254 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1954 



Many novel problems arise when we use the electron microscope to 

 see molecular particles. In it, objects are visible by reason of the elec- 

 trons they deflect, or scatter ; and a molecular particle is able to deflect 

 very few compared with the substrate that must be used to support 

 it. This lack of a necessary contrast in the resulting image can be met 

 by evaporating a few atoms' thickness of heavj'^ metal obliquely over 

 a preparation. Then contrast is enhanced by the little atomic shadows 

 cast by molecular and other detail, and these have the added advan- 

 tage of giving the image a three-dimensional quality that reveals the 

 shapes of what we see. 



There is one very important difference between the practice of opti- 

 cal and of electron microscopy, made necessary by the destructive 

 effects of powerful electron beams. In general, the optical micros- 

 copist does most of his work visually and only occasionally takes a 

 picture. The electron microscopist, on the contrary, uses visual images 

 mainly to find and to focus on fields of interest. By then taking pic- 

 tures on fine-grained i^hotographic film and enlarging them later, he 

 is able to obtain very high and useful magnifications without employ- 

 ing the destructive electron intensities that would be needed to give 

 bright images at these highest enlargements. 



It is impossible to do more here than show what a few typical macro- 

 molecular particles are like, and indicate some of the problems dealing 

 with them that are now claiming our attention. In essence these 

 problems are threefold. They involve recognizing and measuring the 

 particles, determining how they are arranged in the structures they 

 form, and trying to find out how they are produced in Nature. 



The proteins that occur in solution in a living organism offer many 

 examples of approximately spherical macromolecules. When suffi- 

 ciently purified, these substances often crystallize well, and we can 

 study with the electron microscope various stages in the growth of 

 the single crystals they form. The photographs that have been ob- 

 tained from a number of proteins and plant virus proteins show the 

 beautiful molecular order that characterizes all steps in crystal forma- 

 tion (pi. 1, fig. 1). 



Another highly important type of molecular particle is a long fila- 

 ment. The thicker among these are relatively stiff and straight, like 

 the tobacco mosaic filaments in old solutions or the elementary par- 

 ticles of cellulose. Those that are thinner are wavy threads. Solu- 

 tions of the muscle proteins, of collagen from connective tissue, and of 

 the nucleic acids contain such threads; some that have been photo- 

 graphed are less than 10 atoms in diameter. Paracrystalline, rather 

 than crystalline, solids composed of these filamentous macromolecules 

 are among the important framework structures of plants and animals. 

 Their investigation is one of the most rewarding of the current appli- 



