On the antiperiplanar lone pair hypothesis and its application to catalysis by glycosidases.

The idea that sp3 lone pairs of electrons on a heteroatom direct the departure of a leaving group from an adjacent tetrahedral carbon centre is shown to be untenable as a general principle, and the permissive evidence in favour of it to be accounted for by the principle of least nuclear motion. The application of the idea to the majority of glycopyranosidases, which work by a double-displacement mechanism, is shown to require implausible contortions of the pyranose ring.