366 Transactions of the Society. 



. . . that the more widely spaced ruling yields more closely packed 

 diffraction spectra. 



" If now a more powerful optical system, for example, one 

 comprising a larger aperture angle, be employed to obtain the diffrac- 

 tion image, a still larger number of diffraction spectra will at once 

 become visible, but the diffraction image will at the same time become 

 smaller, so that it cannot be observed in the manner above described 

 with a powerful objective. But if the Microscope be provided with 

 a draw- tube, the diffraction image so produced can be enlarged by 

 inserting a low-power objective in the lower end of the draw-tube, 

 replacing the ocular in its ordinary position and focussing the dif- 

 fraction image in the instrument so modified." 



I pause at this point to remark that the effect here described as 

 due to the change of objective can be produced in another way, which, 

 in considering the bearing of these phenomena upon the Abbe theory, 

 it is important to bear in mind, but which Professor Abbe and his 

 disciples appear to have entirely overlooked. It will be observed that 

 in the arrangement described in this passage — which is the lighting 

 arrangement that runs through the whole series of experiments — the 

 substage condenser is focussed upon the infinite distance, so that light- 

 waves originating at the stop are re-formed as plane waves by the 

 substage condenser, and focussed in the principal focal plane of the 

 objective. It is as plane wave-fronts, therefore, that they pass through 

 the diffraction plate. Suppose now that the substage stop were 

 removed farther from or brought nearer to the condenser, so that the 

 light waves given off from its aperture might pass the diffraction 

 plate in the form of spherical wave-fronts, concave in the one case, 

 convex in the other. What difference would that make to the dif- 

 fraction image ? The experiment is easily made, and it will be found 

 that in either case the effect upon the diffraction image is the same 

 as that of changing the objective. The shorter the radius of curva- 

 ture of the wave-fronts when they pass the grating, the narrower and 

 more closely packed will the diffraction spectra be ; and when the 

 image of the source of light is actually focussed in the plane of the 

 grating so that the radius of curvature of the wave-front there is 

 approximately = O, the diffraction spectra will be found to have dis- 

 appeared entirely, having all crowded into and been absorbed by the 

 central image. This experiment cannot easily be performed with 

 the Abbe stop.* because the Microscope is not constructed to facilitate 

 the varying of the distance of the stop from the condenser. But if 

 for the Abbe stop the image of a flame focussed by the concave 

 mirror be substituted, it may very easily be done. The condenser 

 can then be racked up and down so as to vary its distance from the 

 llaine image — which does not for this purpose differ materially from 

 the Abbe slit — and so to give any desired curvature to the wave-front 



* This experiment was exhibited at the meeting by means of a Microscope 

 specially fitted for the purpose. 



