554 
PROFESSOR J. A. EWING ON EXPERIMENTAL 
of the armature. The experiments have a practical value in showing that cores 
which are so thoroughly laminated as to render the induction of currents within them 
unimportant, do not involve any serious loss of energy, and that the efficiency of a 
machine with a soft iron core or cores whose magnetism is periodically reversed need 
not, on that account, be materially less than that of a machine which has no such 
cores. The absence of iron from the armature has been claimed, on the score of 
efficiency, as an important advantage possessed by some types of machine, but unless 
the claim has some other basis it appears to me to be illusory. Magnetic reversal 
does involve some loss of energy, but if the cores are properly laminated so that the 
loss consists almost entirely of the quantity — it is so small as to be practically 
insignificant. 
I shall show later that when a cycle of magnetisation is performed while the iron is 
kept in a state of mechanical vibration the value of —J2>c?<§ is much less than when 
the same cycle is performed with the metal in a state of rest, and indeed almost 
vanishes in soft iron. Hence, in a dynamo, where vibration occurs to a greater or 
less degree whenever the machine is running, the energy dissipated through changes 
of magnetisation is even less than these experiments on still metal might lead us 
to expect. 
[Note added March 25, 1886.—The heating effect of cyclic changes of magnetism, 
especially of reversals, is important from a practical point of view not only in relation 
to dynamo-electric machines, but also to “ secondary generators,” or induction coils 
for the distribution of electrical energy by electromagnetic induction. In § 52 
(below) mention will be found of occasional evidence which these experiments 
furnished that there is a true time lag in the magnetisation of iron : in other words, 
that there is a certain degree of viscous as well as static hysteresis in the relation of 
magnetism to magnetising force. Soft iron, especially in early stages of magnetisa¬ 
tion, exhibits a sluggishness in assuming the magnetic state proper to the magnetising 
current—a sluggishness which does not appear explicable as an effect of the induction 
of currents within the substance of the iron, nor as due to the self-induction of the 
magnetising circuit. The result of this magnetic viscosity is to augment the value of 
— in any rapidly performed cycle, and consequently to increase the evolution of 
heat. It seems highly probable that in the comparatively slow reversals of mag¬ 
netism which the core of a dynamo armature undergoes the additional dissipation of 
energy due to this cause is scarcely sensible. In the case of “ secondary generators,’ 
however, where reversals of magnetism occur far more rapidly, there is nothing to 
show that this source of loss is not a sensible portion of the whole ; but the high 
efficiency which this kind of apparatus has been proved capable of makes it at least 
clear that the magnetic viscosity of the core gives rise to no very serious loss of 
power. 
In connexion with “ secondary generators ” and induction coils generally, the 
bearing of the first part of this paper should be noted, as showing the enormous 
