MISCELLANEOUS PAPERS. 405 



For, if the water continued indefinitely to shrink on parting with 

 its heat, and to shrink still further on being solidified, as lard and very 

 many elements and compounds do, all bodies of water must freeze up 

 from the bottom and continue to freeze upward to a level within the 

 influence of the sun's radiation. Whether this freezing would extend 

 to all bodies of water underground suggests an appalling thought. 



At the present time the temperature of the water in the great deeps 

 of the Atlantic, South Pacific and Indian oceans is 39.2°, which is 

 water of the greatest density and gravity. The temperature cannot 

 get lower than that without the water at the surface first freezing. If 

 the temperature of the water in the great deeps of the North Pacific 

 is not so low as 39.2° throughout the entire year, it is because there is 

 not sufficient arctic current through Bering strait to cool the entire 

 North Pacific ; the antarctic currents have too far to travel, and a great 

 warm stream, the kuroshiico or black current of Japan, passes directly 

 over Kurile Deep, the deepest part of the North Pacific. 



And if there were a zone in the ocean where the water should be 

 39.2° F. at the surface, then all the water along that zone, from surface 

 to bottom, as a wall, would be 39.2". This is impossible, for then 

 along such a zone the water would be heavier than either north or 

 south of it. It would immediately sink, and the lighter waters from 

 both north and south would flow into its place. 



Yet this is precisely what does happen, not as a wall, but as an 

 anticlinal ridge with a gently-falling surface, along a zone from Cape 

 Breton, northeasterly past Newfoundland and Iceland, to the north- 

 ern coast of Norway ; from Sitka, along the Aleutian and Kurile is- 

 lands, to Yezo; and in the southern ocean all around the earth in the 

 latitude of Cape Horn Both of those zones are farther northward in 

 July, farther southward in January. They are the zones of frequent 

 storms. Along those zones the lighter waters of 40"^ F. and upward 

 from the equatorial regions meet and commingle with the lighter 

 waters of 38° and less from the polar regions, becoming an average 

 temperature of 39° or thereabouts. Then, by reason of the increased 

 density and gravity, the waters sink, making room for new waters to 

 flow in from both directions, thus keeping up constant, wide-spread 

 surface currents toward the zones. This is the undoubted cause of 

 both the Gulf stream and the kui'oshiwo, and the principal cause of 

 all oceanic currents. 



TEMPERATURE OF WATER IN OCEANS. 



At the bottom, along those zones of greatest density, the water 

 spreads out over the floor of the ocean and flows (1) toward the poles, 

 warming the waters there, melting the ice on its under surface, and 

 displacing the colder, lighter water, which in turn flows back, loose 



