MANAGEMENT OF THE MICROSCOPE. 159 



each row that lie nearest the centre, the delineation reappears, as it actually is 

 in the grating. 



Experiment 2. If, now, a diaphragm be used having three slits (E), of which 

 the central admits the direct rays only, while the two lateral receive the first pair 

 of the wider spectra and the second pair of the closer, it will be found, on replac- 

 ing the eye-piece, that the whole field is covered with the closer lines, as at F. 

 For the stopping-out of the alternate spectra of the upper series brings into com- 

 bination only that pair which corresponds with the lower, and therefore makes 

 the apparent lineation of the upper half correspond with the real lineation of the 

 lower, by the introduction of an intermediate set of spectral lines, scarcely dis- 

 tinguishable from those of which they seem to be prolongations. 



Experiment 3. Further, by carrying the two lateral slits (as at H) to the dis- 

 tance of the extreme spectra of both rows which distance represents that of the 

 spectra that would be produced by a lineation twice as close as that of the lower 

 half of A, and four times as close as that of the upper half the entire field, when 

 the eye-piece is replaced, is seen to be covered with the doubly-close lines corre- 

 sponding to that distance, as shown at I. 



Experiment 4. On the other hand, by using a single aperture shaped as in D, 

 which is broad enough to admit the innermost pair of spectra in the upper row, 

 but not to admit any of the spectra of the lower row, the field, when the eye- 

 piece is replaced, shows the wide lines (G) of the upper half of the grating, whilst 

 its lower half is perfectly Hank. 



It has thus been experimentally demonstrated that the formation of 

 the true image of the grating is dependent upon the normal re-combi- 

 nation of its diffraction-spectra, while the entire exclusion of these 

 altogether obliterates the lineation. And we thus have now for the first 

 time the scientific rationale of the fact which has long been practically 

 known the relation of the 'resolving power 'of Objectives to their 

 Angle of aperture. For it is obvious that since the 'inflected' rays 

 which form the 'diffraction-spectra' diverge more and more widely in 

 proportion to the approximation of the lines that separate them so that 

 those spectra (as already shown) are carried apart to greater and yet 

 greater distances the separation of those of a very close lineation may 

 be such as to carry them completely beyond the aperture of an Objective 

 which may take in the spectra of a more open lineation (Exper. 4). And 

 thus an Objective may be able clearly to separate lines of 50,000 to an 

 inch, from which no amount of 'coaxing' by oblique or any other kind 

 of illumination can obtain a resolution of lines of 80,000 to an inch. 

 But further, it has been made clear that most distinct ' spectral ' lines 

 can be produced in the Microscopic image, by the re-combination of 

 selected pairs of diffraction-spectra, without any real lines answering to 

 them; and hence, that the images thus formed cannot be regarded as 

 indicative of the actual structure of the obj-ects they represent; the 

 grating, for example, whose real lineation is shown at A, being made to 

 appear (according to the manner in which it is viewed) either entirely 

 blank, as half-blank (G), as having the intermediate lines of its lower half 

 extended over its upper (F), or as having its whole field covered with 

 lines at only half the distance of those of its closest part (i). The same 

 effects of obliteration or duplication of lines may be produced on such 

 objects as the scale of Lepisma saccJiarina (Fig. 417), by using higher 

 powers with suitable diaphragms. It will now be shown that the varia- 

 tions producible by similar treatment in the appearances of cross-lined 

 objects, are yet more remarkable. 



A grating with lines crossed at an angle may be prepared by cementing a 

 cover-glass, with one set of lines ruled through a silver-film on its under side, 

 upon a glass slide having another set ruled on a silver-film on its upper surface. 

 If the two sets of lines are placed at right angles to each other, a rectangular 

 grating is the result (N); if at any oblique angle, the grating is rhombic (K). 



