OF THE KERGUELEN REGION OF THE GREAT SOUTHERN OCEAN. 495 



witli ice and siiow, the glaciers desceuding from the laud would destroy the shore and 

 shallow- water fauna, especially those with pelagic larvae, while those with a direct 

 development would be able to take refuge in the deeper water ; migration from the 

 Antarctic continent towards the equator would thus set in all over the sea-bed in the 

 Great Southern Ocean, and the other physical conditions of the area, such as the mixture 

 of currents of different temperatures and the consequent destruction of pelagic organisms, 

 would combine to furnish this deep-sea fauna with abundance of food and oxygen.' In 

 this way we may account for the apparently greater abundance of life in the very deep 

 sea of the Great Southern Ocean and the North Atlantic and the North Pacific than 

 elsewhere. 



Some geologists have cited the appearances presented by certain ancient conglomer- 

 ates as evidence for the existence of glacial periods during Palaeozoic times,^ but this 

 hypothesis seems to be in direct opposition to the testimony furnished by the coral reefs 

 preserved in Palajozoic strata within the Arctic circle — indeed, coral reefs apparently 

 flourished in all Palaeozoic seas. The vea;etation of the Carboniferous formations likewise 

 indicates a universal tropical climate all over the world both on the land and in the 

 ocean. Palfeontological evidence points to the gradual diminution of temperature in the 

 higher latitudes of the globe during the later geological periods. AVhen considering 

 the climates of the past it is not so much an excess of light and heat in ancient times as 

 a nearly uniform distribution of tropical conditions over the whole globe, and a pro- 

 gressive withdrawal of the then universal torrid zone within its existing limits, that 

 require explanation. 



In seeking for a solution of the problems connected with the geographical distribution 

 of fossils in the geological strata, as well as of those connected with the distribution of 

 existing species over the face of the earth, it has been almost universally assumed that the 

 astronomical relations of our globe have remained stable. This may well be accepted as 

 true for the period covered by human history, but the variation in these relations assumes 

 a very great importance when dealing with the immense duration of geological history. 

 It seems certain that the Palasozoic trilobite looked out on a very different sun from what 

 we now behold in the heavens. In picturing the successive stages in the evolution of the 

 surface features and biological conditions of our globe it is necessary to take into considera- 

 tion the contemporary evolution of the other members of the solar system, and especially 

 that of the central luminary — the sun.^ 



The nebular theory of the formation of the solar system is now almost universally 

 accepted. It starts from a plenum filled with absolutely cold matter, the potential energy 

 of which is at a maximum, and with a rotatory motion at least equal to that of the whole 



' See ante, p. 352. 



2 See A. Geikie, Text-book of Gcoloijij, ed. 3, p. 802, London, 1893 ; A. de Lapparent, Traite de G^ologie, ed. 3, 

 p. 884, Paris, 1893 ; A. C. Ramsay, Quart. Journ. Ocol. Sac, 1855, p. 185 ; M. Nedjiavr, Erdtjeschichte, Bd. 2, p. 193, 

 Leix-izig and Wien, 1890. 



^ See EuG. Dubois, Tlie Climates of the Geological Past and their relation to the EvohUion of the Sun, London, 

 1895. 



VOL. XXXVIII. PART II. (iVO. H')- 3 U 



