ANNIVEESARY ADDRESS OF THE PRESIDENT. 99 



It is perfectly true, however, that the charts hitherto published, 

 for instance, that accompanying the ' Challenger ' narrative, show 

 deep water between the various banks that support the Mascarene 

 Islands and the Laccadives, Maldives, Chagos, and other groups. 

 But the soundings in the portion of the Indian Ocean between these 

 islands are insuf&cient to enable the contours of the sea-bottom to 

 be laid down with any approach to acccuracy ; and I anticipate 

 that, when the contours are better known, a bank will be found to 

 connect the whole series from India to Madagascar. Even should 

 this not be the case, the evidence of land-connexion appears so 

 strong that it may be a question whether the whole of the ocean- 

 bottom between Africa and India may not have sunk to its present 

 depth since Cretaceous times. 



5. South Africa and South America. — The only other hypothesis 

 to which I shall ask your attention is that of an ancient southern 

 continent, and especially the possibility of ancient land-connexion 

 between South America and Africa on the one hand and between 

 South America and Australia or l^ew Zealand on the other. The 

 latter, if it ever existed, must have been, I think, the later of the two, 

 and I will give the biological evidence in its favour first. The 

 most interesting relations are those of freshwater fishes, the peculiar 

 distribution of which has already been noticed. Two families, 

 Haplochitonidce and GaJaxiadce *, are found only in the southern 

 extremity of America, New Zealand, and Tasmania with Southern 

 Australia, and they form a considerable proportion of the small river- 

 fish-fauna of those countries. 



There are some well-marked alliances between the frogs and 

 tortoises of Australia and those of South America. The batrachian 

 family Oystignathidce and the chelonian family Chelydidoi are 

 restricted to the two areas ; but on the other hand no tortoises are 

 found in New Zealand, and the only frog occurring there is a 

 member of a family otherwise confined to the Palaearctic region. 

 Moreover fossil representatives of Chelydidoe have been found both 

 in Europe and in India, so that it is not improbable that the Cysti- 

 gnathidce, which are not very high forms, may also have once had a 

 more extensive range. The land- and freshwater shells, too, afford 

 but little evidence of connexion. If, as Wallace has suggested, the 

 New-Zealand, Tasmanian and Patagonian freshwater fishes or their 

 ova can have been transported by floating ice from the Antarctic 



* A Galaxias has been describad froin the Indian coast, but the determina- 

 tion appears somewhat doubtful. 



