68 R. MARKHAM 



Watson (1954) and later Franklin (1955a) have shown that the rods are 

 hehcal, having 49 subunits arranged helically m every three turns of a helix 

 of pitch 23 A. The rod itself is hollow, with a hole of 20 A radius extending 

 throughout (Caspar, 1956b; Franklin and Klug, 1956; Huxley, 1957). 



The present picture therefore necessitates the assumption of the existence 

 of a groove, some 60 A deep, between the helically arranged subunits, so 

 that the particles fit into each other when close packed. The surface of the 

 rod is probably studded with protuberances, rather like the appearance of 

 a cob of sweet corn having pointed grains (Franklin and Klug, 1956). In 

 spite of the fact that one might expect to observe this type of feature by 

 electron microscopy, so far no surface detail of a convincing type has been 

 seen. 



The approximate molecular weight * of the virus can be obtained by a 

 number of methods, none of which are very accurate; it lies between 45 X 10^ 

 and 50 x 10^. For convenience we shall assume the latter figure, which is 

 close to that obtained by a direct particle count by Williams and associates 

 (1951). 



D. Optical Properties 



The tobacco mosaic virus, being a large nucleoprotein, has a high light 

 absorption which is a summation of three major effects. First, owing to its 

 very size, the virus absorbs light nonspecifically and its solutions are opal- 

 escent. This effect is most obvious in the ultraviolet region, as it increases 

 markedly with decreasing wavelength. Owing to the presence of aromatic 

 amino acids in the protein component, the virus has a specific absorption 

 in the region around 280 m^u, and because of the nucleic acid it has a specific 

 and high absorption in the 260 mju, region (Fig. 8). 



These three separate effects combine to give a characteristic absorption 

 spectrum which will vary somewhat with the apparatus used for the measure- 

 ment (the effect of the scattered light will depend upon the acceptance angle 

 of the measuring device used). The effect of the diffuse scatter can be elimin- 

 ated more or less completely by orientating the virus, which can be done by 

 shearing a suspension, preferably liquid crystalline, between silica plates 

 (Perutz et al., 1950). The orientated material then has an absorption spectrum 

 characteristic of a nucleoprotein. This type of specimen has a marked 

 dichroism, which is an indication of the nonrandom orientation of the 

 constituent absorbing compounds in the virus. 



Attempts have been made to reconstitute the absorption spectrum of the 

 virus from the sum of the absorptions of the constituents (Reddi, 1957a). 



*The term "molecular -weight" is used here and elsewhere to denote the -weight in 

 grams of n = 6.02 X lO^^ particles. Its use is not intended to imply that the particles 

 are themselves molecular in the sense that, for example, glucose is. 



