24 Transactions of the Society. 



In this connection it is of interest to examine in the light of the 

 foregoing deductions the actual appearance of such a microscopic 

 object as the photograph of Pleurosiyma anyulatum, contributed by 

 Mr. F. E. Ives to the Society's Journal, and appearing at p. 529 of 

 the volume for 1902. In the photograph of a somewhat deeply 

 curved valve the focal plane lies at varying distances above and 

 below the object itself, which is accordingly shown in parts as a 

 black dot picture and in other parts as a white dot picture, and the 

 development of the one picture out of the other can be traced in 

 strict accordance with the theory. The critical test of measure- 

 ments in depth cannot of course be applied to a mere photograph, 

 but the appearances of the dot in the different parts of the field 

 correspond so strikingly to what has been above described in con- 

 nection with the theoretical behaviour of a conical wave-front, that 

 even without precise verification these correspondences can hardly 

 fail of receiving the attention of microscopists. 



All this is but preliminary to the attack upon the problems 

 of visibility, resolution and interpretation of the black dot in the 

 microscopic image, but those problems are too large to be discussed 

 at the end of a long paper, and, moreover, as they have not been 

 in any way illustrated by my authorities they do not properly fall 

 to be discussed in this place. Only one thing remains to round 

 off the present theme, and that may be dealt with in a few words. 



Lord Eayleigh has shown, as is above stated (p. 18), that 

 resolving power can be improved by giving a certain gradation of 

 phase to the illumination of the microscopic field. We now see 

 that the diffraction fringe affords us the means of producing such 

 a gradation of phase, and we may accordingly conclude that if we 

 employ a fringe of suitable breadth for the illumination of the 

 stage we shall obtain better resolution than if it be flooded with 

 focal light. Experiment fully justifies this expectation. The 

 following may be taken as an example. 



Take a test object exhibiting features which lie at the limit of 

 the resolving power of the objective, and adjust the instrument so 

 as to secure the best obtainable image of those details. Next, 

 arrange somewhere between the lamp and the condenser a piece of 

 card or other opaque object having a keen edge, so that it can be 

 gradually introduced into the margin of the illuminating beam 

 from the lamp. Now observe the image while the card is so 

 brought slowly and cautiously into the beam. You will see its 

 diffraction fringe steal across the field of the Microscope, and as it 

 does so it will give the most astonishing crispness to the details of 

 the image. Of course such a diffractor introduced from one side of 

 the apparatus is very astigmatic, and tends to produce distorsion by 

 strengthening the shadows which lie parallel to its edge out of 

 proportion to those which lie at right angles to it. But for the 

 purpose of demonstration this is actually an advantage, and for 



