The Able Diffraction Theory. By J. W. Gordon. 367 



in the plane of the diffraction plate. "When the flame image 

 occupies the position assigned by Dr. Zimmermann to the slit in the 

 Abbe stop, and the other arrangements correspond with the specifica- 

 tion of his experiment, the spectra will be seen precisely as he describes 

 them. If, by moving either the condenser or the mirror, the wave- 

 front emerging from the upper surface of the condenser be caused to 

 assume a curved form, the spectra will be seen to diminish in breadth 

 and to crowd together. Proceeding thus by steps the observer may 

 produce with any given objective the diffraction image assigned, 

 according to the reasoning of Dr. Zimmermann's text, to any other 

 objective of higher power than its own. A narrow-angled objective 

 will in this way yield the diffraction image of an oil-immerson objec- 

 tive, and all without distinction will yield one common result. That is 

 to say, proceeding by regular degrees to shorten the radius of curvature 

 of the wave-front in the diffraction plate, the microscopist may marshal 

 his spectra in an orderly procession marching inwards towards the 

 centre until at last they all, from both sides, merge in the white 

 central image. If when this happens he replaces the ocular and 

 looks once more upon the stage, he will see the image of the flame 

 coinciding with the image of the grating, and will have ocular demon- 

 stration that, under these conditions, that is to say, with critical 

 illumination, all objectives are in this respect alike, for diffraction by the 

 object on the stage is abolished altogether. 



The process by which the diffraction pattern disappears under 

 these conditions is worthy of special attention. It is sometimes said 

 that under ordinary illumination the diffraction spectra are unseen 

 because the direct light from one point of an extended source of light 

 obscures the diffracted light from another point. In some circum- 

 stances this may be true, but it is not the explanation of the pheno- 

 mena just described. They show that with critical illumination of 

 the object on the stage, diffraction images disappear, because under 

 these conditions there is no destructive interference by parallel rays 

 of light ; not because they are obscured by direct light. The import- 

 ance of this fact upon the Abbe theory is unmistakable, and will be 

 very obvious as we proceed. At this point it may suffice to emphasise 

 and draw attention to the fact that we have here the means of 

 gathering any desired number of diffracted rays into the beam 

 received by any given objective. For the angle of diffraction is a 

 variable angle, which depends not only upon the spacing of the stria? 

 of an object, but also on the radius of curvature of the wave-fronts 

 which pass through the striae. If a particular object gives too widely 

 divergent diffraction spectra with plane waves, we have only to give 

 to the illuminating wave-fronts a curved form, and we can compel it 

 to moderate its dispersive action even to the extent of abolishing its 

 diffraction images altogether. 



Stated in this way the proposition now put forward sounds, as 

 indeed it is, very elementary. But it is by no means superfluous to 



