Chemistry remains an experimental science. The theory of chemical bonding leaves much to be desired. Yet, the past 20 years have been marked by a fruitful symbiosis of organic chemistry and molecular orbital theory. Of necessity this has been a marriage of poor theory with good experiment. Tentative conclusions have been arrived a t on the basis of theories which were such a patchwork on approximations that they appeared to have no right to work; yet, in the hands of clever experimentalists, these ideas were transformed into novel molecules with unusual properties. In the same way, by utilizing the most simple but fundamental concepts of molecular orbital theory we have in the past 3 years been able to rationalize and predict the stereochemical course of virtually every concerted organic reaction.' In our work we have relied on the most basic ideas of molecular orbital theory-the concepts of symmetry, overlap, interaction, bonding, and the nodal structure of wave functions. The lack of numbers in our discussion is not a weakness-it is its greatest strength. Precise numerical values would have to result from some specific sequence of approximations. But an argument from first principles or symmetry, of necessity qualitative, is in fact much stronger than the deceptively authoritative numerical result. For, if the simple argument is true, then any approximate method, as well as the now inaccessible exact solution, must obey it. The simplest description of the electronic structure of a stable molecule is that i t is characterized by a finite band of doubly occupied electronic levels, called bonding orbitals, separated by a gap from a corresponding band of unoccupied, antiboding levels as well as a continuum of higher levels. The magnitude of the gap may range from 40 kcal/mole for highly delocalized, large aromatic systems to 250 kcal/mole for saturated hydrocarbons. It should be noted in context that socalled nonbonding electrons of heteroatoms are in fact bonding. Consider a simple reaction of two molecules to give a third species, proceeding in a nonconcerted manner through a diradical intermediate I. A + B + [I] + C

Shu Hui Yow

Papers

Claisen rearrangements of heterocycles