RECENT MODIFICATIONS OF ABBE'S VIEWS 6$ 



' My present views may lie thus expressed : With coarse objects 

 the diffracted (bent oft') rays belonging to an incident ray or pencil 

 are all confined within a very nurron- unyidar space ((round that 

 incident ray, and do not appear separated from this except witli 

 objectives of very long focus. The n-hole of such a narrow diffraction 

 pencil is consequently always admitted to the objective toget/ter with 

 the direct (incident) beam, whatever may be the direction of inci- 

 dence, axial or oblique. According to the proposition of p. 7'2 (1) 

 the image is in this case strictly similar to the object, i.e. the effect 

 is the same as if we had a direct delineation by the incident cones 

 of light alone, and as if the image did not depend at all upon the 

 diffractive action of the object. 



' If we have a preparation like a diatom a relatively coarse 

 object, including fine structural details or another preparation con- 

 taining coarse elements and fine ones in juxtaposition, the total 

 diffraction effect may be separated (theoretically and practically) 

 into two parts : (1) that which depends on. or corresponds with, the 

 coarse object (e.g. the outlines of the diatom) or to the coarse 

 elements ; and (2) that depending upon, or resulting from, the fine 

 structural detail or the minute elements. The foregoing consideration 

 applies to (1) : this constituent part of the total diffraction pencil 

 of the preparation which is admitted to the objective complete];/. 

 independently of the limiting action of the lens opening, and hence 

 the corresponding parts of the object (outlines Arc.) are depicted as 

 if there were a direct, delineation, i.e. in perfect similarity even 

 with low apertures. Those diffracted rays within the whole diffrac- 

 tion pencil which are due to the minute elements are strongly 

 deflected from the incident beams to which they belong.' 1 



According to the less or greater aperture of the objective and 

 the axial or oblique incidence of the illuminating pencil or cone, 

 tli in part of the total diffraction pencil will be subject to a more or 

 less incomplete admission to the objective, and the corresponding 

 image will therefore show the characteristic traces of the diffraction 

 image, that is to say, change of aspect with different apertures and 

 different illumination, dissimilarity to the real structure, and so 

 forth. Thus we have practically, in most cases, a composition of 

 the microscopical image, consisting of two superimposed images of 

 different behaviour. But the difference is not to be considered one 

 of jn'iiii'iph, so far as the production of the image is concerned ; for 

 it depends solely upon the different angular expression of the diffrac- 

 tion fans resulting from coarse and from extremely fine elements. 2 



Resuming, then, our illustration of diffraction phenomena as 

 applied to the theory of microscopic vision, we would point out that 

 perhaps the most serviceable illustration for o\ir purpose is a plate 

 of glass ruled with fine parallel lines. If the name of a candle be so 

 placed that its image may be seen through the centre of the plate, this 



1 Letter from Dr. Abbe. 



Thus it appears that both the ' absorption image ' and the 'diffraction image ' 

 are now held to be equally of diffraction origin ; but, whilst a lens of small aperture 

 would give the former with facility, it would be powerless to reveal the latter because 

 of its limited capacity to gather in the strongly deflected diffraction rays due to the 

 minuter elements. 



F 



