Intellijinice and Miscellaneous Articles. 4()i) 



Let us now consider the case in which the system formed of a 

 tube and a core of steel in the neutral state is magnetized by Elias's 

 process. If the magnetization is effected at the ordinary tempera- 

 ture, we find that the tube and core, separated from one another, 

 are magnetized in the same direction ; this was veriiied by jM. Jamin 

 {Comptes Rendus, Feb. 15, 1875). The same result is again ob- 

 tained when the system is magnetized at a high temperature 

 (300-400"), and the tube and core separated immediately after the 

 magnetization is effected ; but when, after having magnetized the 

 system while hot, we leave the tube and core to cool in contact 

 w ith one another, we lind, on separating them when quite cold, that 

 they are generally magnetized in directions the one the inverse of 

 the other ; it is only in a particular case that they are both magne- 

 tized in the same direction. The sign of the magnetization varies 

 with the thickness of the tube, the coercive force of the steel, and 

 the intensity of the current used for developing the magnetization. 



I have performed a first series of experiments, on tubes whose 

 thicknesses were |, |, |, and 1 millim. ; each of these tubes was 10 

 millims. in exterior diameter, and about 300 millims. in length. 

 They had been drawn and manufactured, as well as the cores with 

 which they were supplied, of steel known in commerce under the 

 name of " soft Petiu-Grodet." 



The following Table contains the results obtained in operating 

 upon the tube half a millimetre in thickness. 



I. 



M. 



M'. 



m. 



m'. 



3-4 



+ 4-6 



+ 8-0 



- 1-5 



+ 5-0 



7-5 



+ 20-0 



4-22-2 



-f- 2-D 



+ 2-5 



14-5 



+ 56-0 



-h21-0 



-1-17-2 



-15-0 



20-0 



-h80-0 



+ 27-0 



+ 28-2 



-25-1 



29-0 



+ 80-0 



-h30-2 



+ 28-0 



-28-0 



38-0 



+ 87-0 



+ 34-0 



-h24-0 



-29-5 



The numbers in the column marked I represent the intensities 

 of the current made use of to effect the magnetization. The letters 

 M and M' designate respectively the values of the magnetisms of 

 the core and the tube when these are measured immediately after 

 the magnetization effected before the cooling ; while m and m de- 

 note the values of the magnetisms of the core and tube respectively, 

 measured after the cooling of the system. I regard as positive the 

 direct magnetization, and as negative the inverse. 1 call direct 

 magnetization that which would be communicated by the current 

 at the ordinary temperature, whether to the tube or to the core. 



On inspecting the Table we see that the magnetism m of the core, 

 at first inverse for the very feeble current I=:3-4, becomes direct 

 when the intensity of the current augments, and that, on the con- 

 trary, the magnetism m of the tube, direct with the currents 3*4 

 and 7*5, becomes inverse for currents of greater intensity. 



These results may be regarded as consequences of a more 

 simple fact, which I have noticed at the commencement of this 



