THE MAGNETIC PROPERTIES OP IRON. 
541 
I 
1 way above described, is nearly proportional to the impressed E.M.F., and hence this 
■was assumed to be the case for the Foucault currents also. The other constants 
were then obtained by solving for them by means of simultaneous equations of the 
form e = AF + BF, the necessary data to form the equations being taken from the 
experimental results. It was found in this way that x was practically unity, not 
differing by more than one ])er cent, for the cases taken, and hence it is taken unity 
in the curve. The results of static experiments would indicate that the term BF, 
which represents the energy dissipated due to the retentiveness of the iron, should 
vary more rapidly than the induction, that is, that x should be greater than unity, 
about 1'6 for the experiments of Ewing and others. This result does not seem to 
be reached by the kinetic method here adopted, and an examination of recent work 
on the subject seems to confirm the results here arrived at. In an elaborate series 
of experiments recently published by Mr. C. P. Steinmetz an attempt is made to 
show that the exponent x is always 1'6. Unfortunately his results have most of 
them been reduced on this assumption, together with another based partly on 
calculation for Foucault currents. When the value of x given by the results is deduced 
from a series of equations by the methods indicated above the exponent unity seems 
to agree just as well, sometimes better, than 1’6. An excellent series of experiments 
for this purpose have recently been made by Mr. Alexander Siemens and the results 
communicated to the Institute of Electrical Engineers.* In this paper also the losses 
from Foucault currents are calculated on an assumption of size of wire, specific resist¬ 
ance, &c. The estimate of the energy dissipated seems to be somewhat too small, and 
with a proper correction on it, about 25 per cent., the dissipation given under hysteresis 
becomes simply proportional to the induction. The results of these experiments 
treated by the simultaneous equation method for the determination of the law of 
variation of the different elements of the dissipation give for Foucault currents AI^ 
and for hysteresis BF with almost perfect exactness. The fact that a sufficient 
number of fairly accurate ex^ieriments furnishes data for the mathematical deter¬ 
mination of the law of variation of the different elements entering into the dissipation 
of energy in cases like that here considered seems to have been very generally 
overlooked. 
I 
[Note added June 20, 1893.—Subsequent experiments, the results of which I 
hope soon to place before the Royal Society, show that the constant A in the above 
equation should have been zero, as there is practically no Foucault current loss. The 
exponent x is also shown by these experiments to agree closely with the number 
obtained by Mr. Steinmetz from Professor Ewing’s experiments and, since this 
paper was written, more fully established by his own experiments. 
* “Some Experimental Investigations on Alternate Currents.” See ‘Electrical Engineer,’ February 
19th and 26th, 1892. 
