Abstract
Electronic energy levels in molecules are discussed, by the same methods previously used in discussing atomic levels. The problem in general is much more difficult than for atoms, and less progress can be made toward quantitative solution. For the lowest electronic level in cases where there are definite valence bonds, however, approximate solutions can be set up which should be fairly accurate, and which lead to a definite verification of the ideas of the writer and of Pauling regarding directed valence and the importance of concentrated bond functions. A number of special cases are discussed, involving both spin and orbital degeneracy.