442 H. FRAENKEL-CONRAT 



usually advisable to demonstrate the detectability, by assay, of small amounts 

 of added virus. Tims, in the case of TMV, protein at 1 mg./ml. is generally 

 fomid to give no lesions, and added intact TMV (0.01 ^g./ml.) gives about 

 30 % as many lesions in the presence, as in the absence, of this amount of 

 protein. Thus, it can be concluded that the protein is less than 10"^ times as 

 infectious as TMV, or, in other words, remarkably free from virus contamina- 

 tion (Fraenkel-Conrat and Singer, 1957). 



Tests of the nucleic acid fraction, as previously mentioned, indicate that a 

 10- to 100-fold concentration by weight (0.1 to 1 jug./ml.) is required to 

 obtain the same lesion numbers as given by the standard virus. It thus has 

 to be demonstrated that contamination with 1-10 % of mtact virus is not 

 the cause for this infectivity. This has been done by a variety of techniques, 

 which can be segregated into two groups. They attempt either to exclude 

 the presence of contaminatmg virus, or to demonstrate in a positive manner 

 the nucleic acid nature of the infecting agent. 



The identification of a biological effect with a chemical agent would be a 

 simple matter if the possibility of contaminants could ever be completely 

 ruled out. In the case of TMV-RNA, it can be shown analytically that after 

 removal of salts by dialysis, the dry residue consists almost exclusively of 

 ribonucleic acid, sodium, calcium, magnesium, and traces of aluminum, 

 copper, manganese, and nickel; the metal content (total about 1 % after 

 dialysis) can be further reduced by treatment with EDTA. Loring and 

 Waritz (1957) found calcium, copper, iron, and magnesium in all preparations, 

 and found traces of the latter two to be retained by the nucleic acid after 

 treatment ^vith chelating agents. However, the significance of this is not clear, 

 and we have not found any iron in some preparations. The P/N ratio and the 

 absorption per mole of phosphorus (Ap = 9200-10,000) are all within the 

 range of typical ribonucleic acid preparations. 



Notwithstanding the fact that the viral nucleic acid has long been identi- 

 fied as ENA, the search for traces of DNA has continued. The presence of 

 any such material might be of functional importance, smce DNA has often 

 been regarded as the only genetic substance. There are recurring statements 

 in the literature, the latest by Holden and Pirie (1955), that DNA (up to 

 2.5 %) occurs in the viral nucleic acid, and the microbiological detection by 

 Hoff-jyirgensen (1952) of a little thymidine in TMV appeared to support 

 this suspicion. However, in searching for traces of DNA, the possibility of its 

 being of bacterial origin must be considered, m addition to that of chemical 

 contamination, which renders the interpretation of all trace components so 

 difficult. In analyses of purified preparations, as contrasted with the cruder 

 virus preparations previously used (Holden and Pirie, 1955), Pirie (1956, 

 1957) fomid no DNA; analyses of the infectious RNA performed by us re- 

 cently (mipublished) by the diphenylamine reaction have also shown no 



