852 Proceedings of Royal Society of Edinburgh. [sess. 
Retentivity), than they do under the T h conditions of lower 
permeability. When H is reversed and reaches that particular value 
which measures the coercive force for each set of cycles, the curves 
cross, and the consequence is that the hysteresis loops following 
upon transverse demagnetisation are sheared over to a greater 
extent than the loops following upon co-directional demagnetisa- 
tion. Annealed iron ta,kes its place between these two extremes ; 
but on the whole, and within the limits above given, the hysteresis 
cycles approximate much nearer to those of transversely demagnet- 
ised iron than they do to those of co-directionally demagnetised iron. 
The areas of the three curves, however, appear to remain the 
same, so long as the maximum induction is the same. The 
conclusion is therefore arrived at that the energy dissipated as 
heat when the iron is subjected to cyclic alternation between 
positive and negative maxima is not affected by demagnetisation, 
by co-directional or transverse forces, or by annealing. 
The above summary of the more important results observable in 
the curves plotted for one sample of iron is only to be regarded as 
a preliminary communication on this subject. The publication of 
the curves, and any discussion following thereon, is reserved until 
the experiments have been repeated for other samples of iron and 
also for nickel. It may also be stated that ratio curves have been 
exhibited, the curve under the Ch conditions having been taken 
as unity. When H is sufficiently reduced the curves rapidly 
approximate to one another, and may even cross in some cases in 
the neighbourhood of H = 0’05. Although the initial permeability 
of the various curves cannot differ greatly, it is proposed to repeat 
these experiments at the lowest possible values of H (after the 
sensibility of the ballistic galvanometer has been adjusted so that 
maximum readings may be obtained under the given experimental 
conditions) before attempting a final conclusion. 
All the experiments have been conducted in zero magnetic field. 
The importance of the results obtained, in their bearing on any 
theory of magnetic induction, is obvious. 
As on a former occasion, I acknowledge my indebtedness to the 
Royal Society of London for having placed at my disposal a Govern- 
ment grant for these researches. 
( Issued separately August 29 , 1905 .) 
