researches, led to the general theory of valency, which included in itself the essential features of the older doctrine of equivalents.
Having thus gained the conception of a definite binding power as applicable to elementary atoms or groups of atoms, it followed, as an almost necessary deduction, that the smallest parts of chemical compounds which existed as distinct chemical entities, i. e., the molecules, must have a definite structure: that the parts (atoms) of the little systems must be arranged in accordance with the valencies, or binding powers, of these parts.
Hence, given the number of atoms in a molecule, and the valency of each atom, it became possible to calculate the number of different arrangements of these atoms which could be produced; and careful experiment has often succeeded in preparing all the different, theoretically possible, compounds. The difference of properties of such compounds, i. e., of compounds the molecules of which are constituted of the same number of the same atoms, but differently arranged, is attempted to be indicated in the "structural" or "rational" formulæ of modern chemistry.
Berzelius spoke of compounds composed of parts held together by mysterious bonds: the idea survives in these structural formulæ of today, only we are now able to define what we mean by the smallest part of a compound having a chemical existence, and we have gained certain generalizations which enable us to trace with some degree of accuracy the relationships which exist between the inner parts of these smallest chemical wholes. We appear to be now fairly embarked in the prosecution of molecular dissection, and our chief guide is the theory of valency, itself a development of the dualistic chemistry.
Each elementary atom, I have said, seems to have the power of directly binding to itself a maximum number of other atoms; but it would further appear as if the groups of atoms thus produced had also a certain binding power, but this more indefinite than the atomic binding power, and very variable under different physical conditions. This atomic binding power appears to have a fixed maximum value, but not always to reach the maximum. What is the exact way in which the binding power or valency of the elementary atoms is influenced by definite changes in physical conditions? This is one of the most important unsolved problems of general chemistry.
Then, again, granting the existence of an inner structure to the molecule, granting that groups of atoms do exist in the molecular building, does the fact, that in a certain reaction certain atoms are withdrawn as a group, prove that these existed in the form of the same group in the original molecule? In other words, do our structural formulæ express the relative collocation of atoms within the molecule while the molecule is unacted on by extraneous force, or do they merely roughly represent the condition of things when the molecule is in a state of strain, because of the stress between its parts and
