, the Magnetic Pn^x-rties of Steel and Iron. 

 et VIII. From one wire of •■ Es&usb Silver steel." 

 ss were hardened as in set VII. 



X' 



in length due to the hammering was not m 

 ults of VI, VII and VIII, are represented | 



e see that a cylindrical magnet, magoet- 

 ible to retain more or less magnetism 

 when soft than when hard, according to whether the quotient 

 of its length bv its diameter is greater or less than some value 



Furthermore it is evident that the specific magnetism is a 

 continuous function of the quotient =r. 



Both of the foregoing laws have already been proven for 

 heat-hardened magnets.* 



No further experiments were made to ascertain from what 

 the value of— where the curves cross (a a ) might depend ; Ruths 

 found (« ) to vary between 30 and 40, while in the three cases 

 given here it varies but little from 41. 



With iron no crossing of the curves of the hard and soft 

 magnets was observed; it is piobable, however, that at greater 

 values of (~) the curves do cross, and I hope at an early op- 

 portunitv to study this portion of the subject more fully. 



11. Effect of healing the hardened wires. — The question 

 might well be asked if, when mechanically hardened wires 

 are softened by heating, the increase or decrease in the 

 magnetic moment, which was coincident with the harden- 

 ing, disappears. Experiments to answer this were made by 

 *"frc[iiiig uirc>. the magnetic moment of which when hard had 

 been determined, and then remagnetizing them ; the specific 

 magnetism was found to change in each case in the direction 

 which would be expected from the experiments given in the 

 foregoing pages, and the magnets were capable of having their 

 moments increased or decreased, as the case might be, by a 

 second hardening. 



* Chas. Ruths, Ueber den Magnetismus weicher Eisencylinder, Dortmund, 1876. 



