ORDINARY MEETINGS. XXIX 



under considerable stress the wax near the ends of the glass tubes gave 

 way, and the space thus left in the ends of the tubes had to be packed to 

 make it certain that the portion of the cord actually subjected to tension 

 and torsion was the portion outside the tubes. The observations under 

 the greater stresses are thus considerably less trustworthy than the 

 others. 



The weights used were square plates of sheet lead with an edge of 

 four inches, having a small circular portion cut from the centre, and a 

 slit from the centre to the edge to admit of their being placed on the 

 holder. Their moment of inertia could therefore readily be calculated. 

 The length of the cord was measured by means of a beam compass, 

 reading to .01 inch, and its diameter by a micrometer gauge reading to 

 .001 inch. The time of oscillation was determined by means of a stop- 

 watch reading to 0.2 second. The observations given below were in all 

 cases means of several individual observations, the length and radius of 

 five, and the times of oscillation, of ten. No special effort was made to 

 keep the cord at constant temperature ; but the temperature of the 

 laboratory varied but little. 



The course of the observations was determined by Mr. Macdonald's 

 available time rather than by the fitness of things. He kept the cord 

 stretched slightly throughout the whole series by 482 grm., and from 

 time to time he would apply additional weights, make the requisite 

 observations, and then remove such additional weights, the observations 

 requiring perhaps a couple of hours. The dates of the observations are 

 given in the table below. Unfortunately the length of the cord under 

 its permanent stress was not in all cases, and its diameter was in no case, 

 determined immediately before the application of the additional weights, 

 such observations not being necessary for the purpose originally in hand. 



The following table gives the results of the observations and the 

 calculated values of the rigidity. The observations have been reduced 

 to C. G. S. units, and the rigidities expressed in absolute units of that 

 system. The rigidities were calculated from the formula : 



r* t* ' 



where n is the rigidity, I the length, and r the radius, of the cord, / the 

 moment of inertia of the lead plates, and t the time -of a complete 

 oscillation. In finding the moment of inertia the weight-holder of cork 

 and glass was neglected, as also the small circular apertures in the lead 

 plates. The volume of the cord is given in the table also. 



