FREQUENCY OF HIGH WINDS 321 



result is entirely different from that obtained for Yarmouth. At 

 Framheim calms were of frequent occurrence and the wind blew 

 at 4 miles an hour or less during 42 per cent, of the total time. 

 From this high percentage of calms winds of higher velocity fell 

 off rapidly and regularly. It is important to notice that the shape 

 of the curve for Framheim is similar to that of the curve for 

 Yarmouth from its highest point to its end. This shape is fre- 

 quently met with in scientific work, and indicates that the change 

 from maximum to minimum is regular without any outside factor 

 influencing the natural change from stage to stage. It is the shape 

 of the ' probability curve,' that is, the curve which indicates the 

 probability that anything will occur when it departs from a most 

 probable value. The wind conditions shown in the Framheim 

 curve are the ideal ones for polar work, for the most probable 

 wind is a calm, and the frequency wjth which higher velocities 

 occur decreases rapidly and regularly as the velocities increase. 



We will now turn to the results of a similar analysis for the 

 winds recorded at Cape Evans. Column 4 of Table I. contains 

 the data, and they are plotted on the thick curve of Fig. i. Here 

 we have a curve which commences in a manner similar to that 

 of Framheim: the most frequent winds are those with a velocity 

 of less than 4 miles an hour, and higher winds are less frequently 

 met with. In fact if one had the whole of the Framheim curve 

 and only the first part of that for Cape Evans as far as winds 

 of 14 miles an hour one would say they were similar and would 

 complete the Cape Evans curve along the thin dotted line indi- 

 cated in the figure, making it run parallel with the other two 

 curves. 



But that would be assuming that there was nothing abnormal 

 in the region in which Cape Evans was situated, and that the 

 winds were governed by the same laws as at Framheim. The 

 real curve does not follow this ideal curve, but takes an entirely 

 different shape. The frequency of winds greater than 15 miles 

 an hour does not decrease with the velocity, for all winds with 

 velocities between 15 miles an hour and 34 miles an hour occur 

 with practically the same frequency. It is not until we reach 

 higher velocities than 35 miles an hour that a decrease in fre- 

 quency accompanies an increase in velocity. This shape of the 

 curve indicates that there is some factor affecting the winds at 

 Cape Evans which is not present at a normal station. 



VOL. II 21 



