398 
Transactions of the Boyal Society of South Africa. 
12 noon. The Table attempts to show what the maximum value would 
be at the place of observation. Straight lines were drawn upon squared 
paper from the value of the gradient at its maximum to that of the value 
at its minimum, the horizontal line being the axis of time, and then 
assuming that the potential falls with the time, values of the maximum 
gradient can be found by extrapolation. For example, an observation at 
Crocodile Poort at 11 a.m. gave a gradient of 110 volts per metre, but 
by extrapolation it is assumed that the maximum value of the potential 
gradient would be 210. Proceeding in this fashion, the values given 
in the Table have been deduced, but they must be taken as approxi- 
mate only. The values of Belfast and Louren90 Marques are of course 
definite. 
At Lourengo Marques two and a half days' observations were taken, 
partly at the Observatory, on the beach, and at points in the neighbour- 
hood. The humidity of the air was very great, which interfered with the 
first half-day's and also the next day's work, but the third day, after 
9 a.m., was drier, and gave what may be regarded as a normal set of 
observations. 
A day's results, taken at the Observatory, are given in Table III. 
Table III. 
Lourenco Marques, September 27th. 
Time. 
Pot. Grad. 
Time. 
Pot. Grad. 
Time. 
Pot. Grad. 
5.30 a.m. 
60 
9.40 
174 
1 5.40 
100 
5.40 
60 
9.45 
174 
6.10 
100 
5.50 
60 
10.0 
162 
6.30 
120 
5.57 
68 
10.15 
125 
1 7.0 
176 
6.0 
68 
10.30 
118 
! 7.15 
180 
6.10 
68 
10.40 
108 
7.30 
180 
6.20 
80 
11.0 
108 
1 7.45 
180 
6.30 
75 
11.10 
95 
' 8.0 
165 
6.35 
100 
11.30 
95 
8.15 
160 
6.40 
104 
11.55 
95 
8.30 
150 
6.50 
162 
12.0 
95 
8.45 
110 
7.5 
260 
12.30 p.m. 
95 
9.0 
100 
7.17 
280 
9.15 
87 
7.35 
280 
9.30 
87 
7.40 
158 
9.45 
90 
7.50 
225 
8.0 
200 
8.10 
280 
8.20 
260 
\ 
/ 
8.30 
196 
5-25 
100 
