6 
MR. E. RUTHERFORD ON A MAGNETIC DETECTOR OF 
the period of the oscillation. This is the case whether we are dealing with Leyden 
jar circuits, or the free vibrations, such as are set up in Hertzian receivers. 
An unniagnetized needle, on the other hand, is not appreciably magnetized when 
placed in a circuit where the damping is small, for each successiv'e o.scillation destroys 
the elfect of the previous one. Soft iron wires exhibit a similar effect to steel, only 
it is difficult to use wires of sufficient length to retain their magnetization. The 
effect on the needle is in general a purely surface one, and the amount of demag¬ 
netization does not bear any simple relation to the magnetizing force acting on it. 
After every experiment, the needle was removed and placed in a solenoid and a 
steady current passed, sufficient to saturate the steel. In this way the needle could 
always be quickly reduced to the same state after any experiment, and it was found 
that, using hard steel wires, the results of successive experiments were very consistent 
with one another. 
Passage of a Discharge Longitudinally through a Magnetized ILf/v. 
If a piece of pianoforte wire, several centimetres in length, be taken and placed in 
series with the discharge circuit, in the passage of a discharge, the magnetic moment 
of the needle is diminished, due to the “circular” magnetization of the wire. If the 
needle be dissolved in acid, it will he found that there is a thin skin, apparently 
magnetized in opposition to the original magnetization, due to the resultant action 
of the demagnetizing force of the needle and the magnetic force due to the current 
in the wire. 
The magnetic force H acting at any given point in the wire is given by 
H = 
r 
5 
where y is the current through the conductor flowing internal to the circle described 
through the point, and concentric with the surface of the conductor. The value at 
the surface of the wire is given by 
where a is the radius of the wire. 
Assuming /x, the permeability of the iron wire, as constant, the maximum value of 
the current at any point of the conductor decreases in geometrical pu’ogressioii as the 
distance from the surface inwards increases in arithmetical progression. As will be 
shown later in the part on “ Resistance of Iron Wires,” the current falls off even 
more rapidly than the theoretical law, on account of the increase of the value of /x as 
the amplitude of the current diminishes in intensity from the surface inwards. 
For thin wires the maanetic force at the surface of the wire is much oreater than 
for thicker ones. We should, therefore, expect a thick magnetized wire conveying the 
current to be affected to less depth than a thin one, and this is found to be the case. 
