72 R, MARKHAM 



acidic dyes were taken up unless the virus was denatured. Ginoza and 

 Atkinson (1956) have made similar measurements using the basic dye 

 safranin, of which about 3300 molecules are fixed per virus particle. They also 

 noted that after treatment with carboxypeptidase, about twice as many 

 safranin molecules were bound, the extra number by a site which has a 

 distinctly different afl&nity for the dye than has the original site. 



A number of measurements made by Miller and Stanley (1941) are also 

 relevant to this matter. The number of free amino groups in the virus was 

 estimated by deamination with nitrous acid, and by acetylation, and they 

 came to one for every 9000 to 10,000 molecular weight. As an incidental 

 observation arising from the controls to these experiments, these authors 

 noted the presence of acetyl groups in the intact virus. The number which 

 they estimated was about four per 19,000 molecular weight. So far only one 

 of these four has been identified, namely, the iV-acetyl serine termination 

 of the peptide chains, which will be mentioned later. 



The amino groups were also estimated by Fraenkel-Conrat (1953) by the 

 use of C^^-labeled iV-carboxyleucine anhydride. This reacts with amino 

 groups, including the e-amino groups of lysine. About a thousand groups 

 were introduced into the virus without changing its infectivity or electro- 

 phoretic mobihty. 



In 1944, Best and Lugg determined various constituents of highly purified 

 tobacco mosaic virus; from their results one can deduce that there is one 

 cysteine residue for 16,000 molecular weight, and two tryptophan residues 

 per 20,000 molecular weight. These figures have stood the test of time, and 

 are of considerable use in deducing the size of the minimum chemical subunit. 

 The sulfur in the type strain is in the form of cysteine entirely (Ross, 1940; 

 Best and Lugg, 1944; Knight, 1947), and reacts with iodine quantitatively 

 to give a stable iodine-containing compound having the grouping — S — I 

 (Fraenkel-Conrat, 1955). Much has been made of the lack of reactivity of 

 the — SH group, but this would appear to be due entirely to its isolation 

 from its nearest neighbor by a distance of at least 20 A. 



The — SH grouping also reacts Avith compounds of the type CHaHgCl to 

 give substituents of the type — S — HgCHg. Such groupings having a heavy 

 mercury atom are of considerable use in interpreting the X-ray diffraction 

 measurements (Franklin and Holmes, 1956). One mercury atom is bound 

 per 18,000 molecular weight and at a radius of 69 A. The results obtained 

 with lead are less clear-cut. The virus will take up one lead atom from lead 

 acetate per 17,000 molecular weight and still remains soluble, but the lead 

 is found in similar amounts at two levels in subsequent X-ray analysis 

 (Caspar, 1956b). One set of atoms is at a radius of 84 A, which is at the 

 periphery of the virus, while the other set is at 25 A radius, which is 15 A 

 inside the position of the nucleic acid phosphorus atoms. The superficial lead 



