been formed, is it ready 

 under certain circumstances to exchange one of its elements 

 for another, or to react with some other compound in such 

 a way as to produce either a more complicated compound, 

 or two or more substances which do not resemble either 

 of those from which they are derived? Chemical philos- 

 ophy is occupied with many problems ; but the one which 

 is most distinctly chemical is the determination of the 

 positions which the elements in a compound occupy rela- 

 tively to one another, the architecture of derived substances, 

 as it may be termed. It is necessary to think of matter 

 as composed of atoms, whatever may be the nature of 

 these units of structure. If our powers of vision were suf- 

 ficiently increased, we should see matter, not as we see 

 treacle, but as we see marbles when enclosed in a vase of 

 clear, transparent glass ; with this difference, that the mar- 



