Variation in Value of AUnosioheric Electrical Potential. 899 
The course followed by the potential gradient is thus of the normal 
kind, but the value of the maximum is much less than the previous half- 
day's work would have led one to expect. The atmosphere was steamy 
and probably interfered with the efficiency of the radium collecting-plate. 
From 12.30 p.m. to 5.30 p.m. the gradient was practically constant. 
On the following morning there was a " Scotch " mist, which made 
observations valueless until 8.30 a.m., when the mist cleared away and for 
the rest of the day the indications were normal. Considerable wind in 
gusts was blowing up to 8.30, and dust might have been caught up, thus 
diminishing the positive charge in the air. 
Table IV. 
Lourenco Marques, September 2Sth. 
Time. 
Pot. Grad. 
Time. 
Pot. Grad. 
Time. 
Pot. Grad. 
5.50 a.m. 
30 
6.40 
30 
7.45 
30 
6.0 
0 
6.50 
30 
8.0 
0 
6.5 
35 
7.0 
45 
8.30 
60 
6.10 
30 
7.15 
48 
8.45 
100 
6.12 
0 
7.30 
110 
6.15 
60 
7.35 
62 
6.20 
80 
7.40 
60 
6.30 
30 
After mist had cleared away 
9.30 
400 
11.0 
310 
12.0 
250 
9.45 
382 
11.5 
330 
12.15 p.m. 
205 
10.0 
360 
11.15 
335 
12.30 
260 
10.15 
310 
11.30 
300 
12.45 
250 
10.30 
310 
11.40 
300 
10.45 
320 
11.50 
250 
JOHANNESBUEG TO DuEBAN. 
On the down journey observations could not be taken on account of 
rain until Pietermaritzburg was reached, and then between 6 p.m. and 
10 p.m. a maximum potential gradient of about 200 volts per metre was 
obtained, but as rain was falling in the neighbourhood the observation 
was not of much value. 
Next day observations taken on a clear open space at the south end of 
the town gave the results shown in Table V. 
