on Rods of Iron, Nickel, and other Metals. 391 



coiled with a thick layer of swansdown cloth, and then with 

 600 turns of covered copper wire, so coiled as to have no 

 lateral component. A, 13 were connected to the ballistic 

 galvanometer, and C, D to the battery-circuit and K, &c. 

 All the former effects were produced, but in a much greater 

 degree, the deflexion of the galvanometer =510 divisions. 

 The magnetizing force due to the ring coil was so arranged as 

 to be almost equal to that used in (11) . A minute deflexion 

 could be produced on either closing or opening the battery- 

 circuit after the first effect. 



17. An iron rod 1*5 millim. long, 1 centim. diam., was 

 next coiled with a helix, and all the former effects were pro- 

 duced. The effect was not so marked as when bars of small 

 diameter were used. 



18. The remarkable feature of the phenomenon appears to 

 be that it is only on the entire reversal of the magnetism of 

 the core that the transient current is produced in any marked 

 degree. 



19. None of the effects could be got from brass or copper 

 cores, so that the current from the iron core is due to the 

 reversal of the magnetic matter of the core. In all the 

 experiments, already described, the cores were as free from 

 torsional stress as possible. Similar experiments were next 

 made on cores which had a torsional set. 



20. The marks ^ — ^ 4? ^ in the appended tables 

 indicate right-handed or left-handed torsion given to the core 

 under examination (see Table I.). The torsion was given by 

 fixing one end of the core and then rotating the other a 

 certain number of times, and then deducting the number of 

 turns in the opposite direction that the core took before 

 settling down at its " set " torsional position. It was found 

 that when the cores received 20 turns that they untwisted four 

 turns, so that the whole torsion, at the condition of " set/' 

 was 16 turns. In all experiments tabulated, the marks f 

 indicate an experiment made next after reversal of the current, 



21. In these experiments, with cores subjected to torsional 

 " set," the same markedly large deflexion of the needle of the 

 galvanometer immediately after the reversal of the magnetizing 

 current was clearly shown, but now only large in one direction, 

 the direction depending on the direction of torsion of the 

 core. 



22. In Table I. the relationship existing between the 

 direction of the magnetizing current, the iron cove current, 

 and the torsion of the core are shown. 



In the next experiments (Table II.) the current was passed 





