36 Mr. W. Sutherland on a 



I. \. m. p. I. 



71:7 2*5 37*45 7*29 586500 



0C 16° 51° QQ° 86° 103° 



t/2 2*15 2*35 2-42 2*54 2-65 



nJn Y l'O '854 -804 *727 -m 



The rigidity at 16° is 109 x 10 6 , an uncertain value. 



To get a more accurate absolute value of the rigidity of tin, 

 I carried out some experiments on a large scale on a piece of 

 tin pipe in the same manner as on the lead pipe. 



I mi p. 9. r 2 . r v 



116-8 795-5 7-29 18° -689 -429 



With the three moments of inertia 7176000, 10577000, 

 and 20684000 the periods were -851, 1-029, and 1*418 sec, 

 and the corresponding values of the rigidity are 154*3 x 10 6 , 

 156*1 x 10 6 , and 160-9 X 10 6 ; the mean rigidity of tin atl8° 

 being therefore 157 X 10 6 . This is a more reliable determina- 

 tion than the last. 



Magnesium. — The specimen used was a piece of commercial 

 fine wire. 



I. A. m. p. I. 



54-6 -25 gin. 1*74 525 gm. cm. 2 



6 21° 100° 



t/2 1*42 1-45 



Vm l'O '955 



Rigidity at 21°, 130 xlO 6 . 



To check this value of the absolute rigidity another deter- 

 mination was made on a longer wire with a heavier vibrator ; 

 I 200-6, m -891, p 1*74, I 3395, 6 21°, t/2 7*10. Rigidity 

 131 x 10 6 , a result in good agreement with the last. 



2. Compilation of Data as to Variation of Rigidity ivith 

 Temperature. — The most extensive work yet done on this 

 subject is that of Pisati (Wied. Beibl. i.), who studied silver, 

 copper, gold, aluminium, iron, steel, and platinum from zero 

 to 200°, and in some unannealed specimens up to 300° ; but 

 the results at high temperatures on unannealed specimens are 

 of no use for our present purpose, because mixed up with the 

 pure temperature effect there is an annealing effect. How- 

 ever, from zero up to 100° the proportional change of rigidity 

 of the metals with high melting-point is nearly the same in 

 the unannealed as in the annealed state. The other sources 



