I9IS.] IN AIR CIRCULATION OF THE GLOBE. 195 



statement concerning the prevailing direction of surface winds at 

 the Danish base near Cape Bismarck. More recently (1911) the 

 full meteorological report by Wegener has been issued; and, con- 

 firming the earlier statement, shows that all strong winds come from 

 the westerly (inland-ice) quadrants. The frequency wind-rose to 

 cover the entire period of two years over which the observations 

 extended, is reproduced in Fig. 3.^" If the wind force had been 

 taken account of, the easterly sections of the rose would have almost 

 disappeared, since easterly Vidnds are always light sea breezes, 

 which at an elevation of only 1,000 meters have been completely 

 overwhelmed by the northwest winds." In this rose the dextro- 

 rotatory deviation of the down-slope winds is apparent. 



Early Evidence from Antarctica. — Over the Antarctic inland-ice 

 the law of surface air circulation had been clearly indicated by the 

 results of exploration at the time of my early discussion of the 

 subject. The more important data had been derived from the 

 sledge journeys of Captain Scott, Sir Ernest Shackleton, Profes- 

 sor David and Dr. von Drygalski. As early as 1902 Captain Scott 

 had ascended the Ferrar glacier outlet to the inland-ice above the 

 mountain rampart and pushed west southwestward over it for a dis- 

 tance of two hundred miles, ascending on ever decreasing grades to 

 the farthest point attained, and encountering winds of nearly con- 

 stant direction coming from the south-southwest. The prevalence 

 of such winds was demonstrated by a single set of sastrugi which 

 pointed in the same direction (see Fig. 4).^- Shackleton on his 

 polar journey ascended the Beardmore outlet and for a like distance 

 of two hundred miles over the inland-ice found strong winds blow- 

 ing from the southerly quarter and sastrugi pointed in the same 

 direction. David pushed northwestward from Ross Sea over the 

 inland-ice to the south magnetic pole, crossing over a crest in the ice 

 and descending on low grades during the last stage before reaching 

 the pole. Here the same rule of distribution of currents applies, 



^° A. Wegener, "Med. om Gronland," Vol. 42, 191 1, pp. 324-326. 

 1^ Wegener, " Med. om Gronland," Vol. 42, 1909, pp. 73-75. 

 ^- For this and other references to work published before 1910, see 

 " Characteristics of Existing Glaciers," Chapters XIV.-XVI. 



