Mississippian Meramec reservoirs of the STACK (Sooner Trend in the Anadarko [Basin] in Canadian and Kingfisher counties) play are comprised of silty limestones, calcareous siltstones, argillaceous-calcareous siltstones, argillaceous siltstones and mudstones. Core-defined reservoir lithologies are directly related to independently derived petrophysical rock types based on core porosity-permeability relationships. Machine-learning classification was employed with core and well-log data to predict lithologies in non-cored wells based on well-log signatures. An Artificial Neural Network produced an overall accuracy of 93% by charting the model results in a confusion matrix and was applied to a suite of logs in non-cored wells to generate lithology logs for the Meramec. Using core, wireline well logs and classified lithology logs, the Meramec was divided into eight stratigraphic units characterized as strike-elongate, shoaling-upward parasequences. The bottom three parasequences (lower Meramec) form a retrogradational parasequence set that back-steps to the northwest, with each parasequence capped by a marine-flooding surface and the parasequence set capped by a maximum flooding surface. The upper Meramec is characterized by two parasequences that form an aggradational to progradational paraseqeunce set followed by two transgressive parasequences in a retrogradational parasequence set. Lithology, rock type, porosity, permeability, and water saturation models were generated to explore their stratigraphic and lateral variability and to evaluate their relationships to production, pore volume, and hydrocarbon pore volume. Calcareous-rich lithologies and rock types commonly exhibit lower values of porosity and permeability with higher water saturation. Argillaceous-rich lithologies and rock types have relatively higher porosity and permeability with lower water saturation. Examining the spatial distribution of reservoir properties within the stratigraphic framework, reservoir quality improves moving up in the retrogradational parasequence set then worsens in the overlying progradational parasequence set. The ideal reservoir quality lies in the parasequences below and above the maximum flooding surface where more argillaceous-rich lithologies and rock types exist, resulting in optimal petrophysical properties, higher pore volume and hydrocarbon pore volume.