44 Stress and Strain on the Properties of Matter. [Dec. 18, 



ments on the loss of energy of a vibrating magnet, and the following 

 facts were elicited : — 



(a.) The loss of energy of a vibrating magnet, like that of a 

 vibrating iron wire, is lessened by repeated oscillations, and, after the 

 first adjustment, or after a slight jar, is lessened by rest only. Repeated 

 oscillation produces a subpermanent diminution of loss of energy. 



(&.) The diminution of amplitude due to magnetic causes only is 

 like that due to the resistance of the air, and follows the same law, 

 namely, that for different vibration-periods the diminution is inversely 

 as the period, and therefore is in this respect quite unlike the 

 diminution of amplitude resulting from internal friction. 



During the whole of the experiments a most careful watch was 

 kept upon any effect which change of temperature might produce in 

 the loss of energy, and it was discovered that whilst with wires of 

 tin, lead, aluminium, silver, platinum, nickel, unannealed piano-steel, 

 zinc, copper, brass, German silver, and platinum- silver the loss of 

 energy became greater when the temperature was raised ; with iron, on 

 the contrary, the loss of energy was diminished by the same cause.* 



The marked difference between the effects produced on the loss of 

 energy of annealed iron and of the other metals, by rise of temperature, 

 caused an extended investigation to be made with annealed iron wire 

 with the following results : — 



(a.) The loss of energy of a torsionally vibrating iron wire is 

 permanently diminished to a very large extent by repeated heating to 

 100° 0. and cooling combined with long rest. 



(6.) The loss of energy is very considerably diminished temporarily 

 by rise of temperature, the value of the logarithmic decrement at the 

 temperature 0° C. being about twice as great as the value at 100° C. 

 The loss of energy at the temperature of 100° C. of an iron wire 

 which has been repeatedly heated to 100° C. and cooled to the 

 ordinary temperature of the room is so small that it may be almost 

 entirely accounted for by the resistance of the air. 



An examination was also made of the effect of change of tempera- 

 ture on the torsional rigidity of the metal, and it was found that this 

 change could be represented by the formula — 



n =r (l--0001443^--000001580^ 2 ), 



where rt and r represented the torsional rigidity at t° C. and 0° C, 

 respectively. Both from the observed torsional rigidity and from that 

 calculated from the formula, it was found that the decrease of the 

 torsional rigidity caused by rise of temperature from 0° C. to 100° C. 

 is only about half of that got by Kohlrausch when making a similar 

 investigation. The great discrepancy above alluded to can only have 

 arisen from the different treatment of the iron previously to the actual 

 * Probably also annealed piano-steel. 



