346 



C. Barns — Resistance of Stressed Glass. 



a/b=r/R, where r is a known rheostatic resistance, and R 

 the resistance of the tube, the current in the galvanometer 

 being nearly zero. Then if drr produce the same maximum 

 oscillation of the needle as oR B, it follows that dr/r=dR/R. 

 An accurate chart or table of or r considered as a function of 

 the oscillation is therefore first to be constructed, by aid of the 

 rheostat. This being in hand, the value of dR/R correspond- 

 ing to any oscillation produced by alternately adding and re- 

 moving the load on the tube, is given at once. This method 

 may be made very accurate, and I was able to obtain not only 

 traction effects, but torsion effects as indicated by the following 

 data. Here r is approximately 53000 ohms, R approximately 

 5900000 ohms. In case of torsion, P denotes the load acting 

 during the alternate twisting and untwisting. 



Table II. — Resistance of stressed glass at 100°. 



Preliminary. 





Torsion. 







Traction. 



10 3 x dr/r 



Oscillation. 



P 



Oscillation. 



10 3 



x6R/R 



P 



Oscillation. 



10 s x 6R/R 





cm. 



kg. 



cm. 







kg. 



cm. 





•76 



•35 



5 



•59 





— 1-3 







•22 



— •5 



1-89 



•89 



10 



1-27 





-2-8 



5 



•29 



-•6 



3.77 



1-63 



16 



1-86 





—4-1 



10 



■28 



-•6 



The traction data dR/R are numerically larger than in Table 

 I, and hence lend greater favor to the views just expressed. 

 The torsion data dR/R are of the same sign as the traction 

 data. In other words torsion increases the electrolytic resist- 

 ance. They are of smaller magnitude than traction data and 

 are independent of the load which the tube sustains, so far as 

 I could follow them. 



Table II.—. 



Longitudinal extensioi 



of the tubes, Table 1st. 



Temperature. 



Load. 



L 



6L/L 





°C. 



kg. 



cm 



10- 6 x 





16° 



2 



77 



53 











6 



77 



53 











10 



77 



54 



130 







15 



77 



55 



260 





100° 



2 



77 



58 







Coefficient of expansion -000008 





6 



77 



58 











10 



77 



60 



260 







15 



77 



60 



260 





100° 



2 



77 



57 











19 



77 



59 



260 





100° 



2 



77 



57 











19 



77-60 



390 





8. To interpret the above data, special measurements of ex- 

 tension are necessary. These are given in Table II. They were 



