u6 THE DISSIPATION OF ENERGY 



and the average of these values was then taken. A curve 

 drawn with the scale readings as ordinates and the number 

 of swings as abscissae showed by means of the waviness of 

 the curve if ordinary pendulum oscillations had been appre- 

 ciably started in the apparatus. The values of log y were 

 then plotted against those of log (x+a), and when the proper 

 value of a had been found, so that the points lay practically 

 on a straight line, the constants were obtained. 



CONFIRMATION OF THE EMPIRICAL LAW 



Wires of nine different metals were tested, brass, copper, 

 aluminium, tin, zinc, silver, german silver, platinum, and 

 nickel. Of these, brass, tin, zinc, silver, german silver, and 

 nickel were found at the ordinary temperature to give close 

 agreement, over the very large range of oscillations taken, to 

 the general law, in each case a suitable value of a being found 

 which caused all the points to lie on a straight line. It was 

 found, however, that in the case of the remaining metals, 

 and especially in the cases of aluminium and copper, no one 

 value of a could be found to bring all the points into one line, 

 an 5-shaped curve being obtained in general. When this was 

 first observed, it was thought that the law did not hold in 

 such cases, or at least that it did not hold over the range taken. 

 In attempting to straighten, in this case, one part of the 

 curve, however, it was found that, with a certain value of a 

 the points could be brought to lie on two straight lines in- 

 clined at an angle not differing much from 180. It was 

 further found that this could not be done in every case with 

 the same value of a for the two portions, but, by choosing a 

 slightly different value of a, in every case the points could be 

 brought to lie on two straight lines. The doubling of the 

 line, as will be seen when the metals are considered separately, 

 was found to depend upon controllable conditions, e.g. in 

 brass it occurred when the metal had been brought to a 

 certain temperature in the neighbourhood of 375 C. In 



