Non-reversed Spectrum Interferometry . . 413 



In another somewhat thinner Babinet compensator, the con- 

 stants were : Length, 3-35 cm , thickness, small end -494°'", large 

 end "496 om ; the prism angle is a = -062/3-35 = -0185 radians, 

 also about 1°. 



In this case there was no periodic phenomenon, but in its 

 place the degree of longitudinal coincidence of the axes of the 

 two spectra continually changed. The fringes at once sharp- 

 ened, however, on readjustment of either mirror, indicating a 

 continuous small change of deviation, due to curvature, prob- 

 ably, in the quartz wedge. In the preceding periodic case, no 

 readjustment of deviation sufficed to restore the fringes. The 

 wedge was now detached and used alone. In spite of the 

 relatively large angle (1°), no difficulty was experienced in 

 adjusting or controlling the fringes; but the face curvature 

 just suggested appeared as before, so that readjustment for 

 varying wedge angle was required from time to time. 



8. Micrometer displacement of the second grating. — In the 

 preceding paper* it was shown that if the angle between the 

 gratings G and G' is (f> and the angle between the mirrors 

 M and N (which in a symmetrical adjustment would be 

 180° - (0, + 0„), 0, and 2 being the angle of diffraction at 

 G and G' for normal incidence at G) is decreased by a, so that 

 the adjustment is non-symmetrical, then the displacement &?of 

 the grating G' , per fringe, will be very nearly 



8e= AC0S '** 



2(a — <f>) sin 0, 



if a and <£ are small. Here a is effectively the angle between 

 the mirrors M and JV ; since, if Ji"is rotated 180° on the line 

 of symmetry (normal to the grating G), the two mirrors would 

 intersect at an angle a. The result of fore and aft motion thus 

 depends on the angle a — <j>, and if a = <j>, 8e=cc, per 

 fringe; i. e. fore and aft motion would produce no result. This 

 is necessarily the case when but a single grating is used, as in 

 the earlier methods. In the case of two gratings, however, it 

 is not only difficult to make a perfectly symmetrical adjust- 

 ment of mirrors and grating, but it would not be of any special 

 advantage. Hence the fore and aft displacement e of the 

 grating G' will probably be accompanied by a slow motion of 

 the fringes, from which the angle a — <f> may be computed. 



The following experiments were made with the grating G' 

 on a micrometer slide, moving normally to the face of the 

 grating. With the mirrors, etc., placed so that optical paths 

 were nearly equal, the adjustment screws on M and JV sufficed 

 to bring the fringes strongly into view. Successions of 3 and 



*This Journal, xlii, p. 71, 1916, §4. 



