42 



Mr. H. Tomlinson. The Influence of [Dec. 18, 



IV. "The Influence of Stress and Strain on the Physical 

 Properties of Matter. Part I. Moduli of Elasticity — 

 continued. The Viscosity of Metals." By Herbert Tomlin- 

 SON, B.A. Communicated by Professor W. Grylls Adams, 

 M.A., F.R.S. Eeceived December 9, 1884. 



(Abstract.) 



After a short account of the researches of Sir William Thomson 

 and Professor G. Wiedemann, on the loss of energy of a wire when 

 vibrating torsion ally, the anthor proceeds to describe his own experi- 

 ments on the same subject. The wire under examination was clamped 

 at one end into a stout brass block, which was secured to the extremity 

 of a strong iron bracket projecting from a wall. A wooden box nearly 

 600 cm. in length and 12 cm. square inside, protected the vertically 

 suspended wire from currents of air. The box rested upon another, 

 which measured about 40 cm. each way, provided with a glass window 

 in the front, and a door at the side, which latter was opened only 

 when it was necessary to make a readjustment of a vibrator attached 

 to the lower extremity of the wire, and capable of moving freely in- 

 side the box. The bar of the vibrator was either clamped or soldered 

 to the lower extremity of the wire, and on it were suspended two cylin- 

 ders of equal mass and dimensions ; by sliding the cylinders backwards 

 or forwards on the bar the moment of inertia of the whole vibrator 

 could be altered without changing the mass. The torsional vibrations 

 of the wire were observed by the aid of mirror, scale, and lamp, so 

 that, as the length of the wire was upwards of 600 cm., and the 

 diameter rather less than 1 mm., very small molecular displacements 

 were produced when the wire was vibrating. 



It was found necessary to allow the wire to rest, after the adjust- 

 ments had been made, for a length of time varying with the nature 

 of the metal from one day to several weeks ; as though great care was 

 taken in arranging the wire, it was impossible in many cases to avoid 

 imparting subpermanent torsion, which gradually came out under the 

 influence of rest and repeated oscillation. 



Very great care was taken both in starting the vibrator and in 

 taking the observations, the results of which showed that the diminu- 

 tion of amplitude could in most cases be determined with considerable 

 accuracy. 



The various causes of the loss of energy of a vibrating wire are 

 pointed out in the paper, and subsequent experiments proved that all 

 of these are practically neglectable, except the resistance of the air 

 and the internal molecular friction of the metal. A full account is 



