10 Transactions of the Society. 



surprising that under these conditions the finer details, which high 

 power lenses are specially employed to reveal, should be lost to 

 view, and it is obvious that the remedy must be to cut off all 

 superfluous light from those zones of the system which send it 

 back in greatest abundance to the stage. When immersion 

 objectives are designed with a view to the separating of the focus 

 of reflection from that of refraction ; or when the optical 

 homogeneity of front lens, immersion fluid, cover-glass and 

 mounting medium are duly considered in setting up the object, we 

 shall be able to use cones of condenser light that will fill our 

 objectives, but until these matters come to be considered in 

 connection with the power of refracting surfaces to reflect light, the 

 appearance of any given object under illumination by large 

 condenser cones must be a mere matter of chance. 



Another set of phenomena which are largely, though by no 

 means wholly, explained by the top lighting comprises those 

 connected with oblique illumination. It has been already pointed 

 out that the reflected light from an immersion objective appears 

 to play a very important part in the lighting of the object. When 

 this top light is intended to fall sidelong on the object and to 

 illuminate it by cross lights, it must of course be oblique top light. 

 And this can be secured by shading half the objective. There 

 appear to be a large number of oblique illumination effects 

 explainable in this way. 



The foregoing are general observations. It remains to describe 

 in detail the various applications which I have so far succeeded 

 in making of mercury globules for the purpose of testing the 

 Microscope. 



The first of these experiments relates to the Fresnel rings, the 

 formation of which is illustrated by fig. 3, already described. It 

 may be pointed out that the number of such rings which can be 

 seen depends upon the aperture of the objective. In the formation 

 of interference bands, as a rule, the outer members of the series 

 fade out of view either because of the overlapping of different 

 members or because the foreshortening of the aperture as seen 

 from the outlying parts of the interference image cuts down its 

 light-transmitting power to such an extent that the illumination 

 becomes too weak to be seen. In the case of the mercury globule, 

 how r ever, a different set of conditions obtains. The reflecting zone 

 is most foreshortened, as seen through the innermost rings. Its 

 light-giving power therefore increases as the observer views it 

 through the outer rings of the series, and it seems to be a fact that 

 the limit of the number of rings seen in the Microscope is set by 

 the aperture of the objective. 



It follows from this consideration that the appearance of these 

 rings can be used as a test for the centring of the globule in the 

 optical axis of the objective. It may, I suppose, be taken for 



