Brown — Mechanical Stress and Magnetisation of Nickel. 37 



to shift the maximum to a greater field. As already remarked, these 

 experimenters used a large initial amplitude of oscillation ; and though the 

 actual values of tlie weights are not mentioned, the maximum effect for the 

 light weight took place in a magnetic field of about 93 units, and for the heavi/ 

 weight in a field of about 132 units. The fact, also, that in their experi- 

 ments, the wire under test project;ed at the ends of the solenoid from 

 1 to 4 centimetres, leaving perhaps not more than two-thirds of the length 

 of the wire in a uniform magnetic field, may have had some effect on the 

 final results as well as the large amplitudes employed. 



Oj 09 



c o 

 o o 

 'S " 



is o 



150 



100 



.5 a 



S-5 50 



I-SSIO^ 



3 0X10^ 



4-0X105 



10 20 30 40 50 60 



Fig. 3. — Longitudinal Magnetic Field. 



70 



50 



In order to test the effect of a magnetic field on the internal viscosity of 

 a harder nickel wire, experiments similar to the above were made witli a wire 

 of simple rigidity 756 x 10^ grammes per square centimetre. This wire 

 was also 226 cms. long and 0-168 cms. in diameter, and was hardened by 

 hanging it in a vertical position with a weight on its lower end equivalent to 

 1-5 X 10' grammes per square centimetre ; and then raising it to a bright 

 cherry-red heat by means of a broad Bunsen burner, and so the wire became 



