the Galvanic Current through Iron. 147 



is possible with the average insignificance of the deflections in 

 general occurring, also quantitatively) verified. 



I have characterized the extra currents in iron as the ex- 

 pression of the work of rotation, and described this work by 

 an increase or a diminution of resistance. The value of the 

 resistance corresponding to the closing current does not sud- 

 denly change into the value conditioned by the passage of a 

 constant current or a permanent magnetization (or both 

 together), but is mostly connected with it by a 'phenomenon of 

 afteraction. Even of the nature of this, from the above an 

 idea can easily be formed. If, beside the direction-force, only 

 the principal current acts, it imparts to the molecules a rota- 

 tion-velocity which, according to the value of H, is constantly 

 positive (rotation) or periodically changes its sign (oscillation). 

 ISTow, if even in a magnetic needle moving in a copper shell 

 we observe a rapid conversion of the motion of the mass into 

 thermal motion, the same conversion will result much more 

 rapidly still in a molecule of an iron wire. A heating there- 

 fore takes place, which is only gradually equalized by conduc- 

 tion and radiation ; and the expression of this heating is the 

 phenomenon of afteraction. In fact, heat is generated not 

 merely by longitudinal, but (as Villari* has shown), also by 

 circular magnetizing. Into this I do not further enter; I will 

 only mention that the series of experiments 16-18 of § 4 and 

 the remarks (p. 16) in § 5 apply here. That in those experi- 

 ments not only D and H, but also M acted, has, it is self- 

 evident, no essential influence upon the result. I add a few 

 numbers characteristic of the course of this kind of afterac- 

 tion. 



(1) M = 2B, H=1D, w=150. These numbers were chosen 

 so that 8 was nearly =0. Wire / 4 . The principal current 

 w r as closed after the magnetizing current, t denotes the time 

 between two observations. On account of the considerable 

 distance of the magnetizing current, some seconds mostly 

 elapsed between the closing or opening of the magnetizing cur- 

 rent and the first observation ; hence the absolute alteration 

 of resistance in consequence of the rotation-work cannot be 

 inferred from the numbers, s denotes closing, 6 opening of 

 the magnetizing current. The ratios are graphically repre- 

 sented in PL I. fig. 3 ; the reinforced parts of the drawing 

 correspond to afteractions. 



It will be seen that at the opening the duration of the after- 

 action is less than at the closing. 



* Xuovo Cim. (2) iv. Nov.-Dec. 1870. 



