750 PROCEEDINGS OF THE SOCIETY. 



He had been experimenting with it for the latter purpose since some 

 time with promising results, but was still troubled by difficulties con- 

 nected with producing disks so that the normal phase relationship 

 between the rays passing the central and peripheral portions of the 

 disks might not be disturbed. How sensitive the image was, even to 

 slight alterations or disturbances in this respect, those who had made it 

 the subject of careful experimeut would appreciate. 



Proposals which ignore that important and vital matter altogether 

 were obviously quite untenable. 



One other point : moving the opaque stops over the Ramsden disk 

 in order to find out which parts of the object were refracting light in 

 a particular direction seemed an interesting experiment, but whilst that 

 might apply to the comparatively coarse structure, the question arose 

 whether that lent itself to deductions in the case of the relatively fine 

 structure, because these gave rise to extensive diffracted beams, which 

 were not covered by the stop, but which must be taken into account. 



Mr. Conrad Beck said he was rather surprised to hear Mr. Rheinberg 

 say that putting a stop at the posterior focus of the object-glass would 

 produce similar results, because (as illustrated by a diagram which he 

 drew upon the board) a patch stop placed in that position so as to form 

 a ring of light in the exact centre, will cut off the edge of oblique rays 

 and give a crescent of light at all positions except the exact centre of the 

 field, and it seemed hopeless to find any position except the Ramsden 

 circle where a patch stop could satisfactorily be introduced. Probably the 

 next best place would be on the front surface of the object-glass, but the 

 practical difficulties of this method would not be easy to overcome. 



One great advantage of a discussion such as that before the meeting, 

 seemed that it lifted the subject away from the interminable gratings, 

 and drew attention to the fact that the resolution of objects which were 

 gratings was quite a different thing from that of objects which were not 

 gratings, and the application of the Microscope to practical uses did not 

 lead to the examination of gratings, but to that of objects of irregular 

 contour. He thought no one acquainted with the subject could fail to 

 be struck with the fact that Mr. Gordon's photographs were taken with 

 a lens of 1 • numerical aperture. The subject was one of much interest, 

 and was worth a great deal more attention than it had yet received. It 

 was interesting to note that if they followed the course of the rays 

 through the whole Microscope, they would find there was only one disk 

 through which all the rays passed symmetrically, and that was in the 

 Ramsden circle. 



Mr. Conracly said there were many points in the paper which invited 

 discussion, but he must limit himself to a few. 



Mr. Gordon had once more repeated his idea that the well-known 

 visibility of single minute objects proved the accepted limits of resolution 

 to be wrong. The fact, of course, was that visibility and resolution were 

 quite different things ; the former was merely a question of contrast, an 

 object being seen if it contrasted sufficiently with its background, no 

 matter how small the former might be. For ages men had observed the 

 stars, which, according to present knowledge, were in probably all cases 

 below y^-o second of arc in apparent size ; the limit of resolution for the 



