568 H. MOHN. METEOROLOGY. [NORW. POL. EXP. 



the depth to which the annual periodic variation of the temperature of the 

 surface could extend its influence. The colder under surface of the ice causes 

 the water in contact with it to freeze, and the water lower down to reach 

 its annual minimum of temperature in winter or spring. 



From January to March, the surface minimum of temperature is travelling 

 downwards through the ice. 



In spring the temperature of the surface increases, owing to the fact 

 that the absorption of heat from the radiation of the sun is greater than the 

 loss of heat by radiation and evaporation. A wave of heat is propagated 

 downwards, and raises the temperature of the ice. At the same time the 

 current from the water sends heat from below. The retardation of the wave 

 from the surface causes the isopleths to take a more vertical direction, the 

 vertical thermal gradient diminishes, and the effect of the current from below, 

 unlike that of the autumn, is to cause a decrease. The rise of the tempera- 

 ture in spring almost corresponds with the fall in the autumn. 



At the end of spring or the beginning of summer, the temperature is 

 nearly the same, about 3'5, at all depths of the ice. 



During summer the higher temperature of the air and the surface sends 

 a heat-wave downwards, raising the temperature in the uppermost strata to 

 zero, and continuing its course, with a certain amount of lag, to the deeper 

 layers, where the temperature, between and 1, becomes higher than that 

 of the underlying water, and contributes to raise the temperature of the 

 water to its summer or autumn maximum. In August the temperature seems 

 to be highest at the depth of 1*2 metres, and to decrease from this depth 

 upwards and downwards. In September the temperature steadily increases 

 with the depth, but decreases on the whole in all depths. At the end of 

 October the lower surface of the ice is cooled down to the temperature of 

 the water. 



The reason why the temperature of the ice does not accord with the 

 theory for the temperature of the soil is, as we have seen, to be found in 

 the fact that the ice has not a sufficient thickness, and that there is an almost 

 constant source of heat beneath it in the polar water on which it floats. The 

 ice-covering protects the water from losing heat by radiation and evaporation. 

 The conversion of water into ice on the under surface of the ice goes on at 

 a slow rate, the latent heat given out in freezing being considerable, but every 



