460 
DR. J. HOPKJNSON ON THE MAGNETISATION OF IRON. 
in absolute measure the primary magnetic properties of the materials and the very 
characteristic way in which they differ from each other. 
The curves given include in each case an ascending curve, taken before the sample 
had been submitted to greater magnetising forces ; a curve of residual magnetisation, 
that is, a curve in which the ordinate is the residual induction left after application 
and removal of the magnetising force represented by the abscissa, and two descending 
curves. 
Fig. 4 gives the curves from wrought iron No. I. 
Fig. 5 the same to an amplified scale of abscissae. 
Fig. 6 for steel with '32 per cent, carbon, annealed No. YI. 
Fig. 7 for steel with '32 per cent, carbon, oil hardened No. VII. 
Fig. 8 for steel with '89 per cent, carbon, annealed No. VIII. 
Fig. 9 for steel with '89 per cent, carbon, oil hardened No. IX. 
Fig. 10 for cast iron No. III. 
The descending curves, which express the passage from extreme magnetisation in 
one direction to extreme magnetisation in the opposite direction, may be roughly 
defined by the maximum ordinate to which they rise, and by the points AB in which 
they cut the coordinate axes. The ordinate OB is what is generally meant by the 
residual induction after great magnetising force, or the “ retentiveness.” The word 
“ Coercive Force ” has been long used, but, so far as I know, in a rather vague way 
and without accurate definition. 
I propose to call OA the “ Coercive Force ” of the material, and define it as that 
reversed magnetic force which just suffices to reduce the induction to nothing after 
the material has been submitted temporarily to a very great magnetising force. It is 
the figure which is of greatest importance in short permanent magnets. The manner 
in which the dimensions of the ascending' curves and the curves of residual magnetisa- 
tion vary with the descending curves is sufficiently obvious from inspection. The 
slowness with which iron or steel yields to small magnetising forces is evidently 
intimately connected with the coercive force. Another force is worth noting, viz., 
that demagnetising force which not merely reduces the induction to zero whilst 
applied, but just suffices to destroy the residual magnetism so that when removed no 
permanent magnetisation remains. The area enclosed by the two descending curves 
divided by 47r represents the energy dissipated when the unit volume is magnetised 
to saturation, its magnetism reversed, and again reversed, and so brought to its first 
value. This area differs a little from 4 X coercive force X maximum induction. In 
the cases for which curves are given the results are as follows :— 
I, 
