12 



VIRUSES 



The fact that it crystallizes in the form of dodecahedra, crystals belonging to 

 the simplest class, the cubic, indicates that its particles are spherical, or 

 nearly so. Kecent studies with the electron microscope made by Price, Williams 

 and 'Vyckoff show clearly that this deduction is correct, as illustrated by Fig- 

 ure 8. 



FIGURE 8 - SHADOWED ELECTRON MICROGRAPH OF TOMATO 

 BUSHY STUNT VIRUS PARTICLES. (W. C. Price, R. C. 

 Williams and R.W.G. Wyckoff, Arch. Biochera. 9, 175 

 U94-6J ). 



The fact that these particles are spherical makes it relatively easy to test 

 them for absolute homogeneity with respect to size and shape by means of the 

 ultracentrif uge . 



The rate at which a virus particle will sediment in an ultracentrifuge 

 depends principally upon its size. The bigger the particle, the faster it will 

 settle out. If, in a virus solution, all of the particles are of the sajne size, 

 they will settle out at the saune rate in an ultracentrifuge. This v/ill result 

 in t'le fcrration of a boundary between virus and liquid in the position of the 

 uppermost layer of virus particles, a boundary which will move at exactly the 

 rate of each and every particle. The most ideal type of boundary would be an 

 infinitely sharp one;- such a one could be represented by a plane. In actual 

 practice, however, boundaries are alvra.ys more or less foggy. There are two 

 possible reasons for this. First of all, the virus particles may not all be 

 quite the same size. In such an event, the bigger particles in the boundary 

 region will get ahead of the smaller ones and the boundary will naturally 

 spread out. But, even if all of the particles are of exactly the same size, 

 the boundary v/ill still get foggy due to diffusion. Some of the particles in 

 the solution region will diffuse up into the solvent region. Thus, the problem 

 that confronts one in trying to decide by means of the ultracentrifuge whether 

 or not a virus is absolutely homogeneous is obvious. The boundary spreading 

 during the course of the sedimentation experiment must be measured by highly 

 specialized optical methods, and then one must decide v/hether the observed 

 boundary spreading is due to diffusion alone, or due to inhomogeneity of the 

 particles. Just such studies v/ere carried out with bushy stunt virus, and it 

 was found that the observed boundary spreading could be accounted for quanti- 

 tatively by tlie known diffusion rate of the particles. The same story can be 

 told in the language of geometry, as illustrated in Figure 9« 



