The Bocks of Tristan d'Acunha. 21 



of the space between this and the comparatively shoal water that 

 runs through Fernanda Noronha and St. Paul's Kocks to the coast 

 of Sierra Leone, as the evidence of the Devonian fossils seems to 

 indicate, or whether it bent more to the south, following the line 

 of the Falkland Islands, South Georgia, and the Sandwich group, 

 then coming up through Bouvet Island to the Cape. In both 

 directions it must have crossed some of the great depths of the 

 ocean ; the southern one would be more to the liking of those who 

 still believe in the permanence of ocean basins, but a greater abyss 

 would have to be crossed, whereas the more northerly alternative 

 is still dotted with banks and shoals. The point I wish to bring out 

 is that it is extremely probable that beneath the sea in the South 

 Atlantic there is an old continent sunk beneath the waters, and 

 containing all the various rocks such as granites, gneisses, old 

 schists, and quartzites. 



The mean excess of vibrations of the seconds pendulum on stations 

 in oceanic islands is 5'26 vibrations a day ; in the Falkland Islands, 

 where the rocks are composed of the ordinary materials of continents, 

 there is a defect of 3*85 vibrations, and this would seem to indicate 

 clearly that there is some fundamental difference in the rocks 

 underlying islands like the F'alkland and Comoro Islands and the 

 Seychelles, which contain continental materials, and those of purely 

 volcanic origin. 



In the "United States Coast and Geodetic Survey" there is an 

 account of the measurement of gravity at the base of the volcano 

 Fujisan in Japan. The cone is 12,000 feet high, which was said foy 

 have been thrown up in a single night in the year 300 B.C. The 

 mountain was found to attract the pendulum exactly as a 

 mass of the same size would do if it had been carted thither and 

 piled up.' 1 ' A truly oceanic island is just such a cone as Fujisan, 

 only its base lies so many fathoms deep beneath the surface of the 

 water, and it seems very probable that the mass of volcanic ejacta- 

 menta is sufficiently great, compared with the mass of the pendulum 

 bob, to obscure any effects that might be caused by the density of 

 the floor on which the lavas are piled up. I do not, therefore, think 

 that the revelations of the pendulum need necessarily preclude the 

 existence of an ocean floor made of the same materials as continents. 



Sir John Murray, however, some years ago stated: " There 

 has not been found in the abysmal areas any land made up of 

 gneisses, schists, sandstones, or compact limestones ; nor have 

 fragments of these sedimentary formations been found in the 



* " United States Coast and Geodetic Survey," appendix 22, p. 507, Washington,. 

 1883 ; O. Fisher, " Physics of the Earth's Crust," 1889, p. 251. 



