How hydrides misled chemists

Hydrogen-containing molecules are simple enough to be attractive subjects in experimental diffraction and spectroscopic studies and in quantum computations. Yet, the inferences about molecular structure and force fields originally drawn from studies of these subjects were significantly flawed. In recent developments the original models of structure invoked, such as hybridization, have been superseded. The reasons for this are briefly reviewed. What has emerged to account for molecular geometry, prevailing even over the popular VSEPR theory, is a model of geminal nonbonded interactions.