PHYSICS: TOLMAN AND STEWART 
191 
coil were lead to the center of the wheel and there made connection, 
through wires which were allowed to twist up, with a balHstic galva- 
nometer which measured the pulse of electric current which was pro- 
duced by the tendency of the electrons to continue in motion after the 
wheel was stopped. For the purpose of eliminating accidental effects it 
was found among other things necessary to construct the apparatus 
of non-magnetic materials, to rotate the coil of wire in a space in which 
both the horizontal and vertical components of the earth's steady mag- 
netic field has been neutrahzed, and to compensate for the variable part 
of the earth's magnetic field by connecting in series with the rotating 
coil a compensating coil having the same flux area but wound in the 
opposite direction. For details of the apparatus our complete article 
which has been submitted to the Physical Review for publication must be 
consulted. 
The galvanometer throws were found to be in the direction predicted 
on the basis of a negative charge for the mobile carrier in metals and to 
agree within the Kmits of error with the equation 
vl 
of which the derivation will be given in our complete article. Q is the 
pulse of electricity sent through the galvanometer on stopping, v is the 
velocity of the periphery of the wheel at the instant of stopping, / the 
length of the coil, R the total resistance in the circuit, F the value of 
the faraday, and If is a constant whose value ought to be fairly close 
to that of the mass associated with one equivalent (i.e., with F = 96,540 
coulombs) of electrons. For copper wire the average value of M from 
131 runs was 1/1910 which may be compared with the accepted value 
for the mass of one equivalent of electrons in free space, namely 1/1845 
(i.e., the ratio of the mass of the electron to that of the hydrogen atom). 
Our purpose in carrying out these measurements has not been merely 
to demonstrate an effect which has long been an object of search; we have 
also had in mind the possibility of obtaining from our experiments infor- 
mation as to the nature of the conducting process in metals and indeed 
perhaps further information as to the nature of the electron itself.The 
equation given above which we have tested in this work was derived on 
the assumption that the conducting process in metals is in the nature of 
a drift of 'free' electrons when acted on by an electric field, and the 
fact that the equation seems to fit the experimental facts is to some 
extent a verification of these assumptions. Such considerations are of 
particular interest at the present time in view of J. J. Thomson's pro- 
