The efficacy of chlorine dioxide as a disinfectant was evaluated against cells of Escherichia coli ATCC 11229 in aqueous suspension and adhering to the surfaces of stainless steel AISI 304 and PVC. The concentrations tested ranged from 0.7 to 14 mg/liter; the exposure times investigated were 30 s and 1, 2, 4, and 8 min. When the bacteria were suspended in water with 1.4 mg/liter of chlorine dioxide, a 10(5)-fold reduction of the initial viable count occurred within 30 s; when cells were attached to the steel surface, the same rate of inactivation took place only after 6 min with 7 mg/liter or 4 min with 14 mg/liter of chlorine dioxide. A 5-log reduction was not obtained when organisms were adhered to polyvinyl chloride (PVC). Scanning electron microscope micrographs of contaminated surfaces revealed that the PVC was very rough with pores much larger in diameter than the cells. Time values determining a 90% reduction of the E. coli population (90% killing time) were calculated for each concentration of disinfectant tested in suspension and on the steel surface. If the same experimental conditions were strictly adopted, linear functions of the log of bacterial inactivation could be plotted (log 90% killing time versus log concentration of disinfectant). This work showed that results obtained with suspension tests could not be used to estimate disinfection of hard surfaces.