320 
Proceedings of the Royal Society of Edinburgh. [Sess. 
potential of the tank the value of the potential gradient at ground-level in 
the open is justified. 
In the matter of obtaining the reduction factor it may be here remarked 
that in conjunction with the flame collector a Kelvin electrostatic voltmeter 
supplied by Kelvin & White was used. This voltmeter was specially 
insulated with ebonite and had a scale with a range from 80 to 300 
volts, and has proved very convenient, being very much less trouble to 
read than an electroscope, while giving results in no way inferior to those 
obtained by means of an electroscope. 
I II 
§ 3. Instead of adopting the method of measuring ordinates, the curves 
were divided up into hour intervals ; this was easily done as the timing 
and earthing device mentioned earlier gave a point at each hour on the 
line of zeros. Ordinates having been drawn at each hour, the area included 
between the zero line, the ordinates, and the curve was measured with an 
Amsler’s planimeter reading to *05 sq. cm., and thus the mean for the hour 
was obtained. After measuring a few curves, such facility was attained 
that the time required to measure by this method was not much longer 
than that required to measure ordinates, while a more accurate value of 
the mean is thus obtained. 
Owing to errors in measurement of areas, the mean value of the 
potential gradient for any hour of any one day may be out by about five 
