The Namibian offshore region has been studied using more than 14 300 km of new multifold seismic data to provide insights into the history of lithospheric extension in the South Atlantic. Initial subsidence related to crustal stretching began at least as early as Permo-Triassic times. Continued stretching caused multiphase rifting which migrated northwards with time, as broadly recognized in other South Atlantic basins north of the Agulhas fracture line. Rifting was followed by thermal subsidence in the Late Cretaceous and Tertiary. Geophysical data indicate that south of the Walvis Ridge, major rift faults dip landwards. A structural model based on reflection seismic, gravity and magnetic data suggests that rifting was asymmetric and accommodated by the development of a major easterly dipping crustal shear and mantle wedge. The model suggests that thinning of the crust does not correspond vertically with synrift sedimentation represented in the Central Half Graben. Seaward dipping rift faults north of the Walvis Ridge suggest that the major crustal shear dips westwards. This rift geometry may have been produced by simple shear analogous to that of the Basin and Range model of extension. The analogy is corroborated by major onshore structural features and the distribution of silicic volcanics on both sides of the South Atlantic. Our model suggests that initial Gondwanaland break-up was accomplished by simple shear related to passive stretching of the crust. As there was an association between rifting in the South Atlantic and subduction processes active on the Pacific margin of West Gondwanaland, the stretching may be related to changes in plate boundary stresses. Close to the convergent margin in South America, rifting was influenced by the north-west trending Late Permian accretionary fabric, but the main break was parallel to the Late Proterozoic-Cambrian Pan African-Braziliano sutures of the Lavalleja-Dom Feliciano and Ribeirio Belts. Keywords: Namibia, extensional tectonics, passive rifting, South Atlantic.