MICROSCOPIC VISION. 163 



inch, is only barely resolved the spectrum of the 1st order 

 passes through the same zone on the opposite side (Fig. 5) ; 

 therefore, under these conditions, as there is no spectrum of 

 the 2nd order admitted, it is impossible to duplicate the 

 structure. Although the image of the object probably will 

 be ill defined and defective in certain particulars, it will, 

 however, be correct with regard to the fineness (48,000 per 

 inch) of the structure delineated. When, however, the 

 structure is sufficiently coarse (24,000 per inch) to enable a 

 spectrum of the 2nd order and the dioptric beam to pass 

 through the marginal zone (Fig. 6), a spectrum of the 1st 

 order will pass through the centre of the objective ; spheri- 

 cal aberration will prevent this 1st order spectrum com- 

 bining with the dioptric beam and the 2nd order spectrum 

 situated in the marginal zone, consequently an image of a 

 structure having double the fineness {i.e. 48,000 per inch) of 

 the origioal will be seen. Thus, for example, if the original 

 structure consists of a net having 24,000 meshes per inch, 

 the image will be that of another net having 48,000 meshes 

 per inch. We have confined ourselves to lines and diatom 

 markings in the preceding discussion^ because with such 

 objects demonstration becomes easier, but the laws apply to 

 all objects ; some biological objects upon which theories have 

 been based are of the nature of false ghosts, or, in other 

 words, are merely the duplications of entities. 



It is the duty of every microscopist therefore, be he a bio- 

 logist or a physicist, to critically examine the conditions he 

 is working under, and so make sure that those conditions 

 are such as will not permit the separation and isolation of 

 groups of spectra of various orders by the spherical aberra- 

 tion of the objective. 



Some tests may be applied to determine whether the 

 structure in question is, or is not, a false ghost. 



