154 



CLOUD AND PRECIPITATION. 



in 1902.* Shaw pointed out tliat when thin clouds lose their lieat by radiation evaporation 

 takes place and they disappear. The more rapid the radiation the more rapidly does the cloud 

 disperse. Now during the \\-inter in the Antarctic the conditions are eminently suitable ff>r 

 large and rapid radiation from the clouds. In the first place the clouds are very thin and 



Fig. 5(1. Annual variation of cloixl. 



therefore every part is able to radiate to the clear sky above, secondly they receive no 

 heat from the sun and practically none from the ground. Thus the radiation from the clouds 

 during the winter months is greater in polar regions than in any other part of the world. 

 This explanation is strongly supported by figure 51. In this figure the mean cloud amount 

 has been shown for each month during (a) winds from the south greater than 30 miles an 

 hour, (6) winds from the south between 11 and 30 miles an hour, and (c) winds from calms 

 to 5 miles an hour. The data used are September, 1911, to February, 1912, and March to 

 August, 1911 and 1912, thus the most important part of each curve is based on observations 

 from two years. Except in January the cloud amoimt with high southerly winds is nearly 

 constant from month to month, showing that blizzards are the cause of cloudy skies through- 

 out the year. The cloud amounts during moderate winds fiom the south and during calma 



* W. N. Shaw. ' La luno mange les miages.' Quarterly Journal of the Royal Meteorological Society, 

 Vol. XXVin, page 9.5, 1902. 



