464 E. A. EVANS, JR. 



(Kozloff and Lute, 1957). Similarly, the head protein is fragmented by treat- 

 ment with alkali (pH 10, 4 hours, 37°) (Kozloff, 1957) and by prolonged 

 treatment with hydrogen peroxide (KeUenberger and Arber, 1955). Con- 

 versely, other reagents, such as the zinc-cyanide complexes, iodoacetate, or 

 hydroxylamine, affecting other proteins of the viral structure, are without 

 demonstrable effect on the head protein (Kozloff et al., 1957). 



From approximate calculations based on the phosphorus content of T2 

 and the ratio of P to protein of the virus, the particle weight of the whole 

 protein sheath of T2 is estimated to be about 150,000,000. On the basis of relat- 

 ive dimensions seen in electron micrographs, the tail protein would have a par- 

 ticle weight of about 18,000,000, so that the value for the head protein would 

 be in the neighbourhood of 130,000,000 (Kozloff et al, 1957). Van Vunakis 

 and Barlow (1956), using the dinitrofluorobenzene technique of Sanger, 

 have made an end-group analysis of the protein ghosts obtained from T2 

 and find that alanine is the only amino acid present in the iV-terminal 

 position. Assuming uniform distribution of the alanine molecules, they find 

 one per chain length of molecidar weight of 80,000. Since 80 % of the viral 

 protein is present in the head, these figures would apply primarily to that 

 structure, unless there is a highly asymmetric distribution of alanine. 



2. Treatment of T2 or of T4 with reagents or procedures that lead to the 

 splitting-off of the distal tail (see below) gives rise to virus particles in which 

 the proximal portion of the tail appears contracted. A study of this pheno- 

 menon by Kozloff (1957) has led to the suggestion that the proximal protein, 

 or at least a portion of it, has contractile properties. Comparison of the 

 behaviour of the proximal protein on treatment with alkali, ethylene diamine- 

 tetraacetic acid, and a number of monovalent inorganic ions witli that of 

 myosin preparations under similar treatment shows remarkably parallel 

 behaviour (Kozloff, 1957). While these results cannot be regarded as more 

 than suggestive, they do indicate a line of investigation for further work. 



3. The presence of a lytic agent acting on bacterial cell wall and localized 

 in the proximal portion of the tail protein has been indicated by a number of 

 observations. Weidel (1951) was the first to show that the treatment of pre- 

 parations of bacterial cell waU from susceptible host cell by bacterial viruses 

 caused dissolution of a large portion of the membrane. Barrington and 

 Kozloff (1954, 1956) and Brown and Kozloff (1957) have studied this process 

 in considerable detail. Of particular interest are their experiments in which 

 they study the degree of lysis of bacterial cell walls brought about by various 

 phage strains and preparations. With intact T2 and T4, lysis is found to 

 occur, although with the latter strain tryptophan must be present in the 

 medium as a cofactor. The amino acid is presumably required for the initial 

 attachment between the virus particle and the host cell wall. When the 

 distal protein of the two phage strains is removed by appropriate methods, 



