DISCOVERY 



117 



pelago owing to the present proximity of Australia 

 to that region. 1 This class has its nearest relationships 

 in South America, now separated from it by a whole 

 quadrant of the earth. The third group, finally, is 

 the fauna of the eastern Sunda islands, which is found 

 in New Guinea and which has established itself in north- 

 eastern Australia. This relationship, formerly so puz- 

 zling, is completely explained by the displacement 

 theory. As Fig. i shows, Australia, up to the beginning 

 of Jurassic times, was connected in the west with India 

 and Ceylon, and through them with Madagascar and 

 South Africa. This explains the ancient " Gondwanic " 

 group. After breaking away from India, it was still 

 connected through Antarctica with South America, 

 perhaps as late as Eocene times, and this connection 

 gave rise to the second group. Comparatively recently 



Carboniferous ice age in the southern hemisphere. 

 Traces of inland ice at this period are found in Brazil, 

 the Argentine, the Falkland Islands, Togo Island, the 

 Congo, South Africa, India, Western, Central and 

 Eastern Australia. These traces are to-day so widely 

 separated from each other that they cover nearly a 

 complete hemisphere, and even if the pole be placed in 

 the most favourable position, the traces of ice most 

 distant from it would be in a geographical latitude of 

 only 15° and so be in the tropics. On the other hand, 

 we do not know of anj- certain traces of ice in this epoch 

 in the other hemisphere. This fact has so far consti- 

 tuted a hopeless riddle, and it is no exaggeration to say 

 that it has completely' crippled the development of 

 palaoclimatology. The displacement theory affords 

 a striking solution of the riddle ; at that period all these 



FIG. 3— THE WORI.D IN THE CARBONIFEROUS PERIOD, SHOWING THE POLES. 

 Diagonal Hues denote deep sea ; horizontal lines shallow water: unshaded portions land. 



Australia drifted into collision with the Sunda islands, 

 with the consequence that an interchange of flora and 

 fauna took place. 



In seeking an explanation of former climatic condi- 

 tions, geologists have hitherto been averse to the 

 assumption of large movements of the earth's poles 

 with reference to the land. However, the idea that it 

 is necessary to assume a considerable movement of the 

 poles in early Tertiary times has recently been gaining 

 more and more ground. It is impossible to overlook 

 the fact that all former attempts to map out the 

 position of the poles throughout the earth's history 

 come to grief on one obstacle, namely, the Permo- 



" The dividing line, which defines the extent to which this 

 group has penetrated, has been drawn by Wallace between the 

 islands of Bale and Lombok. 



continents were grouped concentrically around South 

 Africa, and we thus obtain a connected ice cap of no 

 greater area than that of the quaternary ice-age of 

 America and Europe. 



Similar, if less striking, simplifications appear when 

 the position of the pole in other geological periods is 

 determined by aid of the displacement theory, and 

 it is not too much to say that this theory makes it 

 possible, for the first time, to determine the former 

 positions of the pole from fossil evidences of climate in 

 a manner that is satisfactory. 



Finally, the displacement theory may be tested by 

 astronomical determinations of latitude and longitude. 

 It is natural to suppose that the movements are still 

 taking place, and the available estimations of geological 

 time, in spite of their uncertainty, allow us to make an 



