APPENDIX 497 



amounts in test tubes in the usual way. Two drops of the appro- 

 priate bacterial suspension are used to inoculate each tube of 

 agar. Then 0.1 ml. samples of the diluted virus mixture are 

 plated with the following results: 



on B 210 clear plaques-Tl + T7 



on B/ 1,5 113 clear plaques-T7 only 



on B/3,4,7 107 clear plaques-Tl only 



on B/1,5 + B/3,4,7 201 turbid plaques-Tl + T7 



The sample plated on B gives the total number of virus parti- 

 cles, and an experienced eye can in most cases allocate each 

 plaque to either Tl or T7. About 10 per cent of the plaques 

 cannot be positively identified. The sample plated on B/1,5 

 gives only the T7 plaques, and the sample plated on B/3,4,7 gives 

 only the Tl plaques. The sum of the plaques on these 2 plates 

 should equal the total plaque count on B within the sampling 

 error if the efficiency of plating on the indicator strains is 1 00 per 

 cent. The sample plated on mixed indicator strains should give 

 the same number of plaques as the sample plated on B, but all the 

 plaques will be turbid. A Tl phage particle will lyse strain B/ 

 3,4,7 and so produce a plaque, but strain B/1,5 will not be lysed 

 and will grow within the plaque area forming a turbid back- 

 ground which will, however, be less turbid than the area outside 

 the plaque where both indicator strains are growing. Similarly, 

 the T7 plaques will be turbid because of overgrowth by B/3,4,7. 

 However, where a Tl plaque happens to overlap a T7 plaque 

 there will occur a completely clear area because within the area 

 of overlap both indicator strains will be lysed. If when the agar 

 layer hardened a Tl phage particle happened to come to rest 

 very close to a T7 particle, the 2 plaques will be nearly concentric 

 and a completely clear plaque will result. However, the inci- 

 dence of such clear plaques should be quite small. Also it is 

 usually possible to distinguish the Tl from the T7 plaques on the 

 mixed indicator plates because despite the disproportionate in- 

 oculum the B/1,5 strain grows more rapidly and the Tl plaques 

 are more turbid than the T7 plaques. On long incubation the 



