THE BIOCHEMISTRY OF PLANT VIRUSES 71 



investigated. Simple analysis of elementary composition, and even of indi- 

 vidual amino acids, also tends to do this when the precision of the method 

 is not sufficient to distinguish between chance errors and real deviations 

 from the results anticipated on the basis of a simplified structure. It may 

 weU be that the adoption of more refined and sensitive techniques may cause 

 some modification in our ideas about the structure of the virus protein. 

 The present data, however, give a reasonably satisfying picture of the 

 over-all appearance of the virus, but possibly as seen through a slightly 

 imperfect window. 



A. Chemical Composition 



The virus consists almost entirely of protein and ribonucleic acid in the 

 proportion 94.4 : 5.6 (the value 94 : 6 usually quoted is almost certainly 

 incorrect; this will be discussed later). The general elementary composition 

 is approximately: C, 50 %; H, 7 %; N, 16.7 %; S, 0.2 %; and P, 0.54 %. 

 Variation in these values is reflected in the minor amino acid differences 

 fomid ill various strains. The amino acid and nucleic acid compositions are 

 dealt with later. 



B. Evidence Relative to the Existence of Substructure 



Although the virus behaves superficially as if it were composed of nucleo- 

 protein rods having a molecular weight of 50 X 10^, there is a lot of evidence 

 that it is composed of uniform subunits of a less complex nature. Now it 

 should be repeated here that there is considerable evidence, already mentioned, 

 that the virus suspensions are 7iot homogeneous. This does not, of course, 

 necessarily imply that the various rods are made from different subunits; 

 in fact, it is now accepted as a working hypothesis that the subunits com- 

 posing the virus protein are all similar, if not identical. That tliis may not 

 be altogether true may emerge from the subsequent discussion. 



The idea that the virus was made up from small protein subunits first 

 originated from the X-ray diffraction data of Bernal and Fankuchen (1941a). 

 This concept has considerable appeal, because the idea of a unique protein 

 having a molecular weight in the tens of millions seems improbable, to say 

 the least. Evidence for the existence of a fairly small subunit (or set of sub- 

 units) has come from several observations. 



The virus rods are acidic, moving to the anode on electrophoresis at all 

 pH's above about 3.5, presumably because they have carboxyl groups on or 

 near the surface. Oster and Grimsson (1949) measured the uptake of basic 

 dyes by the intact \iius and fomid that, for example, some 6500 molecules 

 of the basic dye acriflavin were taken up by one virus particle, while no 



