Nothing to date has appeared in the literature addressing the relationship of specific bevel dimension and clinical sharpness of curette blades. This study investigated the degree of clinical sharpness of Gracey curettes following four periods of simulated root planning and, using the scanning electron microscope (SEM), determined the approximate number of strokes at which loss of clinical sharpness became apparent. Seventy-five new stainless steel curette blades were standardized and randomly assigned to one of five groups: control, 30-stroke, 50-stroke, 70-stroke, and 90-stroke. With the exception of the control group, blades underwent a simulated root-planing procedure, on extracted, periodontally involved teeth, which employed a device to standardize pressure and stroke length. Blades were then tested for clinical sharpness using light reflection, with two degrees of magnification; and plastic stick, a tactile evaluation. Blades were photographed under the SEM at a magnification of x1,000. Bevel width was measured at 10 standardized locations on the photomicrographs, and a mean was calculated. Nonparametric data were analyzed using the Kruskal-Wallis statistical test and SEM parametric data using ANOVA and post hoc Newman-Kuels tests. Results showed no significant differences among treatment groups when either tactile or magnification sharpness tests were used. Analysis of the SEM data showed no statistically significant differences among bevel dimensions for any of the five groups. A Spearman rank order correlation, used to compare the clinical data to the SEM bevel measurements, showed no correlation of clinical sharpness and SEM-determined bevel dimension.