Abstract Joseph Mawson, a nineteenth-century British railway engineer and businessman in Brazil, discovered fossils from the Cretaceous of Bahia that were described by E. D. Cope and Arthur Smith Woodward. A biographical outline of Mawson is presented. Mawson’s discoveries (especially the giant coelacanth fish Mawsonia , named after him by ASW) are interpreted today in the light of modern geological investigations. Mawsonia apparently lived in fluvial, lacustrine and brackish-water habitats in western Gondwana at the time South America separated from Africa. From the Late Jurassic until the Barremian, Mawsonia was widespread across western Gondwana, but its Aptian–Cenomanian records in South America are restricted to northeastern Brazil (including the Borborema tectonic province and adjacent areas to its north). In contrast, Mawsonia remained widespread in the Aptian–Cenomanian of Africa. Recently published data suggest that northeastern Brazil was still contiguous with Africa in the Aptian/Albian, although it was probably separated from the rest of South America by an epicontinental seaway that apparently followed an unconventional course across the Brazilian interior rather than along the present-day coastline. Aptian–Cenomanian records of Mawsonia and other non-marine taxa (including tetrapods) in northeastern Brazil may therefore represent ‘African’ rather than ‘South American’ biotas.

Abstract The clastic fill of the rifted grabens in the central South Atlantic includes several fields containing over 1 billion barrels of oil in place. The largest fields are mainly hosted by continental fluvial and aeolian sandstones deposited near the base of the rift, before rift topography was fully developed. Footwall erosion progressively eroded back the main fault scarps, and normal faults propagated through the footwall to create terraces along the rift borders. This increased the potential drainage area on the footwall, so eventually, large alluvial fans developed in the hanging wall along the main rift border faults. The basement clast-supported conglomerates scatter seismic energy; and on seismic data, the top of the fan can appear to be the intact crystalline basement. Even when drilled, the top of the fan can still be mistaken for true basement due to the large size of the clasts. Some important oil fields are hidden below these marginal alluvial fans, within prerift and early rift fluvial and eolian sandstones and in lacustrine turbidite sandstone reservoirs. It is suggested that more fields may be found in West Africa using the model of the sub-fan terrace play. The large amount of footwall denudation of dense basement rocks can lead to unloading of the adjacent basin, as well as the footwall, when the crust has a finite flexural strength. The resulting uplift and erosion produce an ‘End of Rift’ unconformity, and it is suggested this process is likely to have caused the so called ‘Breakup’ unconformity on continental margins, rather than this being due to initiation of ocean spreading.