428 Talbot H. Waterman 



cut at right angles to its optic axis; the second is a circular polarization analyzer. 

 This consists of a quarter wave plate and a linear polarizer in that order with the slow 

 axis of the former oriented 45° clockwise from the transmitting axis of the latter. 



When convergent polarized light passes through this optic system, an interference 

 pattern is formed. Four parameters of the pattern are directly dependent on the 

 properties of the light traversing the analyzer. The figure's contrast and extent 

 depend roughly on the intensity and percent polarization of the incident light. Its 

 colour is related to the wavelengths transmitted. Finally, the geometry of the inter- 

 ference figure varies with different kinds of polarized light whether linear, elliptical 

 or circular. 



Such patterns arise because light rays in the convergent beam which pass through 

 the crystal plate obliquely, i.e., not parallel to its optic axis, will be split up into two 

 rays by the calcite's birefringence. These undergo different retardations in passing 

 through the crystal. On emerging from it and traversing the circular analyzer, they 

 form areas of cancellation and reinforcement in a radially symmetrical pattern whose 

 relative size depends on the thickness and birefringence of the crystal. 



With monochromatic polarized light the interference pattern consists of alternating 

 dark and light concentric rings, successive light areas of the pattern corresponding 

 to differences of one wave-length in the oblique ray paths through the calcite. With 

 white light a series of brilliant interference colours in isochromatic rings will be added 

 to this basic pattern. 



If the light entering the analyzer is circularly polarized, the interference figure 

 obtained under the conditions specified will be a series of complete circular isochro- 

 matic curves, starting from a white centre spot and becoming closer and closer together 

 in the higher order rings (Fig. 1a). With elliptically polarized light these rings are 

 distorted into a series of ellipses. 



The interference figure produced by linearly polarized light is broken up into four 

 quadrants at the limits of which the isochromatic rings are interrupted and displaced 

 radially. In two diametric quadrants the arcs are moved outward and two dark spots 

 appear in their central sectors (Fig. 1b); in the other sectors the arcs are displaced 

 inward and the centre remains white. 



The breaks in the concentric rings are parallel to the e and h vectors of the linearly 

 polarized incident light. With a negative analyzer crystal, like calcite, and the circular 

 analyzer axes set as described, the e vector will be parallel to the breaks which delimit 

 the clockwise edges of the two dark centre sectors. 



It should be emphasized that this type of analyzer, although simple, has several 

 important advantages in an application like the present one. Unlike any other polar- 

 ization analyzer, the present device provides the maximum information of which it 

 is capable without the need of rotating it or moving any of its parts. The plane of 

 linearly polarized light can be observed directly by inspection regardless of the angular 

 relation between the plane of the incident light and the axes of the conjoined quarter 

 wave plate and the linear polarizer. 



This is true only when the latter two units are oriented relative to each other as 

 specified. If their axes are not aligned at 45°, or if only a linear analyzer is used with 

 the calcite crystal, the interference figure will change with the rotation of the analyzer. 

 ^s this happens, a dark cross-like isogyre will be superimposed on the isochromatic 

 lines. Its position and intensity will depend on the relation of the analyzer axis to the 



