Late Cretaceous bioconnections between Indo-Madagascar and Antarctica: refutation of the Gunnerus Ridge causeway hypothesis

Abstract

Aim  To evaluate the Gunnerus Ridge land-bridge hypothesis, which postulates a Late Cretaceous causeway between eastern Antarctica and southern Madagascar allowing the passage of terrestrial vertebrates. Location  Eastern Antarctica, southern Indian Ocean, Madagascar. Methods  The review involves palaeogeographical modelling, which draws upon geological and geophysical data, bathymetric charts, and plate tectonic reconstructions, and the evaluation of stratigraphically calibrated phylogenetic analyses to document ghost lineages of select taxa. Results  The available geological and geophysical evidence indicates that eastern Antarctica’s Gunnerus Ridge and southern Madagascar were separated for the entire Late Cretaceous by a vast marine expanse. In the mid–Late Cretaceous, the gap was probably punctuated by land on two intervening physiographical highs, the northern Madagascar Plateau and Conrad Rise, the latter of which, although probably large, was still separated from Antarctica’s Riiser-Larsen Peninsula by c. 1600 km. Recent, stratigraphically calibrated phylogenies including large, terrestrial end-Cretaceous vertebrate taxa of Madagascar and the Indian subcontinent reveal long ghost lineages that extended into the Early Cretaceous. Main conclusions  The view that Antarctica and Madagascar were connected by a long causeway between the Gunnerus Ridge and southern Madagascar in the Late Cretaceous, and that terrestrial vertebrates were able to colonize new frontiers using this physiographical feature, is almost certainly incorrect, as was previously demonstrated for the purported causeway between Antarctica and the Indian subcontinent across the Kerguelen Plateau. Connection across mainland Africa to account for the close relationships of several fossil and extant vertebrate taxa of Indo-Madagascar and South America is another option, although this too lacks credibility. We conclude that (1) throughout the Late Cretaceous there was no intervening, continuous causeway through Antarctica and associated land bridges between South America to the west and Indo-Madagascar to the east; and (2) mid- to large-sized, obligate terrestrial forms (e.g. abelisauroid theropod and titanosaurian sauropod dinosaurs and notosuchian crocodyliforms) gained broad distribution across Gondwanan land masses prior to fragmentation and were isolated on Indo-Madagascar before the end of the Early Cretaceous.