At present with Gerhardt we deny the possibility of giving an account of the position of atoms within the molecule and it seems very natural that chemistry, which is concerned with substances only when they are in a state of change, is powerless to judge this mechanical structure, unless helped by physical methods. On the other hand, however, Laurent has already shown in his classic Méthode de chimie that such knowledge is not completely unattainable and it would be difficult for anyone to claim that it will remain beyond our ability in the future. We may not know what is the connexion between the chemical influence which the atoms within the compound molecule exert upon one another and their respective mechanical position. We do not even know whether in a compound molecule two atoms producing an immediate chemical effect on one another are situated next to each other; but, quite irrespective of the idea of physical atoms, we cannot deny that the chemical properties of a substance depend on the inter-relation of its constituent elements. Assuming that each chemical atom contains only a certain limited quantity of chemical force (affinity), with which it participates in the formation of a compound, I should like to indicate this chemical inter-relation or the manner in which the atoms are bound together in a compound by the term 'chemical structure'.