Palaeomagnetism and the reconstruction of gondwanaland

Abstract

The past existence of the supercontinent Gondwanaland has been regarded as the basis of a now respectable hypothesis which explains the present distribution of rock types peculiar to the southern hemisphere continents and India. Gondwana geologists were clearly attracted to the hypothesis even though specific details of the reconstruction remained unresolved. Much credence has been given to reconstructions based on computer matching of continental-shelf morphologies: the palaeomagnetic data provided insufficient resolution to be used exclusively for detailed intercontinental reconstructions. The geometrical fit of South America to Africa and Australia to Antarctica left little to question whereas the relationship of western Gondwanaland (Africa and South America) to eastern Gondwanaland (Australia, Antarctica and India) has long been a topic for discussion. The detail of that fit depended primarily on the pre-drift position of Madagascar. Recent palaeomagnetic data from Madagascar have provided the unequivocal result that the island was situated off East-Africa near Kenya and Tanzania. The predrift relationship of eastern and western Gondwanaland established previously on morphological grounds has now been verified from the palaeomagnetic data. Palaeomagnetic data for the Mesozoic of Australia have also long been considered anomalous within the framework of Gondwanaland. The data were inconsistent with any form of reconstruction from the geological and morphological viewpoints. However, a new investigation of Mesozoic rocks in Australia has resolved that anomaly and the existence of a single group of palaeomagnetic poles for the Triassic—Jurassic of all continents comprising Gondwanaland confirms that the Smith-Hallam reconstruction is as realistic a model as currently available. The Smith-Hallam model has been used to combine palaeomagnetic data from the Gondwana continents for the Late Palaeozoic and Mesozoic. The large amount of data for the Late Palaeozoic and Early Mesozoic of South America clearly implies that the supercontinent was drifting during that period and the data are best considered in ‘time groups’ that cross Period boundaries. The unity of Gondwanaland has been confirmed until mid-Jurassic time.