no OSCILLATIONS OF A VISCOUS SOLID 



account will be communicated to the Royal Society of 

 Edinburgh. 



The wire to which the diagram refers was one of soft 

 copper, No. 19 B.W.G., and 22*5 cm. in length. The oscillator 

 was in the form of a brass ring, from which brass teeth, arranged 

 at equal angular intervals, projected downwards. These 

 teeth, as the ring revolved, made contact with radial mercury 

 pools in an ebonite plate below. These pools were also 

 arranged at equal angular intervals, but the interval between 

 them was different from that between successive teeth, so 

 that the principle of the vernier came into play. The contacts 

 so made completed electric circuits, by means of which chrono- 

 graphic records were obtained of the instants at which the 

 contacts occurred. It was possible in this way to make records 

 of the position of the oscillator at successive intervals of 2 

 throughout its range. In the special experiment here described 

 much fewer observations were sufficient, and so records were 

 taken at intervals of 12. 



The full curve in the diagram is drawn through the recorded 

 points, and represents the course of an oscillation from the 

 first positive to the first negative elongation, negative values 

 being plotted as if they were positive. In that single half- 

 oscillation the amplitude dropped by nearly one-third of its 

 initial value. Times are represented as angles, the complete 

 time of the half -oscillation being 180. The time of the 

 inward oscillation exceeds that of the outward oscillation by 

 nearly one-fifth of the mean value of the two. 



The dotted curve represents the course which would have 

 been followed had the drop of amplitude been due to a true 

 viscous resistance. 



It follows that any representation of the outward or inward 

 motions separately, either as unresisted simple harmonic, or 

 as viscously resisted harmonic, motion, is of no real value. 

 Yet the comparative accuracy of these empirical representa- 

 tions is of some interest. If the inward motion is represented 



