The abrasive mechanism of polymer wear is dominant in the startup, or "breakin", stage of polymer/steel sliding systems. This mechanism controls the polymer wear rate until the voids in the hard metal surface are filled, much like the filling observed with a file when used on soft metals. This regime of polymer wear is modeled on an event-by-event basis. The model utilizes a digitized profile of the metal surface, bulk polymer properties such as flow pressure and elongation at break, and a few system parameters such as load and slider geometry. The predictions of the model are compared with experimental data. The predicted wear rates are within a factor of 3 of the measured wear rates for polymers with glass transition temperatures, Tg, above the interfacial temperature (rigid PVC and PCTFE). The validity of the model is shown to be related to the ductile or brittle behavior of the sliding polymer. / Ph. D. / Bibliography: leaves 78-80