APPENDIX 481 



particles should have been distributed uniformly through the 

 10 ml. volume of S.G.T., thus actually contributing in the 

 neighborhood of 1 plaque to the 22 min. sample. Owing to the 

 timing of the burst, all 70 particles were contributed to this 1 

 sample. For this reason any point along the rise portion of 

 the single step curve may lie well above the curve. Since there 

 is no compensating error which may lead to correspondingly low 

 counts, these high points must be disregarded in drawing the 

 curve. This is particularly important near the end of the rise 

 period or the early stationary period where a belated burst in 

 the sampling pipet may give an abnormally high plaque count. 

 If this is included in the average value used to calculate the 

 burst size, it may lead to significant error. In the instance 

 cited all 70 plaques were released from the bacterium and 

 dispersed within 30 sec, the time required to withdraw a 

 sample and pour a plate by the agar layer method. This 

 certainly would appear to indicate that ihe phage is released 

 suddenly by disintegration of the host cell rather than gradually 

 by a process of secretion. 



Up to this point it has been assumed in the discussion that 

 release of phage from a bacterium is the result of lysis of the 

 host cell. The fact that in the one-step growth curve the release 

 of phage is not instantaneous but occurs over a period of time, 

 the rise period, has been ascribed to the infected bacteria lysing 

 over a period of time rather than simultaneously. There is 

 direct evidence on this point obtained by other methods. 



Lysi 



1 . Direct observation under the miscroscope. An adsorp- 

 tion tube is set up in the usual way with an actively growing 

 culture of bacteria and an input of phage such that multiplicity 

 is about 3, so that nearly all the bacteria will beco le infected. 

 The depression of a hollow ground microscope slide is filled 

 with melted 1.5 per cent agar so that a slightly convex agar 

 surface projects above the level of the slide. The rim of the 

 depression is heavily coated with petrolatum. Well before 



