Brown — The Fatigue of Nickel and Iron Wires. 165 



the lower side of this non-magnetic load is fixed a vertical steel pin, which 

 dips into a basin of mercury, thus enabling a current to be sent through the 

 wire T. 



There are three independent electric circuits to be dealt with in the 

 experiment, viz. (1) an alternating current circuit through the wires C; (2) a 

 direct current circuit through the solenoid S ; (3) a direct current circuit 

 through the wire T. 



The method of operation in the experiment is as follows : — a certain 

 steady direct current is sent through the solenoid S, which gives a longi- 

 tudinal magnetic field round the wire T; another steady direct current is 

 sent through the wire T, which gives a circular magnetic field, and the 

 combination of these two fields causes the lower end of the wire T to twist. 

 This initial twist is measured by the movement of the light spot on the scale, 

 and will be referred to as D. 



The circuits through S and T are then opened, and the circuit through C 

 closed for say one minute, i.e. the wire T has been subjected to the influence 

 of a transverse alternating magnetic field for one minute. The circuit through 

 C is then opened, and the circuits through S and T closed, with exactly the same 

 values of the currents through them as before ; the twist produced now, as 

 measured on the scale, will be found to be less than its initial value D, and 

 will be referred to as d. 



The closing and opening of the several circuits, and thereby the putting 

 on and off oi the separate currents, is continued until the deflection of the 

 light spot on the scale is no longer diminished. 



Now if F= the fatigue of the wire, 



D = the unfatigued steady deflection or twist, 

 d = the fatigued steady deflection or twist, 



The former papers referred to above show that a good deal of work has 

 been done on the fatigue of nickel and iron wires due to longitudinal magnetic 

 fields, when in different states of rigidity, as well as when under different 

 longitudinal loads. In the present case, therefore, it was thought to be 

 sufficient to test only one nickel and one iron wire for fatigue due to transverse 

 magnetic fields. For the sake of comparison the identical wires that were 

 tested with longitudinal fields were tested under exactly the same conditions 

 with transverse fields. 



2 e 2 



