GEOLOGICAL MUTATIONS. 29 



sinking of all around. The fringe of tertiary beds on the opposite coast of Afiica must have been 

 deposited subsequently to the submergence, and, therefore, the coast of South Africa (and as the fringe 

 extends on both east and west shores, the whole of South Africa, and Madagascar with it) must have 

 suffered, first, a depression sufficient to allow of these tertiary beds being deposited, and then a 

 subsequent elevation so as to bring these beds to view. These tertiaries are referred to the upjoer 

 miocene epoch ; the subsidence, therefore, must have taken place not later than the middle of that 

 epoch. 



The extent of land then submerged, wo cannot specify ; but we may safely infer that a great con- 

 tinent stretched across between Africa and India. The numerous shoals in the Indian Ocean is one 

 indication of this, but a much more important one is the fact of the fixmia of India and Africa, 

 belonging, with few exceptions, to the same families, and these families which are peculiar to 

 those two districts. So far as regards Mammals, abundant illustrations in support of this will 

 be foimd throughout the following pages, passim. This Africano-Indian continent was boimdcd 

 on the north by the Saharan Sea, and by the sea which appears from the nummulitic eocene beds 

 in Arabia, Persia, Beloochistan, and more modern tertiaries in the north of India, to have then 

 covered these countries. 



On still stronger grounds the existence of a large tract of land, where the North Atlantic now 

 rolls, has been inferred. This is what is known as the Miocene Atlantis. 



