The two largest continental fragments of Gondwana are Africa and Antarctica. They are related directly to each other by way of only a relatively narrow strip of ocean, 500-600 km in width and about 6000 km in length. The mid-ocean ridge has remained closely central (apart from a small, early adjustment around the Beira High) such that the conjugate margins of northern Mozambique in Africa and Dronning Maud Land in Antarctica are virutally equidistant from present-day seismicity and topography marking the Southwest Indian Ridge within the Corridor. Precise understanding of the geometry of the creation of this Africa-Antarctica corridor (AAC) is important as it forms part of both the two plate circuits for the dispersal of East Gondwana and West Gondwana respectively.

From early in the break-up until about 121 Ma, published marine magnetic anomalies define the relative positions of Antarctica and Africa closely within the AAC. We have used the continuity of fracture zones across the active mid-ocean ridge to interpolate positions of Aantarctica relative to Africa from M0 (121.4 Ma, GTS2020) to C34 (83.64 Ma) when magnetic anomalies resume after the Cretaceous Quiet Zone (KQZ). However, since the AAC is relatively narrow, even the best observations of marine magnetic anomalies (e.g. Muller and Jokat, 2019) can do little better than define a single point to fix Antarctica to Africa at a given time. In the model, the position of Antarctica may, in reality, be rotated about this point (or the series of points through time) by several degrees without offending the data and the alignment of the fracture zones within the AAC.

This a good example of where a holistic approach to modelling Gondwana is important. If Antarctica is rotated too far west about the fixed point in the AAC, growth of the Weddell Sea is compromised as Antarctica would became too close to the 'toe' of South America. On the other hand, if Antarctica is rotated too far to the east, the space between India and Somalia decreases. We have balanced these two constraints at all times and assumed that the proto-Owen FZ remained either a pure dextral strike-slip feature or a leaky transform during the early (Lower Cretaceous) anti-clockwise rotation of India while constrained between Africa and Australia.

Time lines for the two plate circuits are shown in green in the animation. The circuit for West Gondwana goes (anti-clockwise) from Antarctica to Africa then to South America and the Malvinas Plateau (fragment 516). This circuit closes back to Antarctica across the Weddell Sea and has been adjusted so that Weddell Sea development reproduces the observed pattern of fracture zones. The circuit for East Gondwana goes from Madagascar to Africa, then to Antarctica and then India. This circuit closes with India against Madagascar. We maintain India in its 'fit' position against Madagacsar until towards the start of the Late Cretaceous.