620 



ADAMS AND WILLIAMSON : BIRE^FRINGBNCE AND STRESS 



in Equations (i) to (7) of this paper. Substituting these values 

 in Equation (4) we have calculated the birefringence fiy - n^ 

 caused by a thrust of i kg. per sq. cm. and have placed the values 

 obtained by this calculation in the last column of table 3. 



We are now able to make a comparison between our results 

 and those of Pockels. The first three glasses in table 3 are 

 unusual alumino-borates, but the remaining four form a series 

 of flint glasses with lead content varying over a wide range. 

 In Fig. 4, for the sake of comparison, our results for the flint 



TABLE 3 

 Results of Pockels' Measurements on Optical Effects op Stress 



Kind of glass 



a 



3.6 

 o 



Sodium Alumino 

 borate 



Lead Alumino 

 borate 



Lead Alumino 

 borate 



Light Flint 



Heavy Flint. . . . 



Extra Heavy Flint 



Heaviest Flint. . 



42 



21 



47 

 26 



3Z 

 20 



Birefrin- 

 gence due 



to 1 kg. 

 per sq. cm. 

 calc. from 



Equation 

 (4) 



—4.32 

 -2 .76 



—3.78 



—2.93 



2 .61 



— 1.38 

 I .92 



glass and also those of Pockels have been plotted with PbO 

 content as abscissa and birefringence due to i kg. per sq. cm. 

 as ordinate. 



By inspection of Fig. 4 it may be seen that Pockels' results 

 for the flint glasses on the whole agree very well with ours. It 

 should be noted, however, that the content of PbO does not com- 

 pletely determine the character of the glass and that therefore 

 certain small discrepancies between Pockels' results and ours as 

 indicated by Fig. 4 may not have a real existence. 



It is interesting to note that ordinarily q is less than p. When 

 this is the case the birefringence «„ — w^ by Equation (5) is 

 negative. But for the heaviest flint (PbO = 80 per cent) in- 



