82 F. CRISP ON THE INFLUENCE OF 



and examined apart from each other — one image being that by 

 which the contour outlines of the larger parts are formed, the 

 other image giving the delineation of minute structure. So 

 in the case of an object with lines closer than g^oo" °^ an i ncn 

 apart, the image seen by the eye is formed not simply by the cen- 

 tral dioptric beam, but by the joint action of that and the diffraction 

 images and their exact union in the upper focal plane of the objective. 



How important an influence the diffraction spectra haVe on the 

 formation of the image, may be seen by the following experiments, 

 which, for the moment, I must explain by means of diagrams, 

 though they can be verified after the meeting under the microscope. 



Fig. I represents the appearance in the field of the microscope 

 of an object formed of two sets of lines ruled through a thin film of 

 silver deposited on glass, the lower set being twice as close as the 

 upper ones. 



Fig. II shows the appearance which is presented on removing 

 the eye-piece and looking down the tube of the microscope, the 

 central circle in each series is the dioptric beam, the others being 

 the diffraction spectra. It is important to observe that the distances 

 of the spectra apart are inversely as those of the lines, the upper 

 (widely separated) lines producing spectra twice as close as those of 

 the lower. 



If we now place over the back of the object-glass a diaphragm 

 like Fig. A, the diffraction spectra will be admitted, as well as the 

 central beam, when the slit of the diaphragm is at right angles to 

 the direction of the lines ; when, however, the diaphragm is turned 

 round 90°, so that the slit is vertical, the whole of the diffraction 

 spectra are shut out, and nothing but the dioptric beam is admitted. 



If, therefore, the diffraction spectra contribute, as the theory 

 assumes they do, to the formation of the image, some change 

 ought, of course, to be perceptible when they are thus excluded. 

 It will be found that, as the diaphragm is turned round, and the 

 diffraction spectra disappear, the lines also disappear, and not a trace 

 of them can be seen (Fig. a). 



This demonstrates the importance of the spectra in the formation 

 of the true image, and we may now carry the demonstration a step 

 further, and show what influence they may have in the formation of 

 false images. 



Referring again to Fig. II, if we are able by any means to make 

 the spectra of the upper lines identical with those of the lower, it 



