488 A. TSUGITA AND II. FRAEXKEL-COXRAT 



distribution of lower sedimentation rates, as it might be expected from 

 random frafjments of the predominant molecuhir species, which alone 

 appears to be infectious. That the infectious molecules were single- 

 stranded and of the length corresponding to tiie highest S-values was 

 most clearly demonstrated by Giercr's study of the kinetics of enzyme 

 inactivation of the viral RNA (19r)7, 19o8a,b). Light-scattering and 

 viscosity data support these conclusions. As stated, TAIV-RNA, like 

 <j!.X174 DNA and unlike most other DNA's, exists in single-stranded 

 form in the virus. It is also single-stranded if isolated at low ionic 

 strength. In the presence of salts, and even at low concentrations 

 of divalent cations, RNA undergoes marked conformational changes 

 (Haschemeyer et al., 1959; Doty et al., 1959). The drop of UV absorbance 

 (hypochromism), and the dramatic rise in optical rotation renders RNA 

 similar to DNA, and these effects, as well as alterations in chemical reac- 

 tivity, have therefore been attributed, by analogy, to helical double- 

 strandedness involving about 607^ of the bases (Haselkorn and Doty, 

 1961). However, these effects are freely reversible by dilution, heat, 

 Versene, and other agents, and it thus seems certain that the primary 

 biological role of the RNA as a carrier of genetic information is not 

 dependent upon such conformational features. The same is prol)al)ly ti'ue 

 also for DNA. 



The chemical study of TMV-RNA began with the analysis of its 

 component bases. The four typical bases, adenine, guanine, cytosine, and 

 uracil, were found in the proportions of 1.13:1.00:0.77:1.10 (Knight, 

 1952), and no minor components such as methylpurines or pseudo-uracil 

 have been detected (Littlefield and Dunn, 1958). Attempts to further 

 characterize the RNA in chemical terms have followed two paths. The 

 determination of end groups and terminal sequences may be regarded 

 as the beginning of a systematic sequence analysis. Methods for the 

 identification of terminal residues are available, although they have 

 heretofore not been applied successfully to any RNA exceeding a 

 molecular weight of 100,000 (S-RNA). Such application is dependent on 

 the availability of RNA of the highest possible purity in terms of free- 

 dom from chain fragments or shorter oligonucleotides. For those end 

 group methods which rely on a certain type of enzymatic digestion the 

 absence of other nucleases from both the RNA and the enzyme employed 

 is also of critical importance. These difficulties have largely been over- 

 come in recent studies of this laboratory. Through the use of virus 

 carrying much radioactive label (4-40 X 10'' cpm'mg of C^* or P''-) it 

 was possible to show that neither end of the chain was phosphorylated. 

 Adenine was found to occur at both ends of the chain, and the 5'-linked 

 terminal adenine was preceded by a pyrimidine, probably uracil (Sugi- 



