250 



Supposing the coarser series of lines were focussed for distinct vision 

 and illuminated through a very small circular aperture, upon examining the 

 emergent pencils a white spot of light would be seen in the centre of 

 the field, and on each side of it, in a line perpendicular to the direction 

 of the ruled lines, four spectra with the blue turned towards the centre 

 and the red outside. With the finer lines, which were twice as close to- 

 gether, the dispersion is doubled, and only two spectra are to be seen on 

 each side of the dioptric beam. With an object-glass of greater aperture 

 more spectra are visible, the number being infinite and extending to the 

 margin of the field of the objective used. 



Then came the question — Were these spectra of any use in forming the 

 microscopic image ? Professor Abbe proved that they were by a few 

 simple and convincing experiments. 



A series of diaphragms were adapted to fit a carrier placed in an 

 adapter, so that they could be brought over the back of the objective 

 and easily rotated or changed as might be desired. 



The first experiment was to cut off the whole of the diffraction spectra 

 by means of a narrow slit placed as above described. On examining the 

 object the lines had disappeared, only the outline of the rectangular 

 space they had occupied being visible, and it was stated that no amount 

 of mere magnifying power was capable under these 'conditions of re- 

 solving the lines. It might be objected that nearly all the light had 

 been cut off by the narrow slit. That was granted, but upon rotating 

 the diaphragm until the slit was in a position at right angles to the 

 lines and the diffraction spectra were admitted, the lines were then seen 

 as well as before, while the amount of light was the same as in the 

 former part of the experiment, proving that the diffraction spectra had 

 a very important influence in forming the image, and when excluded, 

 detail exceeding a certain degree of fineness could not be rendered visible. 



The next step was to try whether individual spectra possessed any 

 special powers in resolving or suppressing structural details. 



A diaphragm having a slit somewhat wider than that in the former ex- 

 periment, capable of admitting the two inner spectra only along with the 

 dioptric beam, was now inserted ; the result was that the coarser lines were 

 distinctly visible, but the finer series remained unresolved, which was ex- 

 plained by showing that no spectra belonging to the fine lines had been 

 admitted, therefore no image was formed. Another diaphragm was next 

 used, excluding the spectra last employed, but admitting the next pair, 

 which were the first having their origin in the finer lines, although common 

 to both series. In fact the spectrum of the coarse lines was made to re- 

 semble that of the fine lines. The result was more remarkable than in the 

 preceding experiments, both series appeared to consist of fine lines ; in the 

 case of the coarser lines a false effect being produced, as they appeared to 

 be doubled in number, a spurious line being inserted between each. 



With another diaphragm taking in only the dioptric beam and the two 

 extreme outer spectra, it was possible to produce an image resembling that 



