WRIGHT: MEASUREMENT OF STRAIN IN GLASS 



597 



tative measurements and that therefore a simple method for 

 measuring relative optical retardations accurately should be of 

 value. The method presented below is suggested for this purpose. 

 In this method the glass plate is examined in strong monochro- 

 matic light between crossed nicols; the path difference of the 

 emergent light waves at any given point is measured by means 

 of a bi-quartz compensator of special type.*" 



The most satisfactory source of monochromatic light avail- 

 able is the green line (546 hij.) of a strong mercury lamp, as viewed 



through a Wratten and Wainwright special filter No. 77. The 

 rays from the lamp are rendered parallel by means of a con- 

 denser lens before entering the polarizer. For certain purposes 

 direct sunlight or an arc light may be used and gives results 

 which are sufficiently accurate. 



The bi-quartz compensator consists of two quartz plates and 

 two quartz wedges cut parallel to the principal axis and mounted 

 as indicated in figure 1 a. When viewed in the diagonal position 

 under crossed nicols this wedge presents the phenomenon of 

 figure 1 b, namely a dark, transverse band of exact compensation, 

 on both sides of which the interference colors rise to gray, or 

 other color of the first or higher order depending on the pitch 

 and length of the wedge. The wedge is so ground that gradua- 

 tions on the upper surface of the compensator indicate directly 

 the path differences in ixfx for the wave length (546mm) of light 

 employed. As indicated in figure 1 the bi-compensator is so 



10 F. E. Wright, Amer. Jour. Sci., (4), 26: 370-371. 

 of Washington, Publication 158: 134-135. 1911. 



1908; Carnegie Institution 



