Variation of Hourly Meteorological Normals at Kimberley. 173 



the second day, and altogether negative during the third. Table 7 

 gives the respective annual means for the six hours of observation. 

 The separate monthly means are not definite enough to be worth 

 printing in full. By dividing the cloud types into cirriform, cumulus, 

 and stratiform we get the following cyclonic variation in the mean 

 monthly number of times observed : — 



Day Before. Day. Day After. 



High-level clouds 8'0 8-4 6-6 



Middle-level clouds 11-5 16-7 9-5 



Low-level clouds 4-3 5-8 2-2 



It appears from this that clouds of a cirrus type are not very much 

 affected in quantity by the depression, and that the chief influence 

 is upon the cumulus . and other comparatively low-level clouds. It 

 is worth notice that the variation in the frequency of rain is almost 

 the same as that of the lower clouds. This rain frequency is such 

 that in every 100 depressions rain is observed in 30 on the first 

 day, in 44 on the second, and in 27 on the third ; that is, on the day 

 upon which the trough is passing over the chance of rain is half as 

 great again as it is on the first day. During the three days we are 

 considering the total rainfall is slightly less than three times the 

 daily average of the year. The average daily amount is "051 inch, 

 while the average daily amount in the depression is *046 inch, or 

 90 per cent. To make this, 94 per cent, of the daily average falls 

 on the first day, 149 per cent, on the second, and only 27 per cent, 

 on the third. It seems from this that the rain of the third day, 

 when it does come, is only of about one-quarter the intensity of that 

 of the first and second days. 



It seems to be a fair inference from a comparison between the 

 temperatures, dew-points, clouds, and rain of the day before, and 

 the same elements of the day after the passage of a depression, 

 together with the dust on the first and second days and its absence 

 on the third, that the air currents have an upward tendency in front 

 of the centre and a downward tendency in the rear. 



Table 8 gives the number of times rain has been observed at any 

 hour in the 315 special days under review, i.e., in 105 depressions. 

 We see from this that the relative hourly frequency of each day is 

 very similar to that of the relative daily frequency. Also that 

 in spite of the falling-off on the third day the diurnal curve of 

 hourly frequency is not materially affected. 



Table 9 gives the variation in the velocity of the wind in miles 

 per day, month by month. The greatest daily velocity is attained 

 in the spring, October being the month of maximum velocity, and 



