158 MICROSCOPIC VISION. 



an over-lap in the spectra, and the spectrum becomes im- 

 pure ; so, too, by opening out the illuminating beam there is 

 an overlap of diffraction spectra, and they become so impure 

 that finally all colour is lost. Narrow the illuminating 

 beam to a mere point and the spectra from any diatomic 

 structure will become brightly coloured. (So much is this 

 the case that if a spectroscope is not at hand a coarse diatom 

 such as a pinnularia illuminated by a narrow cone when a 

 suitable objective is employed makes a very good substitute, 

 and a monochromatic screen or a light filter of any kind 

 can be tested by examining the spectra at the back of the 

 objective, when the eyepiece is removed.) This is no means 

 an old fallacy, for one of Prof. Abbe's last papers, if not the 

 last, is for the purpose of advocating the employment of a 

 narrow cone. He says,^ " The resulting image, produced by 

 means of a broad illuminating beam, is always a mixture of 

 a multitude of partial images which are more or less differ- 

 ent (and dissimilar to the object itself). There is not the 

 least rational ground — nor any experimental proof — for the 

 expectation that this mixture should come nearer to a 

 strictly cx)rrect projection of the object (be less dissimilar to 

 the latter) than that image which is projected by means of 

 a narrow axial illuminating pencil. This latter image has 

 the most favourable conditions in regard to similarity to the 

 object, because in its production nothing is lost of the 

 diffraction-pencil but the peripheral portions (which in most 

 cases are of relatively small intensity)." 



This authoritative statement, backed as it is by so great a 

 name as that of Prof. Abbe, is most damaging to the inte- 

 rests of microscopy. Por, apart from the question of the 

 manufacture of false images, it renders unnecessary any 

 improvement in the objective. The quality of an objective 

 1 Journal B.M.S., 1889, p. 723. 



