126 PROGRESS IN MICROSCOPY 



The drawbacks of interference microscopes of this type are the 

 fallowing: a very minute space between L, and Lo (cover-shp and 

 object-sUde) is not feasible. Now, the interference theory shows that 

 the thicker the interferometer, the stricter the source's monochromatic 

 requirements must be. For instance, the microscope, shown in Fig. 3.36, 

 is illuminated by a mercury-vapour lamp provided with a filter iso- 

 lating the green spectrum-line. The source 5 is to be of small size, as 

 using a large source would cause many-angled rays to pass through L^ 

 and Lo- Now, for a given tilt the interferometer has a given optical 

 thickness and the intensity of the phenomenon is, likewise, well de- 

 termined. At another angle of tilt, the intensity varies as a function 

 of the interferometer optical thickness. A whole series of diverse 

 contrast phenomena are thus superimposed. The image contrast is 

 poor and the sensitiveness of the method decreases rapidly. 



Sir Thomas Merton suggested a method permitting the use of 

 a large source. Let us consider again Fig. 3.36, where the surfaces 

 .YiAj and -Yo.Vo are assumed to be quite parallel. It is well known that, 

 in monochromatic light, a plate of selfsame thickness exhibits fringes 

 endlessly (or identically-angled fringes) which are ring-shaped. 



They can actually be observed in the focal plane S' of the micro- 

 scope objective O^. They are also perceived at 5*, as virtual images, 

 however. Let us assume that a diaphragm, pierced with annular 

 apertures is set at 5, these apertures being of such size and in such 

 position that each one isolates a homologous portion in every light- 

 fringe. Under such conditions, all the rings in the plane S illuminate 

 the specimen and give rise to identical phenomena. To every ring is 

 related an image whose contrast is definite; all rings impart the same 

 contrast to the image. Employing a diaphragm pierced with suitable 

 rings, placed at S aff'ords an image whose contrast is as good as that 

 of a pin-point source but with a greatly enhanced brightness and 

 improved quality. 



There is yet another stumbling block in connection with multi- 

 wave interference microscopes: with high N.A. objectives, the slant 

 of the rays passing through the specimen becomes considerable and 

 multiple reflections for these rays are no longer present: a steeply- 

 slanting ray passing through the transparent object is more and more 

 removed from the incident ray as reflections occur. If the object is 

 not very thin, after one or two reflections, the ray issues from the 

 transparent object; multi-wave interferences do not take place any 

 longer and sensitiveness drops. 



