The origin of the ∼40-50 topographically large lake basins on the southern High Plains of Texas and New Mexico has been an enigma. Previous workers have considered deflation or evaporite dissolution at depth and subsequent collapse as the most probable mechanisms. However, the eolian hypotheses have been unable to provide convincing arguments as to how the wind selectively erodes the thick, deflation-resistant calcrete "caprock" that is persistent over much of the southern High Plains. Furthermore, recent detailed studies on some of the basins show no significant evaporite dissolution at depth, and neither mechanism offers a satisfactory explanation as to why the basins are almost universally associated with subsurface topographic highs, or why they are absent where the High Plains aquifer thickness exceeds 60 m. We address these latter concerns and modify the deflation hypothesis by proposing that the calcrete caprock may never have been deposited in the areas now occupied by the basins. The absence of calcrete deposition is proposed to have resulted from high water tables caused by an increase in hydraulic gradient where aquifers thinned above bedrock highs. A high water table close to and/or intersecting the surface prevents deposition of calcrete, and, thus, the uncemented surface would be more susceptible to deflation than the surrounding calcrete-covered areas after decline of the water table. The rise in water table associated with bedrock highs is documented by numerical simulation using boundary conditions and hydrologic parameters representative of the southern High Plains.