A Further Note on the Diurnal Variation of Level at Kimberley. 75 
which is the day of least pressure — have been compared with the daily 
ranges of the pendulum from east to west for the same days. The 
following are the results : — 
TABLE 3. 
Showing the Effect of Barometric Pressure upon the Diurnal 
Eange of the Horizontal Pendulum. 
Barometer. H.P. Range. 
Mean. Mean. 
Day 1 26-129 inches 5-9 mm 
2 26-058 5-8 
3 25-987 3-9 
4 26-075 4-4 
5 26-133 6-9 
The mean daily range of the pendulum being 5-5 mm., it appears that 
when the barometer is lowest during the passage of a depression the range 
is 71 per cent, of the mean, whereas in the wake of the depression the 
range rises to 125 per cent.* The state of the sky is conceivably respon- 
sible for a large part of this difference.! 
Table 4 shows the mean results of a comparison between various 
meteorological elements and the range of the horizontal pendulum. It is 
got (1) by arranging the daily duration of sunshine in order of magnitude 
each month, together with the corresponding values of earth temperature 
range at a depth of 1 inch ; earth temperature maximum ; maximum in 
the sun (black bulb m vacuo) ; range of barometric pressure ; and range 
of the horizontal pendulum from east to west. (2) By dividing the 
monthly sets thus arranged into threes, each including as nearly as 
possible the same number of days, one set including all days with 
little sunshine, another with a medium amount, and the third set with 
a great deal. (3) By taking the annual averages of each of the three 
sets. 
* This is the only relationship between the height of the barometer and the movement 
of the pendulum that I have been able to detect in anything like a definite way. That 
the pendulum at Kimberley does not deviate in any marked fashion from the centre of 
an approaching depression may be due to the fact that our storms usually travel from 
SW. to NE., and therefore cannot affect the gradient at right angles to the meridian 
to the same extent as they would if they travelled from W. to E. Probably, however, 
the views set forth by Dr. Klotz (Report of the Chief Astronomer for 1907-8, Ottawa, 
1910) will make it necessary to reconsider the whole question of the relation between 
the variations of atmospheric pressure and level. 
t The amount of cloud increases gradually as a depression approaches, and decreases 
rapidly as the depression passes away. 
