Notes. 155 



central band, the second maximum occupies a position between 

 A and B only in the case of the upper grating ; in the case of the 

 lower grating it occupies a position on the first band, indicated by 

 the dotted intensity curve of the single slit. 



Fig. 9 is a photograph of the grating, all spectra being excluded 

 from taking part in the image except those underlined a in fig. 7, 

 viz. the zero and the first maximum on the right-hand side. 



Observe that the relative position of the upper and lower set of 

 lines remains unchanged, which indicates that, so far as these two 

 maxima are concerned, there is merely the normal phase -difference, 

 both as regards the upper and the lower half of the grating. 



Fig. 10 shows a photograph of the grating, all spectra being 

 excluded from taking part in the image except the underlined 

 portion b in fig. 7, viz. the first and second maxima on the right- 

 hand side. 



Observe that the lines of the lower set have got shifted, so that 

 the black and white lines run right through from top to bottom. 

 This is because the spectra furnished by the upper half have merely 

 the normal phase-difference, whatever that may be, whilst those 

 furnished by the lower half have an additional difference of half a 

 phase-period. On referring to fig. 8, it will be seen that this is 

 explained by the fact that in the case of the lower grating the second 

 maximum occupies a position on the first band which a single one 

 of its slits would have formed, and, as we learnt above, a change of 

 phase occurs at B. 



Fig. 11 shows a photograph of the grating, all the spectra being 

 excluded except the underlined part c in fig. 7, which comprises 

 the first and second maxima and a small part of the zero maximum. 



Observe that this photograph shows the transition stage from 

 fig. 9 to fig. 10. 



As regards the lower half, the combined influence is seen of the 

 zero and first maximum being normal as regards phase-difference, 

 together with the influence produced by the first and second, and by 

 the zero and second maxima, having an extra difference of half a 

 phase-period. 



As regards all the spectra furnished by the upper half of the 

 grating, everything is normal. 



The effect shown in this photograph is only obtainable by 

 adjusting the part of the zero maximum admitted very carefully, as 

 the change in appearance from that in fig. 9 to that in fig. 10 is 

 comparatively sudden. When a considerable part, or the whole, of 

 the zero maximum is allowed to pass along with the first and second 

 maxima, the effect is essentially as seen in photograph 9. 



Mr. Conrady's experiment, here recorded photographically, con- 

 stitutes, I think, striking evidence of the influence on the images 

 of gratings of the phase-differences amongst their spectra, clue to the 

 relative width of the slits to the bars — a new point in the theory of 

 the Microscope image of considerable importance. 



M 2 



