Page:Outlines of Physical Chemistry - 1899.djvu/46

 26 OUTLINES OF PHYSICAL CHEMISTRY

of certain common properties, but this would lead us too far, and the examples given suffice to show how we apply the principle that the chemical character of a substance depends on its molecular structure, and that from similar properties a similar structure may be inferred. It may be added that this principle is a great aid for the elucidation of the chemical constitution of substances.

From the physical properties of compounds we may also arrive at certain conclusions with respect to the mole- cular structure. We shall recur to this subject when dealing with the densities, melting and boiling points, behaviour to polarised light, heat of formation, electric con- ductivity, and other properties of substances.

Is valency constant or variable ? l

A new hypothesis is generally brought forward in its simplest and most uncompromising form. For a time it succeeds in explaining the facts then known, and by some clever artifice may even surmount the first difficulties. But sooner or later there comes a time when the exceptions become so numerous that the shortcomings of the theory can no longer be denied. Then the theory must either fall or be modified.

The theory of valency has not escaped this fate. Suggested by the investigation of simple volatile com- pounds, applied with ingenuity to the interpretation of innumerable carbon derivatives, the hypothesis of invariable valency had victoriously passed through its first trials and promised to be of almost universal application.

Nevertheless, a certain number of substances were known in which the elements of the sulphur group and of the nitrogen group seemed to be endowed with a valency

1 Note by Translator. — In this connection it may be mentioned that Werner has proposed a theory which supposes that some of the elements possess, besides their valency, another attractive power which he terms co-ordination. The theory was proposed in order to explain some of the metal-amines, and might also be extended to explain water of crystallisation.

�� �