The enigmatic palaeoecology and palaeobiogeography of the giant, horned, fossil turtles of Australasia: a review and reanalysis of the data

Abstract

The distribution pattern of the bizarre Australasian giant, horned, fossil turtles of the clade Meiolaniidae has puzzled biogeographers since their discovery late in the nineteenth century. While their distribution suggests a Gondwanan origin, the lack of fossil evidence from key times and places has inhibited a better understanding of their dispersal pathways to Australia and the south-west Pacific islands in which their fossils have been found. Much palaeoecological speculation related to their dispersal capabilities, ranging from purely terrestrial to freshwater, estuarine, and saltwater lifestyles, has been proposed to explain their enigmatic presence across a wide swath of Oceania. Various lines of fossil, anatomical and ecological evidence now strongly suggest a highly terrestrial lifestyle, and we believe these traits, reinforced by an abundance of marine predators and ever-widening saltwater gaps between land areas during the Late Mesozoic and Tertiary, minimise the importance of saltwater dispersal as an explanation for the observed meiolaniid distribution pattern. Here we propose that the fragmentation of Gondwana provided the main dispersal vehicle for the meiolaniids and that land connections were also used to access suitable habitats and expand their range. The recently recognised continent of Zealandia, along with Australia, South America, and probably Antarctica, transported all known meiolaniid turtles to their present locations. However, ice cover on Antarctica, and the nearly total submergence of Zealandia in essence preclude the current likelihood of fossil discovery in these critical locations. The islands of New Caledonia, Tiga (in the Loyalty Islands), Walpole, and Lord Howe served as refugia for Zealandia meiolaniids as the continent submerged.