OF VULCANIZED INDIA-RUBBER HEBB. 277 



expressed as in the previous formula, / is moment of inertia 

 expressed in Ibs. and inches, and t is the time of a complete 

 oscillation expressed in seconds. The moment of inertia of the 

 plates used in stretching the cords was found by means of the 

 following formula : I = M (a 2 6 2 ) /1 2, in which M is mass of 

 plates in Ibs. and a and 6 are the lengths of sides of plates in 

 inches. The moment of inertia of the brass tube at the end of 

 the cord was found to be negligible. 



The values of the rigidity determined as above would thus 

 be expressed in inch-lb-second gravitational units. 



The effects on the calculated values of the rigidity, of the 

 above possible errors of the component observations, were 

 calculated in a few cases, and were found in the static observa- 

 tions to be between 1.5% and 2.5%, and in the case of the 

 kinetic observations to be between 2.5^/ and 3/. 



My first observations were made on the cord which Mr. 

 Macdonald had previously used. He had subjected it to repeated 

 and prolonged extension, and found that the minimum point 

 which in the earlier series of observations seemed to characterize 

 the variation of its rigidity with tension finally disappeared. 

 Since the time of his experiments it had lain for six months 

 unstretched. 



The static method was the only one employed, the torque 

 being applied by means of the frictionless wheels. The proce- 

 dure was as follows : First a weight was hung on the cord, and 

 left there for at least a day. Then its rigidity was determined, 

 after which the load was increased and the cord left 

 for another day. On the third day the same process was 

 repeated, and so on. It was not until later on in the 

 experiments that the importance of allowing the loads to act 

 for corresponding intervals of time was realized. The following 

 table rives the results obtained : 



