Xir. ELECTRON MICROSCOPY 399 



this order. Wliile the technique was developed initially for metallo- 

 graphic problems and, in fact, the early techniques could not be ap- 

 plied elsewhere, the general technicjue has been continuously varied 

 and improved so that at the present time it is possible to make replicas 

 of all solid surfaces except possibly a few extremely delicate and semi- 

 liquid ones. The study of the external structure of such compact 

 solids as metals, glasses, ceramics, plastics, paint surfaces, wood, parts 

 of plants, flowers, seeds, leaves, many parts of insects, colonies of bac- 

 teria, teeth, hair, skin, blood cells, etc. is now possible. The replica 

 technique also makes it possible to examine the internal structure 

 of manj^ of the same objects. This is done by making replicas of 

 internal surfaces exposed by various methods. Generally speaking, 

 however, this variation is not applicable to delicate biological struc- 

 tures; it can be applied to the structure of bones, teeth, etc. 



Stereoscopic and shadow-casting methods are very necessary ad- 

 juncts to replica work since here the three dimensional structure can 

 seldom be inferred from a single micrograph without a very accurate 

 knowledge of the nature of the replica and without tedious point- 

 by-point measurements of the density in the micrograph. Because 

 all the information provided by a replica is contained in a surface 

 and moreover in only one of the two surfaces that determine its 

 point-to-point thickness and hence its appearance by simple trans- 

 mission imaging, the shadow-casting technique for demonstrating 

 the relief is ideal. It should be mentioned that direct visual inter- 

 pretations of replicas made without these aids are almost invariably 

 erroneous. This is a fundamental limitation due to the normal 

 tendency of the mind to interpret light and shade in terms of illumi- 

 nated objects. 



To study internal structure of the more delicate biological mate- 

 rials, it is necessary to use sectioning techniques. While this has been 

 initiated, it remains one of the potential fields of application of the 

 electron microscope in which success has been limited. The basic 

 difficulty at present lies in the fact that nearly all the histological 

 techniques developed for the light microscope have obvious defects 

 apparent as soon as the resolving power of the light microscope is 

 only slightly surpassed. Added to this is the technical difficulty of 

 cutting sections only a fraction of a micron in thickness without intro- 

 ducing serious distortion in the tissue. 



Very recently, D. C. Pease and R. F. Baker, by the combined use 

 of very small blocks, double imbedding (hardened collodion followed 



