564 
PROFESSOR J. A. EWING ON EXPERIMENTAL 
Effects of Vibration on Induced arid Residual Magnetism in Iron. 
§ 48. It is well known that mechanical disturbance makes iron take up more 
magnetism under an inducing force than it would take up if undisturbed, and that 
after the withdrawal of an inducing force mechanical disturbance reduces the residual 
magnetism. I was, however, unprepared to find these effects exhibit themselves so 
markedly as they did in certain specimens of annealed iron. 
It has been already mentioned (§ 14) that after an annealed wire of soft iron had 
been magnetised, and the magnetising force removed, gentle tapping had the effect of 
almost completely removing the residual magnetism, great as that was. In fact, in 
dealing with very soft annealed iron, I generally found mechanical vibration—produced 
by gently beating against the table—to be the most convenient way of restoring the 
wire under test to its original neutral condition after magnetisation. The residual 
magnetism of soft iron is so sensitive to disturbance that in experiments made with 
the view of observing the retentiveness of the metal, the utmost care must be used 
to avoid accidental errors due to this cause. For example, a long rod of soft iron, to 
which a magnetising force had been applied and withdrawn, leaving about 90 per 
cent, of the induced magnetism as residual, was lying on a table in the laboratory, an 
induction coil round the rod being in circuit with a ballistic galvanometer. A person 
walking quickly along a neighbouring corridor caused by each footstep a distinct 
transient current in the galvanometer, due to the shaking out of successive parts of 
the rod’s residual magnetism. The lightest tap or even rub by the fingers made the 
magnetism leave the rod with a rush; and during the application of magnetising 
force, the most apparently insignificant disturbance, occurring when the magnetisation 
was slight, would give rise to an enormous increase of magnetism, making it suddenly 
rise twenty-fold or more. For these reasons it is necessary, in experiments such as 
those which have been described in the preceding part of this paper, to guard very 
rigorously against accidental disturbance of the piece under test. Methods of 
measurement which involve the removal of the piece from the magnetising solenoid, 
or its movement in any way, before the magnetisation is determined, are quite 
out of the question. All the observations already described were made after I had 
become convinced of the necessity of preventing vibration, by supporting the rods 
and rings of Aon on stone columns, detached from floors, and well out of the way of 
passers-by. 
A number of separate experiments were carried out with the view of testing more 
particularly the effects of mechanical vibration on magnetic susceptibility and 
retentiveness. The results are given below. 
§ 49. (March 17, 1882.) A long wire of soft iron, annealed, 0T58 centim. in 
diameter and 64 centims. or 400 diameters in length, was examined as follows by 
the ballistic method. The magnetising force was raised by steps to a certain value, 
at which the magnetisation was determined by summing up the ballistic throws due 
