134 FORMER LAND CONNECTIONS. 



fauna on the north and west of Peninsular India has on the con- 

 trary a marked European facies. These contrasting faunas indi- 

 cate that Madagascar was still joined to India and must have 

 formed a peninsula at one point at least quiTe narrow, as proved 

 by the deposits of the Narbada Valley. 



The Breaking up of Gondwanaland, 



There is no doubt that Madagascar was severed from Africa 

 shortly afterwards, though there is reason to believe that the strait 

 was quite a narrow one, but on the Indian side the link seems to 

 have remained unbroken until well into tbe Tertiary. 



The general instability of the earth's crust in the Middle 

 Cretaceous is reflected in the further folding of the southern part 

 of South Africa, followed by severe fracturing of the coast-belt, 

 the great faults, which run parallel to the folds, having down- 

 throws on the seaward side of as much as from one to over two 

 miles. 



In addition a great bending of the crust took place along the 

 line following the Lebombo Range, continuing through Eastern 

 Natal into Pondoland. To the west thereof elevation occurred, 

 and the Karroo beds are found at great altitudes, while to the 

 east they are bent down so a? to dip below sea-level. Along with 

 the severe flexuring this coastal strip became broken through by 

 faults, which usually trend north-eastwards, towards Madagascar, 

 so that the separation of that island was doubtless accomplished 

 at this particular time, namely, during the middle of the 

 Cretaceous. 



Contemporaneous igneous activity is evinced by the flooding 

 of the Deccan in India by basalt outpourings of enormous extent, 

 while both Northern India and Western South America were 

 crumpled up and the strata thus folded injected with igneous 

 matter on a vast scale. In Africa the crust was perforated by 

 numerous volcanic pipes into which was squeezed from below the 

 peculiar blue-ground (kimberlite) in which the diamond occurs. 



Development of the Atlantic Basin. 



In places the earth movements were so considerable, even 

 during the Tertiary, that one cannot draw any reliable conclusions 

 as to the former extension of the land-masses merely from a study 

 of the shapes and depths of the present ocean basins; still it will 

 be profitable to note several features that they exhibit. 



In the Indian Ocean the only peculiarity is an irregular ridge 

 from 1 to 1\ miles deep linking South Africa with India via Mada- 

 gascar, the Seychelles, Chagos, Maldive and Laccadive Islands. 

 Mauritius, Bourbon, and Kerguelen are recent volcanic excre- 

 scences upon the floor. 



In the North and South Atlantic, however, the most extraor- 

 dinary feature is the submarine ridge covered by only from one and 

 a third to two miles of water extending down their entire length 

 and always nearly midway between the shores on either side. It 

 forms a remarkably strong piece of evidence in support of the 



