Hundreds of catalysts have been evaluated in bench activity tests, using synthetic engine exhaust gas to determine conversion of NO to N2 and NH3. Many of these catalysts were also exposed to actual engine exhaust in an engine-dynamometer test, and then re-evaluated in the bench apparatus to determine the degree of catalyst deactivation. For catalysts prepared on nonreactive, hydrothermally stable support materials, it was found that the primary causes of deactivation were sulfur and lead poisoning. Subsequently, it was determined that deactivation of a NOx catalyst from sustained exposure to engine exhaust can be closely simulated by impregnating the fresh catalyst with lead acetate to the required lead level, followed by testing the catalyst with SO2 present in the synthetic exhaust gas. These studies have resulted in the development of catalysts showing high initial NOx conversion efficiency and little deactivation in engine tests.