CHARACTER OF THE ASIATIC CHAINS. 89 



because of the tetrahedral depression, no sufficiently extensive land on 

 the south to oppose itself to the effect of the depressed Indian ocean 

 and of the continued elevation of the Manchurian shield — the tetrahe- 

 dral nucleus of this, the largest continent — the many parallel mountain 

 chains of southern Asia and the East Indies have, as it were, flowed 

 southward from this shield like a glacial ice-cap and ended like a lobed 

 ice-front, overflowing and wrapping in great curves around the resist- 

 ant foreland of India. We must assume the elevation of the Manchurian 

 shield on the one side and the sinking of the Indian ocean on the other 

 to have given the slope. 



We have considered but a small fraction of the chains which have 

 flowed out from this Asian center, and it does not seem possible that any 

 thrust from that area can have caused them. 



Since this paper was written I have read the abstract of an article by 

 Suess, in which he defines an ancient horst in the center of southern 

 Siberia bordered by a fault which runs along the Jenissei from Krasno- 

 jarsk north to the mouth of the Stony Tunguska and extends east to a 

 southwest-northeast fault or fold in the Jablonnoi mountains east of lake 

 Baikal. This forms the northwestern part of the Manchurian shield as 

 defined above. He considers it the center from which the Asian curved 

 chains from Saghalin to Java and on through the Himalaya to the Per- 

 sian gulf have flowed outwardly toward east, south, and west, accepting 

 the hypothesis of Keyer. In some way not made clear, North America, 

 and especially the magnetic north pole, is made the center toward which 

 a complementary inflow has taken place, across where is now the Pacific, 

 and Suess suspects that these curved Asian chains." stand in some kind 

 of relation to an outflow of superfluous earth-mass from the pole — 

 that is, with a flattening of the same."* 



In the Banda sea the southward curvature changes to a northward, 

 causing one of the deepest spots in the ocean bottom, and the bending 

 is then to the north away from the old land of Australia, and this north- 

 ward curve is repeated in many island chains on the north to the Mar- 

 shall archipelago. This curve is continued from the mountain range of 

 New Guinea to New Hebrides and New Caledonia. Here the line leaves 

 the continental area with its curvilinear island chains, and enters upon 

 the ocean bottom area with its islands in detached groups, and from 

 Fiji to Tahiti follows the line of volcanic archipelagoes. Off to the north 

 lies the coral island area, separated by Dana's line, and the Kermadoc- 

 New Zealand-Macquari band branches off to the south, as does the 



* See appendix to this paper, p. 96. 



