VELOCITY— DAILY VARIATION. 



103 



The variation during the ^vinter was not the same in the different years. In 1912 when 

 there was an abnormal amount of wind during the winter months the daily variation was very 

 large and quite different from that during the winter of any other of the four years. It is 

 therefore very doubtful whether the variation shown during the winter has any real signi- 

 ficance. 



Daily Variation of Calms.— The frequency with which calms (0—1 nnle per liour) occurred 

 is shown by the curves on figure 33. The data used are for the twelve months March, 1911, 

 to February, 1912. As is to be expected these show the same effect as the mean wind 

 velocity, the calms being most frequent in the night and least frequent in the daytime. 

 There is however one important difference. The character of the curve for the three winter 

 months, May-Julv, is practically the same as during the other seasons of the year. 



It is easy to see why there should be more calms during the night and early morning 

 than in the afternoon during seasons of the year with marked difference of insolation at 

 these times. But why this characteristic should remain during months when the sun was 

 entirely absent is not clear. The effect is most marked and is in all probability quite real. 



Frequency of Whuls of Different Velocities. 



The frequency with which winds of different velocities occur at any place has been 

 strangely neglected in meteorological investigations; although, as we shall show, it is capable 

 of givin" important information with respect to the forces at work producing the air motion. 

 If from a rec(jrd of hourly wind velocities at any place we count the number of times each 

 individual wdnd velocity has been reported it is possible to plot a curve showing the relative 

 frequency with which each velocity occurs. In practice it is sufHcient to choose groups 



of velocities and in the following the groups will be to 4, 5 to 9, 10 to 14 



n to n +4 miles per hour. 



The data for three years' hourly observations of wind velocity at Yarmouth on the east 

 coast of England and for the winter months December, January, and February from two 

 years at Jubbulpore in the centre of India have been treated in this way with the following 



result : — 



Table 58. 



Frequency of Winds of Different Velocities, expressed as a percentage of the whole, at Yarmouth 



and Jubbulpore. 



These values have been plotted on figure 34. It will be seen at once that the curves 

 foi' the two places are very different. At Jubbulpore the first group is the greatest, in fact 

 more than 60 per cent, of all the observations fall within the group 0—4 miles an hour. 

 At Yarmouth on the other hand this group contains only 5-2 percent, of the observations, 

 while it is the third group which contains the greatest number of observations. From the 

 curves it is clear that while at Jubbulpore calms occur more frequently than any wmd velo- 

 city, at Yarmouth the most frequent wind is one having a velocity of about 8 miles an hour. 



