xii PRESIDENT'S ADDRESS. 



the newer type theory, and finally the theory of valence,, mark 

 notable steps in the progress. Each successive theory explained a 

 wider range of facts than its predecessor, and gave place in turn 

 to a theory capable of interpreting a yet wider range. We are 

 here concerned only with the last-named, the theory of valence. 

 This theory attributes to each atom a strictly limited capacity for 

 combining with other atoms, as measured by the number of atoms 

 with whi'ch, it can combine. Accordingly, an atom cannot combine 

 with or, figuratively speaking, become linked to, an indefinitely 

 large number of other atoms, but only with a small number, the 

 atoms of different elements having different capacities in this 

 respect. For example, an atom of hydrogen, or of chlorine can 

 never combine directly with more than one other atom and these 

 elements are therefore called univalent. The capacity of an 

 oxygen atom for combination is exhausted by combining with two 

 atoms of hydrogen or any other univalent element; and hence 

 oxygen is called bivalent. An atom of carbon can combine 

 with a maximum of four hydrogen or four chlorine or two 

 oxygen atoms, that is, carbon is tetravalent. With the aid 

 of this hypothesis it was now possible to interpret experi- 

 mental results by the formulation of relationships between 

 the atoms of a molecule. An example will make this clear. 

 The composition of alcohol is expressed by the formula 

 C 2 H 6 0. Now one-sixth of the hydrogen and all the oxygen 

 are removable from alcohol, and re-appear together again in 

 one of the reaction products. These experimental results can be 

 interpreted by attributing to one of the hydrogen atoms a different 

 relation to the compound from that of the other five, and by 

 supposing that this hydrogen atom is directly combined with the 

 oxygen atom. But as hydrogen is univalent and oxygen bivalent, 

 these relations would have to be expressed by the formula 

 C 2 H 5 -0-H. Interpreting in similar fashion other reactions of 

 alcohol, and assuming the tetravalence of carbon, we finally arrive 

 at the formula CH 3 -CH 2 -O-H which expresses relationships 

 between the atoms. Now formulae of this kind known as 

 structural or constitutional formulae not only suggest new 

 properties, but also methods of synthesis. For example, the 

 iibove formula suggests a method of making alcohol from J .he 



