Rifting and drift of australia and the migration of mammals.

Recently several investigators (1-5) have discussed the immigration of the primitive mammals (monotremes and marsupials) to Australia. The oldest fossils that have been found are probably Early Miocene in age (6). Interest in the subject has been stimulated by the profound changes in the thinking of geologists that have occurred during the past decade on the old problem of a postulated supercontinent, Gondwanaland, and on a possible timetable for its breakup and the dispersal of the fragments to their present geographical positions. Perhaps we have reached the stage where "it is therefore no longer profitable for biologists to speculate about the past arrangements of land masses" (2); at any rate, there is a widespread (although not unanimous) feeling on the part of geologists that we should now concentrate our attention on the problems-especially the timing -of a southern mammalian migration route, that is, via Antarctica, rather than from Asia via Indonesia (7), or even from North America across a postulated former landmass in the present Pacific Ocean (8). In this technical comment I discuss stratigraphic evidence from Australia bearing on the geographic relationship of Antarctica to Australia. I assume for the purpose that mammalian migration was by that route. My conclusion is that the geology of the Australia/Antarctica suture gives no evidence of any significant barrier to the passage of land animals during the time from the Early Permian (270 X 1 06 years ago) to the Early Eocene (49 X 106 years ago). There is some evidence that incipient rifting already had occurred by the Early Permian, and strong evidence that active drifting (separation of the two continents) began in the Late Paleocene or Early Eocene. The distinction between rifting and drifting is important. It follows from this conclusion that geology's contribution to the historical biogeography of southern mammals (apart, of course, from the dating of fossil finds) will consist primarily of the examination of the history of other continental connections, especially the Scotia Arc between Antarctica and South America; those problems are outside the scope of this comment. When Australia is fitted against Antarctica, there seems to be a continuity in a gross sense within the tectonic belts (geosynclines) of the Late Precambrian and the Paleozoic (9). The first evidence of incipient sutural rifting might be found in the first signs of geological trends cutting across those broadly north-south trends. One generally accepted piece of evidence is the presence of volcanic rocks in Tasmania and Antarctica of Jurassic age, which date from about 160 X 106 to 170 X 106 years ago (10). However, there are signs of cross-cutting geological trends which suggest that the formation of a continental suture in the position of the future Southern Ocean occurred in the Late Paleozoic. Wopfner (11, 12) has suggested that Permian sedimentation, including tills and other evidence for glaciation, was controlled significantly by tectonic trends which began in the Devonian, Early Permian ("Sakmarian") sediments with marine-type assemblages of microfossils occur widely in southern Australia, being known from the Denman, Arckaringa, and Troubridge basins and the Renmark Trough (Fig. 1) (13). These microfossils consist almost entirely of organicwalled phytoplankton and benthonic foraminifera with agglutinated tests (14). The stratigraphic sequence as presently