ABSTRACTThe Lower Werra Anhydrite (Zechstein, Upper Permian) deposits of the teba area originated in a deep basin setting, in shallow to deep water conditions. Facies changes occur within small distances and suggest fluctuating boundaries between well defined basins and platforms. This pattern of local platforms and adjacent basins developed during deposition. In basinal areas, the sequence is clearly transgressive, whereas on platforms accumulation kept pace with subsidence after an initial transgression.Nodular anhydrite represents a polygenetic deposit which formed at different times with respect to deposition. Massive anhydrite with pseudomorphs after upright‐growth gypsum crystals suggest rapid precipitation in a subaqueous environment and/or fluctuating, but generally high, salinity conditions. Massive clastic sulphate originated due to periodic high energy events and resedimentation, or due to brecciation possibly connected with salinity fluctuations and the dissolution of halite. Massive, textureless anhydrite is locally porous and passes upward into breccia, indicating a strongly saline environment. Bedded anhydrite is considered to form in shallow water environments and laminated anhydrite in deep water. Bedded anhydrites contain portions which are graded. Intercalations of sulphate turbidites and upright‐growth gypsum suggest fluctuating water depths, with comparatively deep water during turbidite deposition, but shallower conditions during upright‐growth gypsum deposition.The sequence observed in slope zones at platform‐basin margins, detrital (parautochthonous) sulphate sand to graded beds to basinal laminites, indicates that redistribution processes were important. At the onset of the Lower Werra Anhydrite deposition bathymetric relief existed between the central part of the basin and its margins, where carbonate platforms remained subaerially exposed. Formation of local platforms and adjacent basins required a relatively high subsidence rate, as pre‐existing relief cannot account for the total accumulated thickness of the Lower Werra Anhydrite deposits. One implication of this is that the main argument against ‘the shallow water ‐ shallow basin’ evaporite basin model, i.e.,a very fast rate of subsidence, may not be valid for the Łeba Lower Werra