Determining the Elastic Constants of Glass. 557 



method changes of curvature were observed well outside the 

 limits of error of the measurements. 



An attempt was made to confirm these results by deter- 

 mining the change in the value of Poisson's Ratio by 

 measurement of the angle between the asymptotes. The 

 photographs, however, did not enable the angular measure- 

 ments to be made to within more than *5 degree, and no 

 definite evidence of alteration could be found within these 

 limits. This was only to be expected, as a 1 per cent, change 

 in the value of Poisson's Ratio corresponds to only 6 minutes 

 difference in the angle between the asymptotes. Straubel 

 employed much thinner glasses, giving greater curvatures 

 and consequently sharper fringes, and so was able to measure 

 his angles much more accurately. He did not, however, 

 observe any time effect in his experiments. 



The tables which follow give the results of the observations 

 of strain-creep. It was found to be impossible to obtain 

 reliable results for very small loads or for beams whose ratio 

 of width to thickness was small, as in these cases the 

 curvature produced showed too few fringes for an accurate 

 determination of the cross-bending. 



Method of Redwing Observations. 



<?: 



pU-- 



A 



n ~ 



-->\p, 



:<?/ 



V i i 



\ys 



l"*""""* 



>^^ 1 







i ^ 0tt ^ 



-^i 



1 











! 



c 



Arr+S. 



If P, P x represent the positions of the ni\\ pair of fringes 

 counting from the centre, and the length P Pj be An, then 

 if R be the radius of curvature of the surface of the beam, 



2Rx BC = £ A„ 3 . (Neglecting BC in the expression 



2R-BC.) 



Also 2RxAC = iA, 1+s 2 , 



2R(AC-BC) = i(A„ + . s . 2 -A„ 2 ), 



2Rx^\=i(A„ + . s 2 -A„ 2 ), 



»-i- 



(A»+, — A H ") ; 



whore \= the wave-length of light employed. 



This gives a measure of the radius of curvature of tin 4 beam 

 in the region between the fringes P and Q. In order to 

 reduce the effect of errors in measurement o^ the fringes. 



