Humankind has been able to develop a metals-based civilization primarily because the reactive metals (Fe, Ni, Cr, Al, Ti, Zr, . . .) exhibit extraordinary kinetic stabilities in oxidizing environments. From the time of Schonbein and Faraday (1830s), the reason for this stability has been attributed to the existence of a thin reaction product ®lm on the metal (or alloy) surface. This ®lm effectively isolates the metal from the corrosive environment. However, attempts to elucidate the mechanisms of the formation of passive oxide ®lms, which generally comprise bilayer structures consisting of a defective oxide that grows directly into the metal and an outer, precipitated hydroxide (or oxyhydroxide on even oxide) layer, have yielded only a rudimentary understanding of the chemistry and physics of the growth and breakdown processes. In this paper, selected aspects of passivity and passivity breakdown are reviewed, with emphasis on the physical models that have been proposed to account for the experimental observations. One such model, the Point Defect Model, is shown to account for most, if not all, experimental observations, and to provide a robust basis for predicting the occurrence of passivity breakdown in any given system. By combining the Point Defect Model with deterministic models for pit growth and crack growth, it is now possible to predict the evolution of localized corrosion damage in a wide range of systems.