HYDROGRAPHY. 1 ^g 



To try at least whether it might be possible by a quick cooling to produce a layer of ice on 

 water of a salinity of 35 %o, before the whole body of water had been cooled to its freezing point, I 

 made some experiments for this purpose in the laboratory. A large glass-vessel was filled with sea- 

 water of a salinity of 35 %o, and above this a cup of metal containing a freezing mixture (snow and 

 salt) was suspended, and in such a manner that the bottom of the cup was just in contact with the 

 water in the vessel. It appeared then that ice commenced to be deposited on the bottom of the cup, 

 when the temperature at the bottom of the vessel had come down to o°. 



There are, however, in so far as this experiment is concerned, other conditions at hand for the 

 formation of ice than when it takes place in the sea. The bottom of the metal cup is constantly in 

 contact with the water in the vessel. I repeated therefore the experiment several times, but with the 

 modification that there was a little distance between the surface of the water and the bottom of 

 the cup with the freezing mixture. These experiments did not, however, give the results I desired. 

 The water was constantly cooled below the freezing point before the freezing commenced, and if a 

 piece of ice was not put in the water, the formation of ice commenced as a rule at the bottom. Even 

 if a small piece of ice was placed in the surface of the water, it was for all that cooled below the 

 freezing point at the bottom, and the formation of ice then manifested itself thereby that long crystal 

 needles or leaves from the piece of ice were shooting down through the water from the piece of ice at 

 the surface. When these ice-crystals had attained a certain size, they broke off and made their way 

 to the surface, where they were floating and became the starting points for new formations of ice. 



From this will be seen that the experiment does not show anything in the line of whether 

 formation of ice can take place in the ocean-water, as long as the whole body of water has not been 

 cooled down to its freezing-point. Even if such formation of ice be possible, it is for all that likely 

 to be supposed that the main body of the Polar-ice has been formed by freezing of water of very little 

 salinity, for water of a very small degree of salinity would, as we know, keep itself on the surface 

 when cooled to a temperature near the freezing-point. 



It has been stated on page 105 that the specific gravity of the water in the Atlantic varies 

 but very little in the same horizontal layer. The specific gravity of the basin-water in the Denmark 

 Strait is greater than that of the Atlantic, but still the difference must be regarded as small. It is 

 not till the transition from Atlantic-water to Polar-water takes place that differences of importance 

 appear, not only with respect to the magnitude of the specific gravity at equally large depths, but 

 also in so far as the distribution of the specific gravity at each single station is concerned. 



We shall now compare the distribution of the specific gravity for a station in Atlantic-water 

 with the distribution of the specific gravity for a station in the East Icelandic Polar Current, and for the 

 sake of simplicity we will avail ourselves of the values for j (^). The following will then manifest 

 itself: In the Atlantic s (-) is growing gradually from surface to bottom. In the Polar-water j (^) 

 is very small in the upper-layer, but larger in the under-layer than it was found to be anywhere else 

 in the Atlantic. In the ice-water proper s (|) appears at the surface — where the water in 



