ON DIFFRACTION SPECTRA IN THE MICROSCOPE. 3I 



the microscope, and i, d, with wider slit for the group of narrow 

 lines.* When the slits are turned parallel with the lines on the 

 object, the central or direct image alone, is seen ; and when the 

 slit crosses the lines at right angles, the respective diffraction 

 spectra are seen together with the central image. It is necessary 

 to remove the eyepiece and look down the tube in order to set the 

 position of the slit accurately Supposing now that the diaphragm 

 is turned to the position at right angles to the lines and the eyepiece 

 replaced, the lines, will, in each case, be seen as they are known to 

 be ruled, but on turning the slit into parallel position with the 

 lines the^ disappear entirely. This happens obviously, because, in 

 the parallel position of the slit, only one image is admitted, 

 whereas in the crossed position the central and neighbouring 

 spectra work together, and the fine detail of the ruled lines is 

 immediately re-produced in the microscope image ; whilst with 

 the single direct image, the whole group of lines appears as a 

 plain gray band. This proves that a central pencil will not resolve — 

 that is, render visible — any detail as fine as the coarsest of these 

 groups of lines, without at least one diffraction spectrum : as is 

 shewn, more strikingly still, in 



Experiment III. — 



For this the diaphragm i, e, must be employed, and focussing 

 with the same precautions as before, while the centre of the group 

 of lines is brought into the field, both sets of lines will be seen so 

 long as the slit in the diaphragm lies across them, and diffraction 

 spectra admitted. But on turning the slit into a position parallel 

 with the lines it will be seen, by looking down the tube, that 

 besides the two central direct images, only one diffraction image of 

 the wide set of lines on each side of the centre is admitted, but no 

 lateral image of the narrowest of lines is admitted, (pi. II., fig. 9.) 



* The width of these shts is so regulated for each case as to exclude (for 

 each system of Hues) the diffraction spectra, admitting only the central direct 

 image when the slit is situate parallel with the lines. 



