The conception of compound radicle was carried by Berzelius to its utmost limits. We have learned that the Swedish chemist regarded every molecule as composed of two parts; in very many cases each of these parts was itself made up of more than one kind of atom--it was a compound radicle. But the Berzelian system tended to become too artificial: it drifted further and further away from facts. Of the two parts composing the dual molecular structure, one was of necessity positively, and the other negatively electrified. The greater number of the so-called organic compounds contained oxygen; oxygen was the most electro-negative element known; hence most organic compounds were regarded as formed by the coming together of one, two, or more atoms of oxygen, forming the negative part of the molecule, with one, two, or more atoms of a compound radicle, which formed the positive part of the molecule. From this dualistic view of the molecule there naturally arose a disposition to regard the compound radicles of organic chemistry as the non-oxygenated parts of the molecules of organic compounds. An organic compound came gradually to be regarded as a compound of oxygen with some other elements, which were all lumped together under the name of a compound radicle, and organic chemistry was for a time defined as the chemistry of compound radicles. From what has been said on p. 268, I think it will be evident that the idea of _substitution_ is a necessary part of the original conception of compound radicle; a group of atoms in a molecule may, it is said, be removed, and another group, or another atom, _substituted_ for that which is removed. Berzelius adopted this idea, but he made it too rigid; he taught that an electro-negative atom, or compound radicle, could be replaced or substituted only by another electro-negative atom or group of atoms, and a positively electrified atom or group of atoms, only by another electro-positive atom or compound radicle. Thus oxygen could perhaps be replaced by chlorine, but certainly not by hydrogen; while hydrogen might be replaced by a positively electrified atom, but certainly not by chlorine. The conceptions of compound radicles and of substitution held some such position in organic chemistry as that which I have now attempted to indicate when Dumas and Liebig began their work in this field.