22 
MR. E. RUTHERFORD ON A MAGNETIC DETECTOR OF 
We therefore see that, under ordinary circumstances, the resistances may be 
neglected in comparison with the inductances. 
If the branch ACB of the divided circuit consist of a high resistance of short length, 
and consequently small inductance, and the other branch of an inductance N, and 
R“ 
the maximum values of the currents in the two branches are equal, then 2 ^“ — - yi;, 
-*N " — -Li" 
since may be neglected in comparison with E-®, and supposing the value of M to 
be small compared with N. 
If the inductances are unchanged, is practically a constant for all periods, 
and p is therefore proportional to R; 
In practice, one branch of the divided circuit consisted of a standard inductance, 
N, and the other branch of an electrolytic resistance, R. The equality of currents in 
the two circuits was obtained by altering the value of R until the effect on the 
detector needles was the same for both circuits. Since the effect on the needle was 
the same in both circuits, the maximum values of the current are the same, since 
each branch is traversed by an oscillation of the same period. 
Fig. 4. 
It was not found necessary to place the divided circuit in series with the discharge 
circuit, but tlie arrangement was more satisfactory when it was shunted off a portion, 
PQ, of the discharge circuit XPQS (fig. 6), whose period is to be determined. The 
addition of the shunt circuit had no appreciable effect on.the period of the oscillation 
for the equivalent inductance of the two branches QP, QBAP, was slightly less than 
that of QP, and the length of QP was generally not a tenth part of the whole 
discharge circuit. 
ACB, ADB are the branches of the divided circuit. In C was placed a resistance 
consisting of zinc sulphate with zinc electrodes. The amount of resistance in the 
current could be varied by altering the length of electrolyte through which the 
current passed. 
In D was placed a standard inductance consisting of six turns of insulated wire 
wound on a bobbin 10 centims. in diameter. The self-inductance of this could be 
accurately determined by calculation, and was very approximately the same for 
steady as for rapidly changing fields. 
If the inductance L of the resistance branch was small compared with N, the value 
o^p is given by p = R/N, 
