WEATHER. 13 



From beginning to end of a blizzard the wind is extremely constant in direction. Neitlier 

 at Cape Evans nor on the Barrier is there any evidence of a regular change of wind direc- 

 tion during a blizzard. When the blizzard comes to an end, the air motion stops entirely, 

 or there may be a sudden change in wind direction of 180°, and the wind, sometimes of a 

 high velocity, then blows from the north. 



For a typical blizzard snowfall and drift are necessary, but from a meteorological point 

 of view there is no essential difference between a southerly wind with little or no cloud and 

 a blizzard with drift so thick that om, cannot see more than a couple of yards ahead, for 

 evsry grade between these two extremes occurs. In all our statistical work, therefore, and 

 generally in our descriptions, a blizzard will really mean a high southerly wind, which may 

 or may not be accompanied by snow and drift. 



The duration of a blizzard can be anything from an hour or two to several days. There 

 are many examples of a southerly wind rising in a few minutes from a calm to 30 or 

 40 miles an hour and then dying away again within the hour; on the other hand the 

 longest blizzard occurred in June 1912 when from 20 hours on the 7th until 11 hours on 

 the 14th, i.e., for 6 days and 14 hours, the anemometer recorded more than 20 miles in every 

 hour and the mean velocity during the period was 48 miles an hour. 



The temperature during southerly winds in the winter is always much higher than during 

 calm weather. This has given rise to the idea that the temperature during blizzards is 

 abnormally high, and often in Antarctic literature one's attention is called to ' the un- 

 expected warmth of southerly, i.e., Polar, winds.' We shall show that the southerly wind is 

 not a warm wind, but as would be expected it is always colder than a corresponding wind 

 from the north. 



The sequence of weather changes is remarkable. It was a popular opinion amongst the 

 members of the expedition that a blizzard was usually preceded by a northerly wind. As 

 a matter of fact the records do not support this opinion. It frequently happens that a period 

 of northerly wind precedes a blizzard, but it as frequently happens that a blizzard commences 

 without a previous nort.herly wind. On the other hand many blizzards end by the wind 

 changing completely round and blowing from the north. We shall show in the section on 

 Pressure, Wind and Weather that the theoretical sequence of weather is a cycle in which 

 northerly and southerly winds alternate, with the period of the southerly v\inds longer than that 

 of the northerly winds. In reality this is what happens, but the cycle is constantly broken 

 up by periods of calms. Blizzards are therefore associated with northerly winds, but they 

 as frequently occur after as before. 



Weather at Franiheim.— The weather at Framheim is totally diii'erent from that at Cape 

 Evans. The tiiie Antarctic blizzard does not occur at Framheim and during Amundsen's 

 stay only 2 per cent, of the wind observations recorded velocities over 30 niiles an hour — 

 this compares with 30 per cent, at Cape Evans. The frequency of calms at Framheim is 

 nearly twice as great as at Cape Evans and during 42 per cent, of the whole time the 

 wind is 4 miles an hour or less. The direction of the prevailing wind at Framheim is due 

 east, while there is practically no air motion from W.N.W. to N.N.E. 



Although Framheim is only 60 miles further south than Cape Evans its temperature is 

 very much lowei. The mean temperature at Framheim during the period April to September 

 1911 was 19°F. lower than at Cape Evans, and the observations during 1911 and 1912 

 indicate that in all probability the mean annual temperature at Framheim is 15°F. lower 

 than at Cape Evans. 



Weather at Cape Adare.—Q&^e. Adare is situated on the coast of the Antarctic Ocean 

 which is the most boisterous ocean in the worid. In a similar situation about 700 miles 



