402 JVlessrs. Kei lokibe and Sukeaki Sakai : Effect of 



glass tube V from a Kipp's apparatus. The temperature was 

 measured by means o£ a platinum and p'atinum-rhodium 

 thermo-couple placed at and near the middle of the wire, and 

 a milli-voltmeter. 



The moment of inertia of the oscillating system — without 

 the lead ring — was experimentally determined. The moments 

 of inertia of the two flat lead rings were calculated to be 

 25278 and 17780. If the period of oscillation of the system 

 with and without the lead ring be respectively T and T , and 

 the corresponding moments of inertia I and I , we have the 

 relation 



T — T ^O 



x - 1 - rpa rp 2 * 



By using this relation, we found experimentally 



I =8080gr.S?. 



In measuring the period of oscillation the usual method of 

 timing was followed. One of the observers signalled at each 

 complete oscillation, while the other recorded the time by a 

 mean time chronometer. Forty consecutive oscillations were 

 employed for the determination of the period. The first and 

 the last 10 were signalled and recorded; we thus obtained 

 10 sets of time-records for 30 complete oscillations. The 

 mean of these observations gave the period of oscillation, 

 which is accurate to one-thousandth of a second. 



The double amplitude of oscillation chosen for the deter- 

 mination of rigidity was about 45 cm. of the scale reading ; 

 it corresponds to an angle of twist V per unit length of the wire. 

 In the case of metals having high melting-points, the decrease 

 of the amplitude of oscillation at room temperature was very 

 small, and hence the determination of the period could be made 

 very accurately. But in the case of the same metals at high 

 temperatures, or of soft metals even at room temperature, the 

 decrease of amplitude was considerably large, so that the 

 determination of the period of oscillation cannot be accurate. 

 As we have already remarked, the damping of the oscil- 

 lation is caused by the viscosity of the air and that of the 

 wire. In order to obtain the coefficient of viscosity of solid 

 metals, it is therefore necessary to determine separately the 

 damping due to the viscosity of the wire. For this purpose 

 an apparatus as shown in fig. B was used. The suspended 

 system is made to oscillate in a vacuum. S is the test- wire, 

 BB the brass rods firmly connected with the wire, M a 

 plane mirror, IT a pan, and L a lead ring, as in the former 

 case. The whole is enclosed in a bell-jar P with a stout 



