I.] CONTINENTS AND OCEAN-BASINS. 23 



oceanic islands are either of volcanic or coral origin, and do not 

 contain sedimentary rocks; and also that deposits analogous to 

 those laid down in the deepest ocean beds are generally wanting 

 from among the sedimentary series of rocks of which the continents 

 and islands are composed. A further argument was afforded by 

 the discovery that the greater portion of peninsular India and 

 South Africa has been dry land since the Palaeozoic epoch. 



As is so commonly the case in similar instances, the promulga- 

 tors of the doctrine of the permanency of continents and ocean- 

 basins pressed their hypothesis too far; and it is now evident 

 that although the doctrine is true as a whole, and more especially 

 as regards the later stages of the earth's history, yet it requires 

 very considerable modification from the original form in which it 

 was advanced. In the first place, it has been shown that crystal- 

 line granitic and gneissic rocks occur in the Seychelles, which were 

 formerly regarded as true oceanic islands ; and, secondly, deep-sea 

 deposits have been discovered in the West Indies and the Solomon 

 Islands. Moreover, various lines of evidence indicate that during 

 the Jurassic and Cretaceous epochs there was a continuous land- 

 connection between Africa (by way of Madagascar and the 

 Seychelles) and India; while at some time in the Secondary era, 

 in the opinion of Drs Neumayr and Blanford, South America and 

 South Africa were in communication across the South Atlantic. 

 The latter connection appears, indeed, to have been a survival 

 from an older Palaeozoic girdle of land which, from the evidence of 

 fossil floras, seems to have existed in low latitudes round nearly 

 three-quarters the circumference of the globe, and which was cut 

 oft" from the land to the north. There is, moreover, the possibility 

 of a Tertiary connection of Australia with Patagonia by way of 

 Polynesia, to which allusion is made in the third chapter. Then, 

 again, the recent investigations of Dr J. W. Gregory 1 on the fossil 

 corals of the West Indies have afforded strong support to the view 

 that the Atlantic is of comparatively recent origin. After referring 

 to the remarkable resemblance between the existing fauna of the 

 West Indian seas and that of the Miocene deposits of the Mediter- 

 ranean basin, Dr Gregory 2 observes that the sea-urchins, or 



1 Quart. Joitrn. Geol. Soc. vol. LI. pp. 255 312 (1895). 



2 Ibid., pp. 306, 307. 



