112 Transactions of the Boijal Microscopical Society. 



Still, even the above-named Nobert's band is quite within the 

 limits of what might be resolved by the use of blue light, and thus 

 there is no difficulty in understanding how it might be photographed 

 as done by Dr. Woodward. 



Similar principles would of course apply in the case of the 

 very close and uniform markings on the frustules of Diatomaceae. 

 Dr. Woodward's paper and admirable j)hotographs of Frustulia 

 Saxonica, brought before our Society at our meeting last No- 

 vember,* fully bear out all Helmholtz's conclusions, and show the 

 difficulty of distinguishing true structure from interference fringes 

 when the intervals between the real markings are of the same 

 order of magnitude as half the length of the waves of light. This 

 effect is of course altogether independent of the quality of the 

 lenses. It depends on the physical constitution of light itself, and 

 would only be the more perfectly seen with more perfect object- 

 glasses. 



There is also another fact mentioned by Dr. Woodward which 

 merits attention.t He says that for resolving very close lines or 

 linear markings it is a decided advantage to have the lenses some- 

 what under-corrected for colour. As he suggests, this may be 

 partly due to the possibility of making such lenses more correct 

 for spherical aberration, but at the same time it appears to me quite 

 possible that it may also to some extent be due to the fact that , 

 with such a correction it is possible so to have the lines in focus 

 for the blue rays as to take advantage of their shorter wave-length, 

 whilst the interference fringes due to the longer waves are suffi- 

 ciently modified by being out of focus as to obscure the vision less 

 than they otherwise would. 



Taking then all these facts into consideration, it appears to me 

 extremely probable that for object-glasses not made on the immer- 

 sion principle the limit of perfectly satisfactory definition of lines 

 not exactly the same distance apart must be somewhere about 

 gQ^QQ of an inch. With a dry lens having an apertm'e of 140°, 

 or an immersion of 100^, both illuminated by a condenser of equal 

 angle, only the extreme red rays would then serve to produce 

 a very slight indistinctness. Under very favom*able circumstances 

 by varying the angle of divergence of the light passing from the 

 condenser, or by throwing it more from one side than from the 

 opposite, it would be possible to make the dark interference fringes 

 so coincide with dark structural lines that a considerably smaller 

 interval might be distinguished. This, however, would be ex- 

 tremely difficult if not impossible, if the hues were at unequal 

 intervals, since any adjustment of the illumination that gave inter- 

 ference fringes at the proper interval and situation for one part of 



* ' Monthly Microscopical Journal,' 1875, vol. xiv., p. 274. 

 t ' Quart. Journ. of Micros. Science,' viii., p. 229. 



