1267 
of a galvanometer, the deflections of which are proportional to the 
currents flowing through it. 
Let the current delivered by a constant external source of elec- 
tricity, say a single accumulator, be passed through the body 
and cause a deflection of the galvanometer of /, parts of the scale. 
If then the comparison resistance coil of @ Ohms be inserted into 
the circuit, the deflection will fall off, say to /, parts of the scale. 
If the unknown but constant E.M.F. be J”; and R& be the resistance 
of the body. we may state: 
hook «and Vi= [ne ER 
a being the current strength which causes the galvanometer to give 
a deflection of one part of the scale. 
From this we find 
VES 
0. 
hmm — + 
Ll, 
In this method the counter E.M. F. of polarisation is accounted 
for as it is ineluded in V. The accuracy of the method largely 
depends on the rapidity of the galvanometer deflection, and will be 
greater if the time of deflection be smaller. With a stringgalvano- 
meter in which the deflection time can easily be reduced to one or 
a few thousandth parts of a second, the accuracy is practically only 
limited by the accuracy with which the readings of the deflection 
are taken. For during the short time of deflection no change in 
the polarisation occurs, the capacity of polarisation being very 
great. But even if this were not the case we might still find the 
real initial value of the current by EINrHoveN's method of correction. 
This complication proved to be unnecessary. 
By this method it is an extremely simple matter to measure the 
resistance of the body continually during the whole experiment. We 
have only to connect the ends of the resistance o to a shert-circuil- 
ing key, which can be opened and shut several times per second 
by meaus of a small motor. In my experiments this was done about 
5 times per second for about 0,1 second. If the movements of the 
string be photographically recorded in the usual way, we get two 
sets of points on the negative which together form two distinet 
curves: the upper one corresponding to the current through the body 
only, the lower one to the current through the body and the com- 
parison-resistance. At the beginning and end of the record the current 
is broken for a few seconds, in order to mark the zero-line. 
After developing and drying the negative, the curves can be 
measured and # can be calenlated. At the same time we also cal- 
82 
Proceedings Royal Acad. Amsterdam. Vol. XVIII. 
