Dr. 0. Fromme on the Magnetism of Steel Bars. 203 



cated, so that the correctness of its hypotheses has become 

 almost a certainty. But on the ground of these experiments 

 I also remark beforehand that then only can we properly speak 

 of a limit of the residual magnetism, when we presuppose the 

 steel freshly annealed and perfectly unmagnetic. A body even 

 only a little permanently magnetic shows another limit of the 

 residual magnetism, varying according to the amount of it. 

 A perfectly unmagnetic piece of steel, however, belongs to the 

 order of ideal notions ; and therefore it will be impossible to 

 demonstrate Maxwell's assumption by actual experiment (see 



I refrain from discussing the values of k, because I hold 

 them for the most part to possess too little accuracy to deduce 

 conclusions from them, or even to compare them with those 

 ascertained by means of the earth-inductor. 



Thinking that the magnetic state of the ellipsoids could only 

 have been very little altered by the forces hitherto applied, I 

 now subjected the pieces of sort I. once more to induction by 

 the vertical component of the earth's magnetism. As a proof 

 of the correctness of this assumption, the following values of k 

 were obtained, which do not sensibly differ from those given 

 in Table II. :— 



I. 1. 15-592 



2. 11-596 



3. 10-702 



4. 10-004 



5. 8-965 



§ 8. The ellipsoids were now subjected successively to in- 

 creasing forces magnetizing always in the same direction, 

 which continually raised the residual moment to greater values, 



Wiedemann* has already remarked that, when a steel bar 

 is for the first time magnetized by galvanic currents, the tem- 

 porary moments generated in it during the action of the cur- 

 rents increase at first more quickly than the intensity of the 

 currents, and only later commence to approach a maximum. 

 This behaviour should also take place after repeated magne- 

 tizations and demagnetizations of the bars on again magnetizing 

 them. 



Now the experiments show that no change in this behaviour 

 occurs even when the bar is not cleared of its residual mag- 

 netism after every observation, but this is continually aug- 

 mented by ascending forces. Let a series of experiments with 

 ellipsoid II. 4 serve for an example. X denotes the magneti- 

 zing force of the helix, TM the temporary moment, RM the 

 total moment remaining after the cessation of the action, con- 

 * Galvanismus, 2nd ed. vol. ii. p. 350. 



