22 



IOWA ACADEMY OF SCIENCE 



Wire Xo. 3 was first used with four 1 4 widely varying 

 periods of vibration but with a constant load and length of 

 wire. The initial twist was the same in every ease. The periods 

 were varied by changing the moment of inertia of the pen- 

 dulum. 



Figure 32 shows the result of this experiment. It was im- 

 mediately seen that the period-amplitude curves as well as the 

 period-vibration number curves were similar in the four tests. 

 Thus the period of vibration had no effect upon the state of 

 the wire. Further than this, it was found that if each curve 

 - multiplied by a certain factor the four curves fell practi- 

 cally upon a single line (see figure 32). The line has not been 

 drawn, in order that the actual position of each curve may better 

 be shown. The factors were simply the ratios of the periods 

 at a certain amplitude, to any arbitrary number, and in this 

 case they were 1. 1.3. 1.68 and 2.2. Table I below gives the 

 original periods, the periods multiplied by the factor and the 

 corresponding amplitudes. 



TABLE I. 



(l) 



(2) 



K=l 



K=1.3 



