276 Discussion 



produce a small amount of penicillinase, but not enough to allow 

 initiation of growth in the presence of penicillin, so that you may 

 never find them. If you can, by prolonged subculture under your 

 conditions, obtain a population of constitutive penicillinase producers 

 (i.e. of mutation 1) then selection for mutation 2 from such a population 

 may be possible. In the reverse way, you can lose mutation 1 in one 

 mutational step, and mutation 2 cannot then express itself. Therefore, 

 you may find quite a high reversion rate to something which although 

 genotypically not identical with your wild-type strain, is identical in 

 relation to penicillin. 



Pollock : Are you suggesting that mutation 2 is a mutation from what 

 I would call a microconstitutive strain to an inducible strain, i.e. a 

 mutation from a strain that produces small quantities of enzyme, to 

 one that produces a lot of it in the presence of, or after treatment with, 

 the inducer? 



Hayes: Yes. 



Pollock: It is an interesting point that nobody has yet discovered a 

 mutation in that direction. 



Lederberg: I think that can happen, but it may be spurious. The Lac" 

 mutant of Esch. coli K 12 might fit that definition : it produces very small 

 amounts without an inducer, and you can get reversions to the Lac"*" 

 which will produce large amounts. I do not believe that there is any 

 fundamental genetic reason why this should not happen. I think one can 

 find quite a lot of examples. 



Hayes: I suggest this model to account for the fact that although 

 selective techniques have failed to demonstrate mutations from sensitive 

 to penicillinase-producing resistant, you do appear to get the reverse 

 mutation occurring at quite a high rate. 



Lederberg: The rates at which you can get mutations for the loss and 

 for the regain of the production of different enzymes may be expected 

 to be quite variable in different circumstances. 



How many different types of penicillinase-negative cultures have been 

 looked at for their ability to give penicillinase-positive forms ? Did not 

 Rountree find, again in this one phage type, that at least the penicillin- 

 sensitives of certain phage types had a much higher rate of mutability 

 to resistance? 



Barber: I don't know whether one can talk about rate of mutability 

 when Rountree did not show the actual mutation in single strains. She 

 was postulating a higher rate of mutability on the simple evidence that 

 clinical infections were turning up with penicillinase-producing strains 

 only of this type. 



Lederberg: Has she had some laboratory experiments ? I had in mind a 

 paper on streptomycin resistance from the same laboratory (Barbour, 

 R. G. H., and Edwards, A. (1953), Aust. J. exp. Biol. med. ScL, 31, 561). 



Barber: In the paper you refer to Barbour and Edwards found that 

 staphylococci of phage-group III yielded streptomycin-resistant and 

 dependent variants more readily than did strains of other phage- 

 groups. From these findings they suggested that mutation to penicillin 

 resistance might also be higher among strains of group III, but since 



