Abstract The dissolution rate of rutile in hydrofluoric acid was measured to examine the importance of line and point defects on the reaction rate. Well-annealed rutile was shocked with an explosive charge to induce a high density of dislocations (4 × 10 11 cm −2), paramagnetic point defects (≈2 × 10 19 cm −3), and lattice strain as measured by X-ray line broadening (≈ 2 × 10 −3). Subsamples of the shocked material were then thermally annealed in order to prepare samples with a wide variation in dislocation density and point defect concentration. The unshocked starting material had a dislocation density of about 10 6 cm −2, no detectable lattice strain (<10 −5), and no detectable point defects as measured by electron spin resonance (< ≈1 × 10 16 cm −3). In spite of this large variation in dislocations and point defect concentrations, we observed only a factor of two variation in the reaction rate for shocked and annealed material. Existing theories suggest that etch pits nucleate and grow at the outcrop of a dislocation on the crystal surface. Assuming that the measured dislocation densities can be used to estimate the density of surface outcrops, and that each outcrop dissolves to form an etch pit, the relation between dissolution rates and dislocations can be interpreted in terms of a mean rate of etch pit growth. The mean rate of etch pit growth in shocked samples is less than 9 × 10 −27 cm 3 s −1. In experiments with single rutile crystals, however, observable etch pits grow at rates on the order of 1 × 10 −19 cm 3 s −. This discrepancy suggests either that: (i) the observed etch pits grow much more rapidly than the mean rate; or (ii) bulk dislocation density is not a useful measure of potential sites for etch pit growth on rutile crystal surfaces.