.8^. THE PERMANENCE OF CONTINENTS. 741 



of the assumed zoological and botanical migrations, though not all, 

 might have taken place equally well by either route, but the 

 geological evidence appears to me only consistent with the former 

 existence of a belt of land across what is now in places deep sea. 



This geological evidence is, briefly : (i) That South Africa, India, 

 and Australia, in late Palaeozoic and early Mesozoic times, formed 

 parts of one continental area, and that this area was separated by a 

 wide sea, probably of considerable depth, from Northei;n Asia, 

 Europe, and North America, which were also connected by land ; 

 (2) that in Cretaceous times part of Western India, a little north of 

 Bombay, and also the South Coast of Arabia, were part of a sea that 

 extended over a wide area in South-western Asia and in Europe, and 

 that the Khasi Hills in North-eastern India, Trichinopoly, south of 

 Madras, and Natal, in South Africa, were on the shores of another 

 sea, divided from the first by a land barrier ; (3) that, as shown by 

 Neumayr, there are indications of this same land barrier in Jurassic 

 and Neocomian times. I do not see how these facts can be explained 

 by land connections within the present 1,000-fathom limit, and as it 

 has been shown that a depression of part of the old barrier, the 

 Mozambique Channel, to a depth of upwards of i ,000 fathoms, has 

 taken place within comparatively recent times, almost certainly since 

 the Miocene, and probably since the Pliocene period, it is not impro- 

 bable that another part of the old barrier may have undergone depres- 

 sion to a greater depth, even to 2,000 or 3,000 fathoms in places, since 

 the Cretaceous or Eocene. It must not be forgotten that the con- 

 necting barrier, which I believe to have once existed, is represented 

 by a tract of ocean still dotted over with islands and shoals ; that the 

 soundings are far from sufficient to show the true contour of the 

 ocean basin ; and that there is, so far as I am aware, no geological or 

 biological indication of probable land connection in Palaeozoic or 

 Mesozoic times in the direction preferred by Dr. Wallace. 



The other point is of no great importance. Dr. Wallace says 

 that my suggestion of a possible Mesozoic girdle of land, chiefly in 

 low latitudes, from Peru to New Zealand and the Fiji Islands is in- 

 consistent with the permanence of the great ocean basin of the 

 Pacific. He must, I think, have OAerlooked the circumstance that 

 the supposed girdle in question was explained as passing through 

 Australia, India, Madagascar and Africa, and not across the Pacific 

 Ocean [Proc. Geol. Soc. 1890, p. 106). 



The real difference between Dr. Wallace's views and mine is, I 

 think, this, that Dr. Wallace regards ocean permanence as an estab- 

 lished law, only to be disregarded on the clearest evidence, while I 

 look upon it as a theory supported by some important and valuable 

 data, but by no means proved ; and I contend that there is ample 

 proof that even if the law of perm-anence prevails it is not universal. 

 As a result of this, I think that every questionable case must be judged 

 upon its merits, and if the evidence tends to show that land formerly 



