The Winds of Kimberley. 85 
and direction together—for, as we have seen, hourly observations of 
welocity were not commenced until March, 1897. There are seven- 
teen columns:— | | 
1. The hours of the dag. 
2. The mean hourly velocity. 
_ 3,4, 5,6. The mean hourly velocity for each quadrant of daeetions. 
-7,8,.9, 10. The hourly departure, Q,—M, from the mean velocity 
M of the same hour (where Q is a quadrami: and 12 its nae 
‘smoothed in threes by Bloxam’s process... 
11, 12, 18, 14. Numbers such as those of Table 7, but for ao 
years instead of three, also arranged in, quadrants. 
15. The hourly values of the humidity of the air for the see 1898. 
16. The mean hourly rate of evaporation from a free water surface. 
1%. The hourly values of the mean barometric pressure of the air. 
It is to. be understood, of course, that every particular value in the 
‘table is, with the exception of those in the humidity and pressure 
-columns, for the hour ending at the time indicated on the same row 
‘by the first column. In the humidity and pressure columns the 
values apply to the hour. : 
The diurnal curve of wind velocity contains two maxima at 2 pe. m. 
‘and 10.45 p.m.),* and two minima (at 5 a.m. and 7.30 p.m.). The 
‘second maximum is.of more than ordinary interest. It is strongly 
smarked in the winds of the second and third quadrants, but is not so 
conspicuous in those of the first and fourth. In the diurnal .curve 
of evaporation from a free water surface there are also, as 1t happens, 
‘two maxima in the hourly rates fallmg at the same hours as those 
iof the wind velocity.. The first is obviously due to the high, wind 
‘velocity and low humidity during the warm hours of the day. Now 
the humidity curve rises rapidly, particularly during the winter 
amonths; from:just before sunset until about 8.30 p.m., after which, 
for three hours or so, the rate of increase.is not large. The second 
maximum in the rate of evaporation is not, therefore, difficult to 
see ay This matter, however, is merely mentioned in passing, 
Columns 11-14 give essentially the same rule as Table 7. But 
the chief interest of Table 12 is to be looked for in columns 7-10. 
The numbers they contain are in remarkably regular sequence, and 
would doubtless be even more so in a more extended series, They 
discover in a most decided fashion the important fact that for any 
hour of the day the mean velocity of the wind from any quarter 
decreases relatively to mean diurnal curve, with the deviation of the 
«The actual value of the second maximum is only'approximate in consequence 
of the uncertainty introduced by the SSG om ne the record of the ese strap 
between the hours XXI. and XXII. 7 
