368 Scientific Proceedings, Royal Dublin Society. 



A single example out of many experiments will show the 

 working of this method. A piece of manganese steel wire 2*15- 

 millimetres diameter and 44 centimetres long, was suspended (by a 

 platinum wire attached to a torsion head) in a uniform and con- 

 stant magnetic field. On turning the torsion head, the upsetting 

 angle was found to be 60°, and the torsion required for this was- 

 158°. Hence 158° - 60°, or 98°, is the actual force of torsion em- 

 ployed. With a piece of fine iron wire of precisely the same 

 length and 0*2 millimetres diameter, the upsetting angle was 70°, 

 and the torsion required 320°. Hence 320° - 70°, or 250°, was 

 the force required in this case. The ratio of the forces of the two 

 "bodies are therefore as 1 : 2-5. The ratio of the volumes of the 

 two substances will be as the squares of their diameters, or as 

 1 : 115. Assuming the magnetic moment increases as the volume 

 of the bodies, the ratio of the forces multiplied by the ratio of the 

 volumes will express the ratio of the susceptibilities of the two- 

 bodies, which gives 



1 : 287. 



There are, however, some objections to this method of experiment,. 

 as the upsetting angle is not the same, and hence the magnetic 

 distribution at two different angles will not be alike in the two* 

 cases. 



The following method is free from this objection. The force re- 

 quired to turn each of the two substances through a given very small 

 angle, when they are suspended in a magnetic field of constant 

 strength, is found: this value (less the angle of deviation), multiplied 

 by the ratio of the volumes of the two bodies, will give the number- 

 sought. A mirror was attached to the cradle supporting the body, 

 . and by means of a lamp and scale, a very accurate measure of the 

 angle through which the substance was turned could be obtained. 

 A constant current of 1\ amperes was used to magnetize the 

 electro-magnet, a uniform field being obtained between two large 

 upright pole pieces. The torsion required to turn the manganese 

 steel through 18° was in one experiment 42°*5, and in another 42° - 3, 

 or a mean of 42 "4°. The torsion required to turn the fine iron wire 

 through the same angle was 94°. This, less the angle of deviation, 

 gives a ratio of 1 : 3. In a more powerful field the numbers were 

 66° and 68°, or a mean of 67° for the manganese, and 155 c '5 and 



