146 Tlie Electron Microscope 



points of the object are now illuminated with waves incoherent 

 between themselves ; and vet this makes verv little difference to 

 the microscopic image. This is the so-called critical illumination. 

 The violent controversy which followed between supporters of 

 the equivalence theory and Abbe's theory ^ was decided, in 1926^ 

 by M. Berek who showed that not only in the case of critical 

 illumination, but also in most other methods of ilkimination used 

 in practice, it is justified to consider the object, or rather its 

 interstices, as self-luminous, and that Abbe's original theory 

 gives accurate results only if the illuminating beam is narrow 

 in comparison with the objective aperture. L. C, Martin ^^ 

 showed, in 1931, that the two theories are not irreconcilable, 

 and that Abbe's method can be applied also to wide angles of 

 illumination, by analyzing the illuminating light into plane, co- 

 herent waves. But as this generalized Abbe theory has not yet 

 been fully worked out, Abbe's theory will be used here in the 

 restricted, original meaning of the word. 



In electron microscopy the di^nding line between the equiv- 

 alence theory and the Abbe theory is somewhere near a half- 

 illuminating angle of 5.10"^. If the cathode is imaged on the 

 object the angle will be about ten times wider, as has been shown 

 in chapter 5, and the instrument operates entirely in the region 

 of the equivalence theory. But the modern tendency in electron 

 microscopy is to restrict the half-illuminating angle to 5.10"^ or 

 even less. Though this should not affect the resolving power, 

 the best resolutions have been found in fact with very restricted 

 illuminating angles, presumably because the contrast was much 

 better. The reason for this is not quite clear, but it appears very 

 likely that incipient Fresnel fringes are a great help in empha- 

 sizing, and even overemphasizing small contrasts. Another 

 possible explanation is that phase contrast plays a certain part, 

 and this phenomenon is strongest in almost parallel illumination. 



Whereas these simple outlines of the theory are sufficient for 

 the understanding of most fundamental phenomena in present- 



* For a concise history of this controversy, cf. Sir Herbert Jackson and 

 H. Moore.80 



