78 
Transactions of the Royal Society of South Africa. 
done at Kew and other observatories, for the irregular variations caused 
by the dust v^ould make it impossible to do this with advantage. I must 
therefore merely put the observations on record as a means of shov^ing 
what a strong influence dust can exert locally on the electrification of the 
atmosphere. 
The Table. 
The table gives the record of observations, showing the value of the 
potential gradient at the end of each hour. The maximum gradient is not 
necessarily shown, as of course it may occur between the hours, and in 
that case it would be found by reference to the curve. The numbers given 
are usually for the positive potential gradient in volts per metre for open 
space, but those numbers to which an asterisk is affixed denote that the 
potential gradient is negative. If a blank space occurs in the table, it 
indicates in most cases that a dust-storm is blowing and that the charge is 
negative and too great to be indicated by the instrument. The maximum 
values recorded for the positive gradient during the period under observa- 
tion do not exceed 330 volts per metre. The comparative lowness of 
these results as compared with those for places nearer the coast may be 
accounted for by the fact that Bloemfontein is 4,500 feet above sea-level ; 
and the author has shown " that the potential gradient varies inversely 
as the height of the observing station above sea-level. Higher results 
have, however, been observed at other times of the year. The much higher 
values of the negative gradient have been approximately determined in an 
indirect way. 
During strong dust-storms sparks have been seen to pass between 
insulated and earth-connected conductors, and by finding the maximum 
length of spark attained a rough approximation to the difference of 
potential between the two conductors may be made. A piece of apparatus 
was constructed which during the passage of a strong dust-storm acted as 
a very efficient electrical machine, and was capable of giving sparks 
between its terminal up to 1'5 cm. in length. It consisted of two con- 
ductors : one coated with a radium preparation to take the potential of the 
air, and another hollow conductor arranged near the ground to catch the 
dust — it might even be resting on the ground. A brass sphere about 4 cm. 
diameter served for the aerial collector, and an empty paraffin tin with the 
bottom covered with fine wire gauze for the other. Wires from these 
were carried to a couple of insulated spheres to act as a discharger (see 
Fig.). When this apparatus was set up on the open veldt, a torrent of 
sparks passed between the discharging spheres when a dust-storm was 
blowing, and by measuring the length of the spark a fair estimate of the 
difference of potential could be obtained. If the spark was of a length of 
* TravH. Boy. Sac. South Africa, Part II., 1912. 
