16 Mr. John W. Gordon [Feb. 17, 



practical experience, the empirical process of trial and error, rather 

 than of scientific deduction, for the practical opticians have treated 

 Helmholtz's investigation of the laws of the microscope with most 

 singular neglect. 



Before passing away from this branch of the subject, it will be 

 interesting in this room to refer to certain experiments which at the 

 present time are in a very different position, for they are as yet 

 quite immature, and the special interest attaching to them arises 

 from the circumstance that they have been suggested, and very 

 recently, by certain theoretical conclusions deduced by Lord Rayleigh 

 from the wave theory of light. It has been commonly assumed by 

 earlier writers, and was assumed by Helmholtz in his discussion of 

 the theory of the microscope, that the instrument would develop its 

 utmost resolving power when the stage was so illuminated that each 

 part of it should shine as nearly as possible with independent hght. 

 To make the meaning of this proposition clear, suppose that you fix 

 your gaze upon a candle flame ; it all appears to be of one colour 

 and one brightness, but you know that the uniformity is rather 

 apparent than real, and results from the averaging by the eye of an 

 immense number of impulses received from every point upon the 

 luminous surface, which impulses are individually by no means all 

 alike, but range between wide hmits of variation in colour and 

 brightness. In a word, every point is, in respect of the light which 

 it emits, independent of every other point, and this independence of 

 its various parts, down to the minutest into which it can be sub- 

 divided, is characteristic of a self-luminous body. Next, in contrast 

 with this, suppose that the light of the candle is received not directly 

 from the flame itself, but indirectly by transmission through a sheet 

 of ground glass, or by reflection from the surface of a sheet of while 

 paper. In that case it is plain that adjacent points upon the lumi- 

 nous surface — the ground glass or the white paper, as the case might 

 be — would not be independently illuminated. They would all 

 shine at any moment with the same borrowed light derived equally 

 from all parts of the candle flame. Therefore, as the same com- 

 ponents at all times enter into the hght transmitted from every part 

 of the secondary source, it will follow that this secondary source will 

 shine with a really uniform — i.e. very approximately uniform — 

 illumination in all its parts. 



The question now proposed may be formulated thus : It being 

 open to the microscopist to choose such a form of illumination as 

 may best suit the object with which he is dealing, let us suppose that 

 the object is extremely minute, and that it is of first importance, 

 therefore, for him to develop the full resolving power of his instru- 

 ment : will this be best accomplished by lighting the stage of the 

 instrument so that every part of it shall shine, as in a candle flame, 

 with independent light, or, as in the case of the white paper, with 

 light of more or less uniform phase ? 



