The functional surfaces of porcelain restorations are often ground to adjust occlusion. This removes the surface glaze and introduces flaws. Re-firing the restoration before final placement produces a self-glaze layer on the surface that may reduce the extent of damage caused by grinding. The objectives of this study were to determine whether re-firing after grinding increases the strength of dental porcelain and to determine whether the effectiveness of this treatment was dependent on the initial flaw size. Six groups, each containing 12 disk specimens, were prepared from experimental body porcelain (No. 36, J.F. Jelenko & Co., Armonk, NY). The specimens were fired under vacuum, ground to a thickness of 1 mm with 240-grit SiC abrasive, and polished through 600-grit on the surface to be subjected to tensile stress. Control Group A consisted of 12 non-indented specimens. Flaws were induced in the remaining groups by means of a Vickers indenter under the following loads: 3.9 N (Group B), 7.8 N (Group C), 11.8 N (Group D), 15.7 N (Group E), and 19.6 N (Group F). Following indentation, half of the specimens from each group were re-fired. The flexure strength of each specimen was determined by means of a piston-on-three-ball biaxial fixture. The TTEST procedure of SAS (1985) indicated a significant difference between the mean strength values of glazed and non-glazed specimens in Group E only (p = 0.55, 0.24, 0.13, 0.07, 0.01, and 0.69 for Groups A to F, respectively). ANOVA with the GLM procedure of SAS (1985) revealed that the mean strength values of groups subjected to five indentation loads were not significantly different (p = 0.45 for glazed specimens and p = 1.00 for non-glazed specimens). These findings support those of Fairhurst et al. (1992) for specimens with smaller flaws that were induced by grinding with 1 micron abrasive grit. The results of the current study indicate that re-firing of porcelain with large surface flaws does not significantly increase the flexural strength.