Mineral dissolution is among the most important sets of chemical reactions that occur at and near the Earth's surface. In recent years, efforts to understand mineral dissolution have shifted to include both macro-scale processes, in which batch reactors are used to determine bulk dissolution rates, and micro-scale processes, in which surface techniques such as Atomic Force Microscopy (AFM) are used to directly observe processes occurring on mineral surfaces during dissolution. Significant progress that links the physical processes of mineral dissolution to solution chemistry has been achieved through this approach. This study embraces the same strategy and aims at exploring the kinetic and thermodynamic constraints of mineral dissolution reactions and the relations between surface processes and bulk dissolution rates. An experimental approach through in-situ observations by fluid cell Atomic Force Microscopy (AFM) of the real-time processes was used to investigate crystallographic direction-dependence behaviors of calcite and gypsum during dissolution. AFM observations show simultaneous retreat and advance of the + and - steps on calcite {1014} cleavage faces, indicating that steps in those two directions may have different solubilities. This result is consistent with the thermodynamic prediction that solubility (Kb) is face dependent as long as energy changes associated with the attachment of individual growth units in symmetrically unrelated directions are different. Furthermore, the measurements of critical step lengths on calcite {1014} cleavage faces display a clearly defined inverse dependence on saturation state that closely follows the theoretical curve. These findings suggest that the direction-dependence of Kb should be taken into account if the thermodynamic and kinetic properties of monolayer steps are of interest and the size-dependence of Kb should not be ignored when nano- to submicron-sized crystals are concerned. Moreover, experimental observations of gypsum surface dissolution process reveal that the only type of etch pit formed on the (010) and (010) surfaces of gypsum is that embraced by the [100] and [001] steps. This result indicates that previously reported pits composed of [101] and [001] steps may be incorrectly defined due to neglecting the differences in the atomic structures of the (010) and (010) cleavage planes. In addition, the experimental observations of gypsum dissolution show that gypsum dissolution on {010} faces is not characterized by the formation of deep etch pits, even at conditions far from equilibrium. Further study suggests that this occurs when less stable steps propagate as step trains at a higher speed that out-paces the vertical growth rate of etch pits. Another approach involving an ex-situ mix-flow reactor was used to investigate calcite dissolution kinetics at near and far from equilibrium conditions. Experimental results suggest that the relation between dissolution rates and the Gibbs free energy is highly nonlinear and rather sigmoidal. This finding indicates the classic Transition State Theory (TST) based rate equation may not be sufficient to depict the relation between dissolution rates and the Gibbs free energy for calcite. Using the same approach, dissolution experiments were conducted to study the role of dislocation density in controlling mineral dissolution. The experimental observations of this research show that calcite dissolution rates did not increase with the dislocation density at saturation conditions near and far from equilibrium. This result suggests that the effect of dislocation density on mineral...