This work on chlorine was followed up, in 1813, by the proof that the class of acidifiers and supporters of combustion contains a third elementary substance, viz. iodine. As Davy's views regarding acids and salts became developed, he seems to have more and more opposed the assumption that any one element is especially to be regarded as the acidifying element; but at the same time he seems to admit that most, if not all, acids contain hydrogen. Such oxides as sulphur trioxide, nitrogen pentoxide, etc., do not possess acid properties except in combination with water. But he of course did not say that all hydrogen compounds are acids; he rather regarded the possession by a substance of acid properties as dependent, to a great extent, on the nature of the elements other than hydrogen which it contained, or perhaps on the arrangement of all the elements in the particles of the acid. He regarded the hydrogen in an acid as capable of replacement by a metal, and to the metallic derivative--as it might be called--of the acid, thus produced, he gave the name of "salt." An acid might therefore be a compound of hydrogen with one other element--such were hydrochloric, hydriodic, hydrofluoric acids--or it might be a compound of hydrogen with two or more elements, of which one might or might not be oxygen--such were hydrocyanic acid and chloric or nitric acid. If the hydrogen in any of these acids were replaced by a metal a salt would be produced. A salt might therefore contain no oxygen, _e.g._ chloride or iodide of potassium; but in most cases salts did contain oxygen, _e.g._ chlorate or nitrate of potassium. Acids were thus divided into oxyacids (or acids which contain oxygen) and acids containing no oxygen; the former class including most of the known acids. The old view of salts as being compounds of acids (_i.e._ oxides of the non-metallic elements) and bases (_i.e._ oxides of metals) was overthrown, and salts came to be regarded as metallic derivatives of acids. From this time, these terms--acids, salts, bases--become of less importance than they formerly were in the history of chemical advance. In trying to explain Davy's electro-chemical theory I have applied the word _affinity_ to the mutual action and reaction between two substances which combine together to form a chemical compound. It is now necessary that we should look a little more closely into the history of this word _affinity_. Oil and water do not mix together, but oil and potash solution do; the former may be said not to have, and the latter to have, an affinity one for the other. When sulphur is heated, the yellow odourless solid, seizing upon oxygen in the air, combines with it to produce a colourless strongly smelling gas. Sulphur and oxygen are said to have strong affinity for each other.