126 FORMER LAND CONNECTIONS. 



on these lines the geologist is confronted by several difficulties, the 

 most formidable of which — one that has hitherto been insuperable — 

 arising out of the extraordinary location of the ice-centres, namely, 

 in the temperate girdle, and their peculiar attitudes to one another. 



For explanation I am advancing in all seriousness the view, 

 revolutionary and heretical as it will appear to orthodox geologists, 

 that Gondwanaland was a much smaller continent than as usually 

 conceived, that its centre lay somewhat further to the south, that 

 the Carboniferous ice-sheet was an almost continuous mass, and 

 that the land fragments' still preserved represent portions of the 

 ancient continent forcibly torn apart, subsequently modified in 

 outline by erosion, deposition, etc., and now separated by vast 

 stretches of ocean. 



The customary view is that these stretches of water have 

 developed as areas of down-warping, the main outlines of the lands 

 having been determined by extensive faults and modified by sub- 

 sequent marine erosion. 



That the continents might have originated by the actual tear- 

 ing apart of one or more much larger masses is no new doctrine. 

 Although generally dismissed as fantastic, it has been very ably 

 championed recently by Wegener, and, when the hypothesis is 

 studied in detail, the evidence in its support is found to mount up 

 so remarkably as to become almost overwhelming; only a few of 

 the arguments in its favour can be presented, however. 



It will forthwith be realised that by thus supposing the several 

 units to have been spaced much closer together in the past, the 

 numerous remarkable lithological and palaeontological resemblances 

 between them become more explicable, while the difficulties that 

 beset migration become much reduced. Moreover, the areas known 

 to have been capped by ice become roughly grouped around the 

 South Pole, not far from, if not well within, the present northern 

 limits of drift-ice, and a serious stumbling block to the interpreta- 

 tion of the Carboniferous Ice Age is thereby removed. (See Fig. 1.) 



This hypothesis indeed becomes the key to the understanding 

 of the past, and during the rest of this discourse it is this reduced 

 and modified conception of Gondwanaland that will be referred to. 



The Deposits of Gondwanaland. 



With the melting of the ice-sheets the borders of the lands 

 were submerged and a series of strata laid down varying in places 

 from marine through estuarine and lacustrine to continental types 

 over very considerable portions of the area and to a thickness of 

 many thousands of feet occasionally. 



During this lengthy period the geographical and climatic con- 

 ditions changed repeatedly over the entire continent, as would be 

 expected. In South Africa land prevailed in the north of the 

 Union, and the muds and sands washed down therefrom collected 

 in an ever-deepening trough in the south. This latter probably 

 formed an immense extent of flats, now flooded, now dried up or 

 dotted with lakes and pans; entombed in the silts thus laid down 

 over them are the fossil remains of animal and plant life. This 



