The late Neoproterozoic to Cambrian history of East Africa is characterized by two major thermal events. The periods from ~750 to 620 and ~580 to 510 Ma are thought to relate to the closure of the Mozambique Ocean, resulting in the amalgamation of Gondwanaland. There are several schools of thought as to which period actually represents final ocean closure. One model envisages early closure by 620 Ma, involving the collision of ‘African’ cratons with other fragments. Tectonism between 580 & 510 Ma is in this case considered post-collisional, representing orogenic collapse. In the second model, final closure did not occur until ~550 Ma, and tectonism before 620 Ma is explained by the early accretion of continental fragments, or by magmatic underplating.

One approach to resolve these ambiguities is to determine the style of tectonism at 550 Ma. In the early closure model, tectonism between 580 & 510 Ma should be driven largely by extension, while in the model for final amalgamation at ~550 Ma, shortening related to convergence should dominate. Malagasy geology has been interpreted to support either model.

Central Madagascar comprises Palaeoproterozoic supracrustals and ~2.5 Ga basement, intruded by 570-510 Ma granitoids. We find no evidence of major magmatism between 720 and 570 Ma. The main phase of WNW-ESE shortening had started by 550 Ma and involved mostly thin-skinned thrusting. Around 540 Ma, possibly signifying a lock-up stage, strain partitioning increases and deformation in the cover is dominated by sinistral transpression. This is associated with vertical movements between blocks bound by NW-SE and N-S trending strike-slip faults, similar to the style of deformation responsible for exhumation in southern Madagascar. This sinistral transpression continued until ~530 Ma, when lateral escape could no longer be accommodated.

Thus, Central Madagascar provides evidence for continued convergence, rather than for a post-collisional extensional setting between ~550 & ~530 Ma. Along the length of the suture zone(s) of Gondwanaland, oblique convergence was accommodated by horizontal shortening coupled with dominantly sinistral strike-slip. Therefore, we favour the model for late stage closure of the Mozambique Ocean but do not exclude a multiphase assembly, possibly involving the accretion of continental fragments other than Madagascar to East Africa by 620 Ma.