Tidal Phenomena at Inland Boreholes near Cradock. 75 
amplitude generally less than 2 millimetres. Occasionally also there 
appears here and there a wave-like movement of 3 or 4 millimetre 
amplitude and 1 or 2 hour period. 
I have examined a number of records taken by the tide gauges at 
various points along the South African coast, and find that a frill of 
minor fluctuations of similar periods forms a characteristic feature of the 
marine tide curves, and I am given to understand that they are generally 
ascribed to marine seiche movements. 
The amplitudes of the great 12|-hour period waves were obviously 
subject to considerable variation. A general inspection brought to light 
the fact that these amplitudes attained a maximum of about 1^ inches on 
June 17, which happened to be the date of full moon, and attained a 
minimum of less than f inch about June 10, when the moon reached its 
first quarter. Another minimum was attained about May 26, when the 
moon reached its last quarter. 
I had expected to find that variations of barometric pressure would 
exert an influence on the water-level in the tank A, and accordingly 
barograph records were taken on the farm simultaneously with the water- 
level records. 
A comparison of the barograph records with the water-level records at 
first did not yield absolutely certain evidence of a connection between the 
two. During the June fortnight the barometric pressure remained very 
steady, and the barogram did not show any variation amounting to more 
than "2 of an inch, and mere general inspection did not disclose any 
movements of the water-level curve that could be attributed to barometric 
change. During the May week, however, a greater variation of atmo- 
spheric pressure occurred, the barogram showing a gradual fall, extending 
through the greater part of the week, while the mean line of the water- 
level curve gradually rose during the same period. It was possible that 
this might have been a fortuitous coincidence, and it became obvious that 
much longer records would have to be secured if such a connection was to 
be established beyond doubt. Longer records would also be required to 
put on a more satisfactory basis the connection suggested above as existing 
between the great wave amplitudes and the lunar phases. 
Messrs. Rayner and Roberts have from the beginning evinced a keen, 
intelligent interest in my investigations and a gratifying sympathy with 
the spirit of the inquiry. They have always been anxious to place all 
reasonable facilities and help at my disposal, but during these early years 
they were engaged in a very severe struggle with the difficulties of semi- 
arid farming. Their livelihood was dependent on their control of the 
water of these wells, and they were then unable to permit the uniformity 
of conditions necessary for the securing of a continuous record of two or 
three months. In particular they found it necessary from time to time to 
