320 H. K. SCHACHMAN AND R. C. WILLIAMS 



RNA (see Section II, 5, b, in), this latter value would correspond to a macro- 

 molecule containing 61 % protein and 39 % RNA. Recent analytical studies 

 have shown that the pure virus contains 35 % RNA (Markham and J. D. 

 Smith, 1954). Combination of the partial specific volumes with the diffusion 

 and sedimentation coefiicients gives 3.0 X 10^ and 5.0 X 10® for the mole- 

 cular weights of the two components. This difference in molecular weight is 

 due to the presence of RNA in one component and not in the other. The data 

 indicate, therefore, that one component is pure protein while the second 

 contains, in addition to that protein, a large amount of RNA. Smce the 

 particle size is the same, the assumption of an empty protein shell for one 

 and a shell filled with RNA for the other seemed plausible. Evidence for 

 this was provided by the electrophoretic studies. Apparently the charged 

 groups of the nucleic acid were sufficiently buried within the structure that 

 the electrophoretic mobility was not influenced by the presence of RNA. 

 Finally, the immunological reactions seemed to preclude any contribution of 

 RNA in the antigen-antibody reaction commensurate with the RNA content. 

 Low-angle X-ray scattering measurements (Schmidt et al., 1954) have 

 been particularly valuable in corroborating the conclusions from the hydro- 

 dynamic data of Markham. The diameter of the nucleoprotein particles as 

 determined by X-ray scattering was in excellent agreement with that calcu- 

 lated from the diffusion data. Like the situation described for bushy stmit 

 virus, this means that the virus particles must have a spongelike structure. 

 Drying of the particles does not cause them to shrink to the volume calcu- 

 lated for a imiform, compact solid with a molecular weight of 5 X 10® and 

 a density equal to (1/0.666) gm./cc. From the scattering data for the "top" 

 component, Schmidt et al. (1954) concluded that the protein particles were 

 not uniform spheres. A satisfactory fit of the data was achieved for a spherical 

 model having a large central cavity of lower electron density. With this 

 "hoUow sphere model" and a ratio of inner to outer diameter (for the hollow 

 core and periphery of the particle, respectively) of 0.75, Schmidt et al. (1954) 

 obtained a diameter for the protein particle which was in agreement with 

 that calculated from the diffusion data. Further support for Markham's 

 view of the structure of the two particles comes from the recent electron 

 microscopic studies of Cosentino et al. (1956). They noted that the nucleo- 

 protein particles were rigid and nearly spherical in shape, having a diameter 

 of 260 m/x, even in air-dried specimens. The particles representing the "top" 

 or protein component, however, were flattened when viewed singly, having 

 diameters as large as 360A. In clumps, these particles did not flatten, 

 retaining their diameter of 260A, but they became perceptibly dimpled in 

 appearance, thus showing their hollow nature. Other structural details 

 whose significance is not yet fully evaluated are provided by the electron 

 microscopic investigations of Kaesberg (1956) and Steere (1957). 



