16 ON THE ANNUAL AND DIXJRNAL DISTRIBUTION OF THI 



V. The principal maximum occurs with the wind from S.S.W. from 

 1 1 A.M. to 4 P.M., namely, during a portion of the time when the dura- 

 tion of the south wind is above the average, and it occurs with the 

 N.N.W. and north winds mostly at the hours when the duration of the 

 north wind is above the average, a second maximum vibrating from 

 east to E.N.E. during the whole of the day and night. From 9 a.m. 

 to 11 A.M., and from 4 p.m. to 7 p.m., namely, when the north and 

 the south winds respectively are near their averages as compared with 

 other winds, and when the winds in the N.W. quadrant are more 

 equally distributed among its several points, the easterly or second 

 maximum surpasses in value the westerly or principal maximum. 



The character of the diurnal periodicity of the different winds is more 

 apparent in table IX., in which the duration of each wind at each hour 

 is expressed in terms of the average duration of that wind in the twenty- 

 four hours. 



If the columns corresponding to the four cardinal points be examined, 

 it is found that the west wind, during the night, is mostly above the 

 twenty-four-hour average, and below that average during several hours 

 of the day ; but the range is small, the maximum being to the mini- 

 mum in the ratio of 1.36 to 1. 



The east wind from 8 a.m. to 9 p.m. is above the twenty-four-hour 

 average for that wind, and is below the average from 9 pm. to 8 a.m., 

 its diurnal range, or the ratio of the maximum to the minimum, being 

 2.40 to 1. The north wind is above the average from 10 p.m. to 

 9 A.M., and below the average from 9 a.m. to 10 p.m., and has a range" 

 of 3.44 to 1. The south wind is above the average from 10 a.m. to 

 7 P.M., and below it from 7 p.m. to 10 a.m. Its range is 4.82 to 1. 



Calms occur eight times as often between midnight and 1 a.m. as 

 they do between 1 p.m. and 2 p.m. The hours of maximum and mini- 

 mum frequency of calms are very nearly the same as those of minimum 

 and maximum mean velocity, a correspondence which, as appears from 

 table VII., does not hold in the case of the annual distribution of 

 calms. 



