accompany the Magnetization of Iron, Nickel, and Cobalt. 53 



mens I have met with, has very slight retentive power when 

 magnetized, whereas the cobalt has a high degree of coercive 

 force. 



I. 

 Enclosing either of the bars within a helix of wire, a sound 

 was emitted as soon as an interrupted current traversed the 

 helix. The sound with cobalt was far the more powerful of 

 the two, and was even more pronounced and more metallic than 

 with a corresponding bar of iron. This fact, I believe, has not 

 been noticed before. It is easy to obtain these sounds by 

 merely using the coil of an electromagnet and drawing the ter- 

 minal of the battery wire over a coarse file in a distant room. 



II. 



In order to examine whether the metals lengthened by mag- 

 netization, I had a special apparatus made for me by Messrs. 

 Yeates, of Dublin and London. The instrument is a modifica- 

 tion of the arrangement used by Dr. Tyndall, and described by 

 him in his { Researches on Diamagnetism ' (p. 240). Instead 

 of being mounted vertically, the iron bar in my instrument is 

 placed horizontally within the coils of a powerful electromagnet. 

 One end of the bar is rigidly pressed by the end of a micro- 

 meter-screw, which is mounted on a sliding brass support that 

 can be adjusted to any length that the bar under experiment 

 may be. The other end of the bar presses against a system of 

 levers, by which the least motion of the bar is largely multiplied. 

 On an axle moved by the last lever a mirror is fixed ; and upon 

 this a beam of light is thrown, the reflected image being received 

 on a distant scale. The whole is mounted on a firm mahogany 

 base rather more than a metre long. The momentary warmth 

 of a lighted match against the iron bar drives the reflected image 

 through 2 or 3 feet of the scale. 



The arrangement is shown in fig. 1 . The soft iron bar, 1 1, 

 Fig. 1. Fig. 2. 



r 



