The theory of “Valency” or “Atomicity,” while explaining satisfactorily many chemical phenomena, still leaves unexplained many facts, such, for instance, as the varying atomicities of several elements, the composition of double salts, water of crystallisation, &c. This arises from assuming that chemical affinity acts in an atomic indivisible manner, which is not warranted by the facts. If we take, for instance, nitrogen in the compound, NH3 and NH4Cl, it acts in the one case as a triad, while in the other it is said to be a pentad. Now, ïn the case of the NH4Cl in this view, we have the chlorine represented as directing all its affinity to the nitrogen, for which it usually has such a feeble affinity, while it has no action on the hydrogen, for which its affinity is usually so intense. Now, if we get rid of the idea of the atomic action of affinity, we may represent the chlorine dividing its action among all the four atoms of hydrogen, giving on the average ¼ to each; and the nitrogen, in the same manner, giving ¾ to each; and thus, the whole molecule is bound together.