90 TEMPEKATURE. 



We now come to one of the most characteristic differences between the temperature in 

 the north and the south. From April to August in the south, and from October to February 

 in the north, no insolation is received. The temperature in the south during these months 

 is nearly stationary, the fall being only 4°F. on the Barrier, while in the north the tempera- 

 ture continues to fall, and February is 24 9°F. colder than October. Here again we have an 

 effect of the different local surroundings. The Barrier is only a few miles from a sea, the 

 temperature of which, during the winter, is, as we have already shown, 25°F. warmer. Over 

 the Ross Sea, there are vertical convexion current.'', while over the Barrier there are none, 

 therefore, the temperature of the sea governs the temperature of the upper atmosphere. Every 

 blizzard which blows removes the cold air from the Barrier surface, and raises the temperature 

 to that in adiabatic equilibrium with the warm upper air. The fall of temperature during 

 the winter over the Barrier is, therefore, retarded, and the temperature remains nearly constant. 

 In the north, the land which was warm in the summer is relatively cold in the winter, 

 therefore, there is no great diti'erence in temperature over a short horizontal distance, and the 

 upper air is likely to be much colder than the upper air over the Barrier. The temperature 

 over the greater part of the north polar regions falls at the same rate, and, therefore, there 

 is no local source of heat after insolation ceases, hence until insolation again commences 

 after the winter, the temperature continues to fall. 



The curves for the yearly variation of temperature at the few places in the Ross Sea 

 area given on figure 28, page 84, become of great interest in the light of the previous 

 discussion. 



December is the warmest month at the Barrier, Framheim and McMurdo Sound, showing 

 that at these stations the temperature follows the insolation very closely. At Cape Adare, 

 however, the maximum temperature occurs in January, but Gape Adare is much more affected 

 by open water than the other stations, and, therefore, its temperature cannot follow the insola- 

 tion so closely. The effect of open water is clearly seen in the rates at which the tempera- 

 ture falls after the summer maximum. The Barrier is furthest away from open water, and its 

 temperature falls the most rapidly. Framheim is so situated in the south-east angle of the 

 Ross Sea that it has the Barrier on three sides and open sea only on the fourth ; the tem- 

 perature fall there is, therefore, not so rapid as on the Barrier, but much more rapid than 

 in the McMurdo Sound, where the sea is open in most years until the end of March. Cape 

 Adare is almost entirely surrounded by open water until May or .June, and the lag of tem- 

 perature is greatest at that station. 



The character of the curves during the winter months is most interesting. The tempera- 

 ture at both Cape Adare and McMurdo Sound continues falling until it is between — 10°F. 

 and — 15°F. The simplest explanation is that this is the temperature at which the partially 

 frozen Ross Sea gives up sufficient heat to counterbalance the radiation to the clear winter 

 sky, and, therefore, it is the lowest mean temperature at Cape Adare and McMurdo Sound 

 which are so much under its influence. 



If we neglect for the moment the low temperature of August at Framheim, we see that 

 the winter temperatures are very similar at Framheim and on the Barrier ; this, as explained 

 above, is due to the blizzards which remove the cold air over the Barrier and replace it by 

 air in temperature equilibrium with the upper air whose temperature is governed by the 

 relatively warm Ross Sea. The mean temperature of August 1911 at Framheim may .safely 

 be judged to be abnormal ; a longer series of observations would, doubtless, show that during 

 the months May to August the mean temperature at Framheim would be nearly stationary, 

 as it is at other stations in the Ross Sea area. 



