494 BACTERIOPHAGES 



members of the T group and some were susceptible to all 7. The 

 resistance patterns of the shigella strains and their mutants were 

 the same as those of strain B of E. coli as far as tested. However, 

 this group of host cells certainly merits more intensive investiga- 

 tion than it has received. 



Indicator Straijis and Their Properties 



An indicator strain is a strain of bacteria which is resistant to 1 

 type of phage but susceptible to another type, thus enabling the 

 investigator to assay 1 type of virus in mixtures with the other 

 type. The ideal indicator strain must be completely resistant to 

 the 1 virus strain, not merely unable to support plaque forma- 

 tion. For instance, the B/3,4,7,2,6 strain described by Luria 

 (1946) would not be a suitable indicator strain for Tl and T5 in 

 the presence of T2 or T6 because, although the latter viruses do 

 not form plaques, they do kill the bacteria and hence would in- 

 terfere with the recording of Tl and T5. This point can be 

 checked by assaying low concentrations of the virus to which the 

 indicator is sensitive in the presence of very large amounts of the 

 virus to which it is resistant, and comparing the efficiency of 

 plating relative to the assay on the usual host strain. 



The efficiency of plating of the phage on any prospective indi- 

 cator strain must always be determined, since it is frequently ob- 

 served that the efficiency of plating is reduced in mutants relative 

 to that in the parent strain. For instance, the efficiency of plat- 

 ing of phage T2 on certain B/3,4,7 strains is only 20 per cent of 

 that on strain B. Relative efficiency of plating can be readily de- 

 termined by assaying suitable dilutions of the virus in parallel on 

 the indicator strain and the usual host strain. If efficiency of 

 plating is constant and not too low, the indicator strain can still 

 be used with a suitable correction factor. The relative effi- 

 ciency of plating of infected bacteria is usually higher than that of 

 free phage, so both should be determined. Also, as already 

 mentioned, the efficiency of plating of the phage to be recorded 

 should be determined in the presence of a large excess of the 

 nonrecorded phage to be certain that there is no interference. 



