596 WRIGHT: MEASUREMENT OF STRAIN IN GLASS 



polariscope, upon light crossing the Une of strain at right angles 

 is due exclusively to strain-generated change of velocity of the 

 component ray which has its plane of polarization parallel to 

 the line of strain;" that in the case of plates cut parallel with, 

 or normal to the direction of strain, the difference of retardation 

 for obliquely incident rays varies directly as the cosine of the 

 angle of refraction. This relation, in particular, indicates how 

 closely similar in its optical behavior a strained glass plate is to 

 a uniaxial mineral. S. Czapski^ measured the relative strain 

 in glass rods by a dioptric method based on Brewster's observa- 

 tion that a cylindrical rod behaves optically as a meniscus lens. 

 F. Pockels^ studied, by means of a Jamin interferential refractor 

 and a Babinet compensator, the optical effects produced in a 

 glass plate by elastic deformation of given magnitude. He 

 found also that a lead silicate glass high in lead shows anomalous 

 optical behavior in that it acts under load as an optically positive 

 substance; also that for each color of light there is a composition 

 of lead silicate glass for which no optical effect is produced either 

 by compression or dilatation. Recently E. Zschimmer and H. 

 Schultz^ have investigated, by means of a Babinet compensator 

 in combination with an interference method based on Lummer's 

 theory of interference curves of equal inclination, the optical 

 behavior of glass heated and cooled under different condi- 

 tions. Their method, like that of Czapski, is so applied that 

 it furnishes an average or integration value of the relative re- 

 tardation of the waves after transmission through any particular 

 part of the plate. The method is well adapted for the purpose 

 but the apparatus is complicated and its manipulation evidently 

 requires a considerable degree of skill. 



It is of interest to note that in practically every paper cited 

 above a new method is proposed for measuring the optical effects 

 observed. A careful study of these and other papers and also 

 of the general theory of the themial behavior of cooling glass 

 plates indicates clearly that further progress in this interesting 

 and technically important field depends chiefly on careful quanti- 



7 Ann. d. Phys., 42: 319-331. 1889. 



8 Ann. d. Phys., 7: 745-771. 1902; 9: 220-223. 1902. 

 3 Ann. d. Phys., (4), 42: 345-396. 1913. 



