334 OPTICAL PROJECTION 



different widths at different distances. On focussing the 

 slits upon the screen, conspicuous colour will be seen where 

 the images overlap, and that these are due to interference is 

 readily shown by covering up half of the bi-prism. 



190. Billet's Lenses. If difficulty be found in procuring 

 a good bi-prism, similar results can be obtained with the arc- 

 light by employing a convex lens cut in half, and the two 

 halves separated by a small distance, filled with a strip of 

 some opaque substance. The section is of course arranged 

 perpendicularly, parallel with the slit. Such a split lens is 

 easily mounted on any sort of stand, and is used in place of 

 the bi-prism. I have not however found it give sufficiently 

 distinct fringes with the lime-light. A lens of about 2 

 inches diameter and 6 inches focus is perhaps as effective as 

 any. 



191. Diffraction. There is not light sufficient, unless 

 with a powerful arc, to show on the screen the spectra from a 

 single slit ; but gratings, either original or photographed, give 

 fine projections. Of straight-line gratings on glass, copies of 

 Norbert's 3,000 lines to the inch pattern are most generally 

 useful, but the 6,000 also gives fine projections. It is sufficient 

 to place a metal or card slit about 2 mm. wide in the optical 

 stage, and focus on the screen ; then hold the grating in the 

 rays. Two gratings crossed give beautiful crossed spectra, 

 using a small hole instead of a slit. The effect is much 

 brighter if the aperture is used on the open optical front 

 with all the light from the condensers converged upon it as 

 far as possible, or with the attachment shown in fig. 95, and 

 is focussed on the screen by the loose lens. With this aid, 

 the diagonal spectra will be seen, as well as the primary 

 crossed ones. By interposing coloured glasses the spectra 

 may be reduced to coloured images of the slit, the red ones 

 being farthest apart. 



A circular grating, used in the same way to diffract a 

 small round aperture, gives circular rainbows. I have had 



