338 Discussion 



time two different colicins, and in which the cohcinogenic factors seem 

 to be hnked. As a rule, however, when a strain produces two different 

 colicins the two cohcinogenic factors are quite independent. For 

 example, if you cross an F+ which produces colicin A with an F~ which 

 produces B, 100 per cent of the recombinants produce colicin B like the 

 F~ parent; and some but not all of them will at the same time also 

 produce the one from the F+ parent. Very often, a strain which is doubly 

 cohcinogenic can transduce only one of its cohcinogenic properties. 



Pollock: Has there been any chemical separation of colicins of the same 

 strains from one another? 



Fredericq: No, the evidence for different colicins produced by the same 

 strain is most indirect, based on their action on resistant mutants and 

 on their susceptibility to proteolytic enzymes and so on, but there has 

 been no chemical separation. 



Pollock: Could there be one molecule with two properties? 



Fredericq: It seems to be quite excluded because they have different 

 rates of diffusion. The technique which gives good results in separating 

 colicins according to their proteolytic susceptibility is to inoculate in a 

 plate a streak of the cohcinogenic strain and a perpendicular streak of a 

 strain which produces a proteolytic enzyme, e.g. a strain of Proteus. If 

 they are allowed to grow for 48 hours and then the whole plate is inocu- 

 lated with the indicator strain, it will be found that the inhibition zone 

 which should be regular around the cohcinogenic streak is destroyed in 

 the vicinity of the proteolytic strain. In the case of a strain which 

 produces only one colicin this curve is quite regular, but when it pro- 

 duces two distinct colicins — which, as a rule, have different areas of 

 diffusibility — the destruction curve in the external part of the inhibition 

 zone is quite different in shape from the destruction curve in the internal 

 part. 



Cavalli-Sforza: Does Esch. coli K 12 itself produce any colicin? 



Fredericq: Not the original K 12. Of course the property of producing 

 many different colicins can be transferred to it, but originally it did not 

 produce colicin. 



