SECTIONAL TRANSACTIONS.— C. 327 



the Coal Measures of Europe and North America, which must have been included in 

 a single zoogeographical province. The known Tetrapod fauna is clearly that of 

 the peculiar habitat presented by a coal-swamp. The direct continuation of this 

 fauna is known in Stephanian rocks in France and Czecho-Slovakia and in the Rothlie- 

 gende. In the latter horizon it is accompanied by rare stragglers from the dry land 

 in the persons of Pelycosaurian reptiles. 



These forms are far more abundantly represented in the Artinskian rocks laid 

 down under semi-arid conditions in Texas and New Mexico. Thus in Lower Permian 

 time Europe and North America still formed one province, and two facies of its 

 Tetrapod fauna are known. 



No land vertebrates are known from any rocks in North America between the 

 Artinskian and the Middle Trias. In Europe the copper-bearing sandstones of the 

 Urals and the Kupferscheifer, which are earlier than the Zechstein, yield reptiles. 

 The very incompletely known Ural fauna includes the amphibian Zygosaurus closely 

 allied to a Texan Artinskian animal and a series of reptiles which are morphologically 

 intermediate between the Pelycosaurs and the earliest South African Therapsids. 

 The Kupferscheifer reptiles are of unknown affinities. The only Upper Permian 

 reptiles known in the northern hemisphere are those from the Dvina and from Cutties 

 Hillock, Elgin. These faunas contain Pareiasaurs and DicjTiodonts, agreeing in evo- 

 lutionary stage with those from the Cisticephalus zone of South Africa. 



Thus it seems clear that the South African fauna of the Tapinocephalus zone, 

 which is associated with Olossopteris, is a derivative of that from the Copper Sandstones 

 of the Urals, which itself is a continuation of the dry land facies fauna of the Lower 

 Permian of Europe and North America. There is no evidence to show that in Upper 

 Permian times the Tetrapods of South Africa differed from those of the northern 

 hemisphere, and some evidence that it did not. 



Thus, from the standpoint of a student of Tetrapoda, there is no basis for a belief 

 in a Gondwanaland. 



Dr. A. L. DU ToiT. — The existence of Gondwanaland is proved by a wealth of 

 evidence, geological, palseoclimatical, palaeontological and zoological. It remains an 

 open question whether the several sections of the continental fragments were con- 

 nected by narrow land-bridges, or bj' broader connections of continental width ; or 

 whether they have drifted apart as advocated under the well-known displacement 

 hypothesis. 



The boundaries of Gondwanaland naturally varied at particular geological 

 epochs between the Devonian and the Jurassic, but, as seen on a globe, the southern 

 continent would have occupied an extent fuUy the size of the land areas of the old 

 world ; under the Displacement hypothesis this would be cut down to about the land 

 areas of the new world. 



Life in Gondwanaland shows its first important divergence from contemporary 

 life in the northern hemisphere just after the Lower Carboniferous, and was due to 

 the onset of glacial conditions in the south. The cosmopolitan plant life suffered 

 partial or complete extinction in the south, and was replaced by the Olossopteris- 

 Oangamopteris flora ; while in the northern hemisphere a break occurs between the 

 floras of the Lower and Upper Carboniferous. In the Lower Permian of parts of 

 Gondwanaland, however, relics of this northern flora occur. The Oangamopteris- 

 Olossopteris flora is not absolutely confined to Gondwanaland : migration into Angara- 

 land and Russia became possible in Upper Permian times. Early in the Triassic 

 plant life in Gondwanaland dvi-indled to some half-dozen genera, although there is no 

 sign that the environment was unsuitable to animal life. Diu-ing the late Triassic 

 immigration from Eurasia accounts for the strong resemblances between the land 

 faunas of both the northern and southern continents. Floral evolution also was 

 stimulated by this connection, in spite of a progressive tendency of the climate 

 towards aridity. Now appears the Thinnfeldia flora in Chili, Argentina, South Africa, 

 E. Australia, Tasmania and Tonkin. Many of the genera are of southern aspect, 

 while several are apparently confined to Gondwanaland. During the Jurassic period 

 exchange between the north and the south appears to have been freer. 



Outstanding lithological parallelisms occur in the Rhaetic of South America and 

 South Africa : each has an areally immense fine-grained sandstone formation, largely 

 aeolian, followed by vast outpourings of basalts and rhyolites, associated with wide- 

 spread intrusions of dolerite, which are found also in Tasmania and Antarctica. 

 Through the emission of these great volumes of magma the bonds connecting 

 Gondwanaland were weakened and its fragmentation dates from about the beginning 



