The basin experienced rapid subsidence at the same time that sea levels were periodically rising and falling as a result of late Paleozoic glaciation.

As sea level dropped and the basin was cut off from the open ocean, dolomite was again deposited, then gypsum or anhydrite, then halite (which makes up most of the thickness of the cycle), then potash.

This evaporite facies of the formation exists mainly in the subsurface, with only scattered surface exposures of highly deformed beds of the less soluble minerals.

[3] To the northeast, near the Uncompahgre uplift, the Paradox Formation abruptly transitions to clastic rock assigned to the undivided Hermosa Group.

To the southwest, the basin gradually shallows, and the evaporite beds are replaced by limestone of the carbonate shelf facies of the Paradox Formation.

Here each cycle consists of black shale, then carbonate mudstone, then highly fossiliferous silty limestone, then algal mounds, ending with a cap facies of sediments deposited in a very shallow, high-energy environment.

Baker, Dane, and Reeside examined well logs from throughout the region and concluded that the evaporite beds filled a subsurface basin some 150 by 80 miles (240 by 130 km) in extent trending northwest to southeast.

Because surface exposures were few and poor, they described a section for the formation from a well drilled at Schafer dome, about 10 miles southwest of Moab, Utah.

[6] In 1967, Don L. Baars, J. William Parker, and John Chronic further refined the definition of the Paradox Formation, establishing its age as Atokan to Missourian, and defining a subsurface reference section at depths of 5,420 to 9,600 feet (1,650 to 2,930 m) at a well northeast of Egnar, Colorado.