192 MAEION EXPEDITION TO DAVIS STRAIT AXD BAFFIX BAY 



estimated, are aiiiuially transft)rmed from solid to liquid and then 

 back to solid aoain in the northern Xorth Atlantic along a front 

 from Spitsbergen to Labrador, and a phenomenon of such mag- 

 nitude must bring into play a seesaw of enormous quantities of heat 

 energy. 



Several theories have been advanced, based on this effect of the 

 latent heat. Pettersson (1904). and (1927). as a result of laboratory 

 experiments with ice melting in a tank of salt water see fig. 84. p. 126) , 

 having reference to the distribution of temperature and salinity in 

 the wT)rkrs oceans, has presented a very fascinating theory explain- 

 ing (a) the major oceanic system of circulation, and (h) the source 

 of the great reservoir of cold bottom water. In order to test the 

 extent of the latent heat effect he placed a block of ice in a tank of 

 salt water and colored the water in such a manner that it would 

 clearly show any signs of circulation. Soon after the ice began to 

 melt three distinct movements were perceived, viz. an indraft of 

 relatively warm water in the mid-depth of the tank and two cold 

 outflows from the ice. one on the surface of the water and the other 

 toward the bottom. Applying this to the general circulation of 

 the North Atlantic (the tank representing the Atlantic Basin, its 

 contents the ocean, and the block of ice the real front of the melt- 

 ing ice stretching from Spitsbergen to Xewfoundland), he concludes 

 that the current toward the ice, at mid-depths, corresponds to the 

 warm layer which, dejiending on the total depth, is actually found 

 in most northern seas. The outward expansions of melting ice, sur- 

 face and bottom, correspond, resjiectively. to the cold icy surface 

 currents. Labrador and east Greenland, that flow into the Atlantic 

 and to the great reservoir of cold bottom water that fills most 

 of the North Atlantic. The fact that the circulation established in 

 the lal)oratory has so striking a counterpart in the longitudinal 

 distribution of warm and cold water in the Atlantic is j^roof for 

 Pettersson that the latent heat of melting ice is, (a) the main ener- 

 gizing agent responsil)le for North Atlantic circulation, and (h) the 

 chilling effect wliich initiates great downpourings of cold bottom 

 w^ater. 



Nansen (1909) (1912). on the contrary, is of the opinion, after 

 many years of careful observation at sea and in the polar regions, 

 that ice melting does not greatly affect the major processes of circu- 

 lation nor chill the su])ply to the great bottom masses of the ocean. 

 He believes : 



(a) The phenomena should be considered as action and reaction. 



(h) The processes are distinctly superficial when one considers 

 the relatiA'ely great average de]ith of the ocean. 



((")The stratified condition characteristic of the ocean prevents 

 the influence of melting ice from extending any significant field of 

 forces downward into the deeper strata. 



(d) The fact that ice in northern seas is usually separated from 

 the cold bottom water by a warm insulating layer precludes the 

 possibility of anv relationship between the ice and the bottom water. 



(e) The winter freezing is from top, downward; the summer 

 melting is in similar order. 



Finally Nansen (1909. p. 321) claims to have discovered the source 

 of the o-reat bottom water, not in the cooling effect of melting ice but 



