304 Transactions of the Boyal Society of South Africa. 
This table gives the mean monthly deviations from the mean position of 
the pendulum for each hour of the day. It appears from this that the 
movements in the ground which set up corresponding movements on the 
seismograph at Kimberley are very great. Assuming that the mean 
position of the pendulum indicates the undisturbed level of the ground, 
then the maximum westerly elongation of the pendulum occurs at 6\ h., 
a.m.; the maximum easterly about 4^ h.,p.m. ; the median positions a 
little before 11 h., a.m., and 9^ h,, p.m. Geometrically these movements 
may be represented on the hypothesis that the hemisphere facing the sun 
bulges out, the bulge forming a sort of meniscus to the geo-sphere. The 
enormous rises and falls of the surface of the land that such a supposition 
would postulate are, however, mechanically difficult. Moreover, it is 
impossible that they should be due to the alternate expansion and con- 
traction of the earth's crust, by day and night, in the heat of the sun's 
rays and the nocturnal cooling. For the direct heat of the sun during the 
course of the day does not penetrate below a depth of 3 feet, and even 
then it penetrates so slowly downwards that the maximum earth tempera- 
ture at a depth of 1 foot is not felt until some two hours after sunset, 
while it takes more than as long again before the wave of temperature 
reaches the 2-foot depth. Thus it is out of the question that the direct 
heat of the sun can affect the levels at the depth of the base of the pier on 
which the pendulum is fixed. 
Professor Milne, who has made and discussed many elaborate experi- 
ments and observations for the purpose of ascertaining the cause of the 
diurnal variation of level as it has been observed in Japan and other 
countries, has found that it may be met with in cases 12 feet below^ 
the ground-level ; but that in a space sheltered by trees from the sun's 
rays it is scarcely appreciable. He found further that when the character 
of the country to the east of a pendulum differed from that to the west — 
say when there were many trees or buildings on one side, but bare ground 
on the other — the diurnal variation of level was large. The general con- 
clusions to which the observations at his disposal point is that the move- 
ments which take place during the day are due to the removal of a load 
from the side of the station most exposed to the effects of radiation. 
This load may be represented by aqueous vapour carried upwards and 
dissipated. And a very important observation in this connection is that, 
at many stations, on wet or cloudy days diurnal waves are absent." It 
is known that in a damp country the diurnal variation of temperature may 
easily set up an alternate loading and unloading of the upper strata of the 
earth's crust. During the night, when radiation has cooled the surface, 
dew will be deposited from the air ; and, besides, the flow of moisture in 
* J. Milne, " Movements of the Earth's Crust," Geographical Journal, March, 1896. 
« 
