THE PHYSICAL PROPERTIES OF INFECTIVE PARTICLES 263 



materials use is made of a much more elaborate treatment employing equa- 

 tions derived by Fujita (1956). To use this theory, both the concentration 

 dependence of the sedimentation coefficient and some parameters describing 

 the shape of the boundary are required. As a further test of homogeneity, 

 the bomidary shape should be measured accurately and compared with a 

 Gaussian curve. Deviations from a Gaussian shape are an indication of 

 inhomogeneity. 



It should be recognized that these measurements reveal heterogeneity, 

 or its absence, with respect to sedimentation coefficient only. As such, any 

 observed heterogeneity may be attributable to variations, within the popu- 

 lation of solute molecides, of either density or molecular size and shape. In 

 order to accentuate the effect of differences in density among the solute 

 molecules, tests should be made in solutions of higher density so that the 

 buoyancy term, (1 — Vp), is made closer to zero (Cheng and Schachman, 

 1955a). By this means, variations among the virus particles in a given pre- 

 paration with regard to their nucleic acid content, for example, may be 

 subjected to direct experimental inquiry. This is feasible because of the 

 high density of nucleic acid. If the relative bomidary spreading is indepen- 

 dent of the density of the medium, there is no heterogeneity with respect to 

 density. Conversely, any observed enhancement of the boundary spreading, 

 as the density of the medium is increased, indicates that the sedimenting 

 material is heterogeneous with respect to density. The increasing emphasis 

 on the study of the structure of incomplete viruses (lacking in nucleic acid) 

 makes this type of investigation important. The reagent used to increase 

 the density must be inert and exhibit no specific interaction with the virus. 

 Evidence of this same type can be provided by sedimentation equilibrium 

 experiments in density gradients (Meselson et al., 1957). 



V. Sedimentation in Multicomponent Systems. As aLready indicated, the 

 equations presented above are restricted to two-component systems. How- 

 ever, it is often necessary to investigate viruses or their degradation products 

 in solutions containing large quantities of a third component, such as sucrose 

 or urea (Bechhold and Schlesinger, 1933; Smadel et al, 1938). If it is known 

 that the sedimenting material does not interact preferentially with either of 

 the components of the solvent or, alternatively, there is no interaction (like 

 solvation) at all, the solution then can be considered as a two-component 

 system described by the equations presented earlier. Earely, if ever, does this 

 situation obtain. Even for those systems containing inert materials like 

 sucrose there is, as Kauzmann (see Schachman and Lauffer, 1949) has 

 pointed out, preferential interaction between the sedimenting material and 

 water because of steric exclusion of sucrose from the immediate vicmity of 

 the sedimenting solute molecules. The existence of this effect complicates 

 those numerous investigations aimed at measuring the hydration of viruses. 



