Be'own — The Fatigue of Nickel and Iron Wires. 169 



The results are better seen in fig. 3 in the form of curves, where, as before 

 L and T indicate the curve showing, the effect of the longitudinal and the 

 transverse fields respectively. 



The results given in the present paper complete the work (as far as 

 the torsional oscillations and fatigue of nickel and iron wire are concerned) 

 which the author set himself to do a few years ago. There are one or two 

 things arising out of the work that are not important enough to form 

 the subject of a separate paper, but which for the sake of future workers 

 should perhaps be put on record ; I have, therefore, attached them here as 

 appendices. 



Appendix I. 



Subsidence of torsional oscillations in a wire when carrying an electric 

 current. 



i. Nickel Wire. — A No. 16 s.w.g. nickel wire of rigidity about 715 x 10 6 

 grammes per sq. cm. with a longitudinal load on the lower end of 10 5 

 grammes per sq. cm. was tested for damping or subsidence of torsional 

 oscillations in direct and alternating magnetic fields both longitudinal 

 and transverse, of value 17 units, the alternating fields having a frequency 

 of 50 per second. The increased or decreased values of clamping mentioned 

 below for both nickel and iron are arrived at by comparing the amplitude of 

 the 70th oscillation in each set of observations. 



The following series of experiments were made : — 



A. When the wire was under the influence of the vertical component of the 



earth's magnetic force only. 



B. When the wire was in a direct longitudinal field. 



C. When the wire was in an alternating longitudinal field. 



D. When one ampere (direct) was through the wire and a direct longitudinal 



field round it. 



E. When one ampere (direct) was through the wire and an alternating 



longitudinal field round it. 



F. When one ampere (alternating) was through the wire and an alternating 



longitudinal field round it. 

 G-. The same as in (F) with a transverse alternating field superposed. 

 Comparing results obtained under conditions : — 



A and B, in B the clamping is increased 4A per cent. 

 B and D, in D the damping is decreased 12 per cent. 

 C and E, in E the damping is increased 3 per cent. 

 F and G, in G- the damping is decreased 4 per cent. 

 C and F there is no change. 

 ii. Iron Wire.— A No. 16 s.w.g. iron wire of rigidity about 810 x 10 6 

 grammes per sq. cm. with a load of 10 5 grammes per sq. cm. was put through 



