228 Dr. D. K. Morris on the Magnetic Properties 



magnitude of the standardizing throw. The difficulty was 

 much reduced by diminishing the number of observations, 

 though the number taken (just eight all round the cycle) 

 barely suffices to determine the hysteresis loop. 



The group of experiments at one temperature was con- 

 cluded by a measurement of the resistance of the iron. 



Throughout the whole of such sets of experiments, and 

 particularly when working immediately below the critical 

 point, where the magnetic properties vary so rapidly, the 

 most scrupulous attention was paid to the temperature that it 

 should not, if possible, be allowed to alter more than a fraction 

 of a degree ; and this attention was the more necessary when 

 working with the larger magnetizing currents, for their 

 heating effect in the primary circuit was quite perceptible. 



The time-period, of the ballistic galvanometer needle being 

 about eleven seconds, magnetic "creeping" effects lasting 

 much more than a second would not be recorded. 



In reducing the observations for each curve of cyclic mag- 

 netization, the mean of the corresponding throws taken during 

 the ascending and descending halves of the process has in 

 every case been taken ; though this eliminates any systematic 

 unsymmetry which the curve might have possessed. 



Discussion of Results. 



The results of the experiments are given in the accompany- 

 ing Tables and Curves (Plates II. & III. and Woodcuts, 

 pp. 247-254) ; the series being in each case arranged in the 

 order in which they were taken. 



I. Curves of Magnetization by the Method of Reversals. — 

 The effect of annealing (at 1050°) on specimen A is strikingly 

 shown in the curves marked 1, and in the corresponding per- 

 meability curves 9 and 10. Its effect on the maximum 

 permeability and corresponding field is shown in Table IX. 

 The ultimate result of the annealing was to increase the 

 maximum permeability at ordinary temperatures three-fold 

 (/z,= lo55 to 4050) whilst reducing the corresponding field to 

 little more than half its former value (3*72 to 2*02). 



The first annealing of specimen B affected the permeability 

 in a similar manner (compare curves No. 3). The experiments 

 on this specimen, however, were carried out with a view to 

 showing especially the difference between annealing at a red heat 

 and at a low white heat. The curves of magnetization at 

 different temperatures obtained after annealing respectively 

 at 840° and 1150° are numbered 7 and 8 ; whilst the 

 corresponding permeability-temperature curves for different 



