Notes. 151 



centre lines, and the lines would, therefore, be shown in exactly the 

 same position, and otherwise, also, much the same as before. But 

 I also stated in ithe paper that with relatively broad slits the 

 second spectrum arrived at the same points in the opposite phase, 

 but couIl. not reverse the result of the direct light and of the first 

 spectrum, owing to the comparative weakness (under these con- 

 ditions) of the light of the second spectrum. But if in this case 

 the direct light were blotted out, it is at once apparent that the 

 remaining two spectra would meet in the centre-lines of the geo- 

 metrical images of the slits opposed to each other in phase. They 

 would, therefore, come to complete interference, and produce 

 practical darkness where there was maximum brightness before ; 

 and, in accordance with the general theory of interference, they 

 should produce maximum brightness where, in the presence of 

 the direct light, there was practically complete darkness. In other 

 words, the change from direct to dark-ground illumination should 

 cause the lines in a grating of relatively broad slits to change their 

 position by half an interval. 



These were the conclusions which quite recently I put before 

 Mr. Eheinberg when he happened to call on me, and I was delighted 

 to hear that he had the instrument and the gratings that should 

 enable us to submit my theory to the test of actual experiment. 



The instrument referred to is Abbe's " Demonstration Micro- 

 scope," and the grating which was selected for the experiment is a 

 " reciprocal " one, i.e. one having relatively narrow slits in one 

 half, relatively broad slits in the other half, the interval from centre 

 to centre of the slits being the same in both rulings. It is, there- 

 fore, an ideal object for this crucial experiment, inasmuch as it 

 presents the two cases side by side in the same field, and under 

 precisely the same conditions. 



It is as gratifying to myself, as it must be disconcerting to the 

 gentleman who tried to disprove my theoretical conclusions, that 

 the latter are borne out by experiment to the fullest possible 

 extent, for — 



1. The admission of the second spectrum brings out the relative 

 width of slit and dark interval, in accordance with my theory and 

 true to nature. 



2. The surprising result predicted by theory duly follows when, 

 in the presence of the first and second spectrum, the direct light is 

 cut off. 



The bright lines forming the images of the relatively narrow 

 slits retain their position, and are only slightly affected in the 

 ■direction of a moderate change in width and brightness, but the 

 lines forming the images of the relatively broad slits immediately 

 change their position, so as to appear where the dark interval was 

 before, exactly as had been deduced theoretically. 



It will, no doubt, prove a difficult matter to maintain the 



