Transform faults constitute conservative plate boundaries, where adjacent plates are in tangential contact. Transform faults in the ocean are marked by fracture zones, which are long, linear, bathymetric depressions. One of the largest transform offsets on Earth can be found in the South Atlantic. The 1200 km long Agulhas Falkland Fracture Zone (AFFZ), form by this, developed during the Early Cretaceous break-up of West Gondwana.Between approx. 41°S, 16°E and 43°S, 9°E the Agulhas Falkland Fracture Zone is characterised by a pronounced topographic anomaly, the Agulhas Ridge. The Agulhas Ridge rises more than 2 km above the surrounding seafloor. The only equivalent to this kind of topographic high, as part of the AFFZ, is found in form of marginal ridges along the continental parts of the fracture zone, namely the Falkland Escarpment at the South American continent and the Diaz Ridge adjacent to South Africa. But the Agulhas Ridge differs from both the Falkland Escarpment and the Diaz Ridge in the facts (1) that it was not formed during the early rift-drift phase, and (2) that it separates oceanic crust of different age and not continental from oceanic crust.A set of high-resolution seismic reflection data (total length 2000 km) and a seismic refraction line across the Agulhas Ridge give new information on the crustal and basement structure of this tectonic feature. We have observed that within the Cape Basin, to the North, the basement and sedimentary layers are in parts strongly deformed. We observe basement highs, which point towards intrusions. Both the basement and the sedimentary sequence show strong faulting. This points towards a combined tectono-magmatic activity, which led to the formation of basement ridges parallel to the Agulhas Ridge. Since at least the pre-Oligocene parts and, locally, the whole sedimentary column are affected we infer that the renewed activity began in the Middle Oligocene and may have lasted into the Quaternary. As an origin of the renewed tectono-magmatic activity we suggest modifications in spreading rate and direction as a result of the Discovery hotspot chain activity starting ~ 25 Ma (Kempe and Schilling, 1974) and the significant deceleration of the African plat since at least 19 Ma (O'Connor et al., 1999).Kempe, D., Schilling, J.G. (1974), Discovery Tablemount basalt:Petrology and geochemistry. Contrb. Mineral. Petrol., 44, 101-115.O'Connor, J.M., Stoffers, P., van den Bogaard, P., McWilliams, M. (1999), First seamount age evidence for significant slower African plate motion since 19 to 30 Ma. Earth Planet. Scie. Letts., 171, 575-589.