6 " Transactions of the Society. 



It may, perhaps, be thought that I have handicapped my 

 apparatus by making use of yellow light. For, it will be said, is 

 it not the fact that the resolving power of blue light is greater than 

 that of yellow light in a proportion wliich approximates 2:1? Now, 

 as a matter of speculative theory that is true, but for practical 

 purposes it is unimportant, for no lens at present made goes any- 

 where near the wave-length limit of resolving power in its per- 

 formance except upon ruled gratings. Lord Rayleigh, in a paper 

 published in the Journal of this Society in 1903,* drew the distinc- 

 tion, and showed that whereas - — , is the limit of grating in- 



2 sin u o o 



terval which can be resolved in a uniformly ruled surface, a single 

 dark line lying in" a wide bright field is theoretically visible if it 

 has a breadth equal to one-fourth of this magnitude, or under very 

 favourable conditions of illumination equal to as little as -j^. 

 For tlie case of an isolated black dot, the calculations have not yet 

 been made, but it will probably be found that a dot is theoretically 

 visible even Avhen its diameter is less than that of the narrowest 

 visible dark line. Now it so happens that dark lines and dark dots 

 in a bright field are the most important of all forms to the practical 

 microscopist, ibr they are the forms wliich Ijacteria assume. We 

 may, therefore, be well content to confine our attention for the 

 present to the work which can be done with vivid yellow light. 

 This evening I hope to satisfy you of this, among other things, that 

 the undeveloped resources of yellow light are so vast that it is not 

 only possible and pleasant, it is as wise as it is pleasant, to put by 

 the blue screen for a time and see what can be done by examining 

 objects which it is perfectly easy to see. 



The first of such objects which I have to submit to your notice 

 is a Fleurosigma formosvni, photographed under three conditions 

 of illumination (plate I.). In fig. 1 it is imaged by a narrow beam 

 of light, having a semi-aperture equal to one-third of its focal length 

 (N.A. = • 33). In fig. 2 it is imaged by a hollow cone, having a 

 semi-aperture of l'3/(iSr.A. = 1*3), from the centre of which the 

 first-mentioned cone has been stopped out ; and in fig. 3 it is 

 imaged by the full solid cone, of which the two preceding are 

 component parts, having S A. =/, that is to say, N.A. = 1. 



A comparison of these three figures suggests many interesting 

 conclusions. In the first place, the contrast between figs. 1 and 2 

 is very striking. The dot, which is densely black in the one, is 

 brilliantly white in the other, and the field which in fig. 1 is 

 featureless, is covered in fig. 2 with delicate markings. The 

 significance of this is quite unmistakable. The dot is, clearly, a 

 region in which the incident light is strongly refracted at the 

 surface of the diatom. It is thus thrown towards the edge of the 



* See this Journal, 1903, p. 474. 



