330 THE BAGTERIOPHAOE 



but the results recorded were undoubtedly due to technical errors (Spat 1924, Borchardt 

 1924, Gildemeister and Herzberg 19246, Meissner 19246, Gercke 1925, Bronfenbrenner 

 and Korb 1925a). 



The size of the lytic particle in a number of phage preparations has been deter- 

 mined and it has been found that different strains of phage are each characterized 

 by a certain relatively narrow range of particle sizes. 



The size has been estimated in various ways. Filtration through membranes of 

 approximately known pore-sizes indicated diameters of the order of 20-50 mft (Stassano 

 and de Beaufort 1925, Bechhold and Villa 1926, Zinsser and Tang 1927). Elford and 

 Andre wes (1932) used carefully graded collodion membranes, and showed that different 

 pure -strain phages had each a different particle size and that size was independent of the 

 particular strain of bacterium upon which they were being propagated. One phage, for 

 instance, had a mean diameter of 8-12 m/<. The diameters of a group of phages lay 

 between 50 and 75 m//, whUe other strains gave intermediate values. Yaoi and Sato 

 (1935) recorded similar results. Schlesinger (1932-33a, 1933), Elford (1936), Mcintosh 

 and Selbie (1937) and Paic, Krassnoflf, Haber, Reinie and Voet (1938) estimated the diameter 

 of phage -particles by measuring their rate of sedimentation in a high-speed centrifuge. 

 The results were consistent with those observed by filtration and confirmed the existence 

 of a wide range of particle -size. Wollman and Lacassagne (1940) estimated a similar 

 range of sizes, from the results of inactivation of phages by X-radiation (see also Luria 

 and Exner 1941). 



Reference to Chapters 2 and 41 wiU show that the larger phages are big enough to 

 be demonstrated by modern microscopical methods. Merhng-Eisenberg (1938) and Eisen- 

 berg-Merhng (1941) obtained photomicrographic images in visible Ught of staphylococcus 

 and Bad. coli phages, which on photometric measurement yielded estimates of size similar 

 to those obtained by ultrafiltration and ultracentrifugation studies. Barnard, using the 

 technique of photography by monochromatic ultra-violet Ught, has obtained photographs 

 of a phage of this type, which show it to be composed of uniformly sized particles with 

 a diameter of about 50 m/i (see Biu-net 1933e). These large phages produce m lysed 

 broth cultures a tiu-bidity that can be detected by its Tyndall effect, and the intensity 

 of this effect has been found to provide an acciu-ate measure of the concentration of the 

 phage in the filtrate (Schlesinger 1932-336, Schuurman and Schuurman-ten Bokkel 

 Huininck 1936). 



Electron micrographs reveal not only the size of phage particles, but also a structure. 

 Ruska (1941) noted " sperm-shaped " particles m his photographs, and more recently 

 Luria, Delbriick and Anderson (1943) have shown that three of four Bad. coli phages 

 they studied had an opaque " head " consisting of a pattern of granules, about 80 m/i 

 in diameter, and a less opaque " tail," about 120 w/< long (Figs. 50, 51, 52). 



Our conception of the lytic particle, of a size that closely characterizes the 

 various types of phage, as the ultimate unit of phage, does not accord with recent 

 descriptions of certain physical and serological properties of Bact. coli phage. 

 By diffusion experiments with dilute solutions Kalmanson and Bronfenbrenner 

 (1939) deduced a particle weight of 1,500,000 for a coli phage. The particle weight 

 of the corresponding lytic unit was 27 times as great. Again, the maximum 

 amount of antibody that combined with a lytic unit of phage was found to be 

 far more than could be accommodated on the surface of a simple lytic unit. By 

 postulating sub-units of 1,500,000 particle weight, the quantitative serological 

 data fall into line with those from reactions of other antigen-antibody systems 

 (Hershey, Kalmanson and Bronfenbrenner 1943). The lytic unit appears to be 

 made up of about 30 independent diffusing units, bound by 25 per cent, of carrier 



