MEASUREMENT OF POTENTIAL DIFFERENCE 247 



of dynes corresponding to one division of the micrometer head 

 can then be calculated. 



The natural procedure in taking a measurement would be first 

 to short-circuit the instrument and bring the attracted disc to 

 the zero position by means of the micrometer head B, the read- 

 ing of which is noted. The potential to be measured would then 

 be inserted between the terminals of the instrument. This would 

 cause the attracted disc to move downward. The disc would 

 then be returned to its zero position by turning the head B, the 

 final position of which is read. The difference of the micrometer 

 readings gives the stretch of the spring. The force F is deter- 

 mined from this stretch and the calibration of the spring. 



It will be found that if the P.D. is small, the lower plate 

 must be brought very near to the attracted disc. For example, 

 if the potential difference is 500 volts, F, 100 dynes, and A, 

 100 sq. cm., the distance between the plates will be only 3 mm. 

 It is practically impossible to make and adjust the plate and disc 

 so that such a small value of S can be measured with certainty. 

 Slight irregularities of the surfaces and lack of parallelism of the 

 plates wduld vitiate the results. 



Tli is difficulty has been overcome by Lord Kelvin's method of 

 using an auxiliary high potential. Suppose that the guard ring 

 and disc are charged to a potential of 10,000 volts, the at- 

 t met ing plate being connected to earth. When the disc has been 

 brought to its sighted position, corresponding to F = 100, 

 S = 6.666 cm. If the plate be now connected to earth through 

 a 500-volt battery, thus making the potential applied to the 

 instrument 9500 volts, to return the disc to its standard position, 

 B remaining fixed S must be made 6.366 cm. That is, the lower 

 plate has to be moved through a distance equal to that which 

 would exist between the plate and disc if the P.D. were directly 

 measured. The advantage attained is that in both measure- 

 ments the attracting plates are so far apart that there is practi- 

 cally no uncertainty as to the value of S. When used as just 

 suggested, the instrument is said to be employed heterostatically; 

 when only the P.D. to be measured is employed the electrometer 

 is said to be used idiostatically. 



The expression for the P.D. when the instrument is used 



heterostatically is P.D. = 1,504(5 - S'\& The distance 



