Transient, non-steady-state responses of hematite dissolution rate to pH-jumps, from high to low pH, contain information about dissolution mechanisms and can he used to improve our understanding of dissolution processes operating under variable natural conditions. Our data show that, following each downward pH-jump, the hematite dissolution rate jumps up but then decays exponentially to a new steady state over a period of about 36 h. This requires that, after a pH-jump, the nature of the surface Fe sites themselves, and not only surface charge, gradually changes. Our results are consistent with the depletion of a reservoir of Fe sites active for dissolution on the hematite surface after a jump to pH 1, and show that such active sites can be reproducibly regenerated during returns to higher pH. We interpret the data with regard to long-standing crystal growth and dissolution models [e.g., Burton- Cabrera- Frank, BCF (Burton, W. K.; Cabrera, N.; Frank, F. C. Philos. Trans. R. Sec. London Ser. A 1951, 243, 299-358)] that assume the existence of "adsorbed nutrient" that is structurally distinct from metal centers in the solid surface structure. The general concept behind the model should be applicable to other minerals as well as hematite.