Abstract
Cases are found in which the Strutinsky prescription for single-particle corrections to the liquid-drop model fails. Estimates of the spontaneous-fission lifetimes of super-heavy elements are found to be overly sensitive to small changes in the assumed single-particle level spectrum.