Tiieories of Microscopical Vision. By A. E. Conrady. 551 



I indeed venture to suggest that " aplanatic cones " or " critical 

 light " would be more scientifically described and specified as 

 "concentric illumination." 



There is yet another advantage accompanying the use of ex- 

 tended cones of illumination, viz., the certainty that the objective 

 is free from serious rests of spherical aberration, for only a well- 

 corrected objective will bear a wide illuminating cone. The 

 danger of utterly false images is a very grave one, when only a 

 very narrow beam of light is employed ; we may then obtain a 

 sharp image although there is considerable spherical aberration, 

 and as the latter is equivalent to inequality of the optical paths 

 "between conjugate points, it will be seen that the phase-relation 

 between the direct light and the diffraction-spectra, which I have 

 shown to play a most important part in the formation of images, 

 will be entirely falsified by spherical aberration, and that mis- 

 leading images must result. 



It only remains to bring forward some strong evidence in 

 favour of the position which I took up in the early part of my 

 former paper, i.e. the claim that all microscopical images were due 

 to the diffraction produced by the object. 



The chief theoretical arguments in favour of this somewhat 

 revolutionary postulate were given in the former paper, and have 

 not as yet been called into question ; there is, however, experimental 

 ■evidence tending in the same direction. 



The first of these experimental facts is one of which I myself 

 often make practical use in the testing of Microscope objectives. 

 It is this : if we examine a broken specimen of Pleurosigma angu- 

 latum (showing the familiar postage-stamp fracture) with a wide 

 " aplanatic cone " of light, using a dry objective, we obtain a re- 

 markable result if spherical aberration is present, i.e. if the wrong 

 tube length is employed. 



At one focal adjustment the broken edge is clearly discernible, 

 whilst by varying the adjustment the dots may be brought into 

 view. As the fracture and the structure are really in the same 

 plane, this is utterly inexplicable on the basis of the spurious disk 

 theory ; it is irreconcilable with the assumption that the object 

 behaved as if it were self-luminous, for in that case all parts of the 

 object would have their images formed by the same process and in 

 the same plane. 



The diffraction-theory on the other hand explains this quite 



easily and naturally. 



the broken edge produces a narrow fan oi: diffracted light 

 closely surrounding any ray of direct light; the image of the 

 broken edge is due to" such confined pencils of diffracted lighl 

 passing through the axial portion of the object-glass ; for owing to 

 spherical aberration affecting (when of fairly moderate magnitude) 



