The first-named author [this Bulletin, I9'48, v. 23, 201] has previously reported that the percentage of penicillin-resistant strains among cultures of staphylococci isolated from various lesions had risen in 1 year from 14.1 to 38. It is here recorded that in the same hospital this percentage has recently been 59. All these strains were penicillinase-formers. There is some evidence that their frequent occurrence was in part due to cross-infection: a single phage type predominated in 2 surgical wards and was also found in 2 carriers there. Resistant strains were much commoner in patients who had had previous treatment than in those who had not. When resistant strains replaced sensitive ones during treatment they were found to be of different phage types in 2 cases and of the same type in 1.

The findings are discussed from three points of view. On the question of how such strains arise, the authors favour Barber's previously expressed view that they originate by selection and not by mutation. Whether massive penicillin dosage can successfully combat such infection is answered by a qualified affirmative (each of 2 persons having septicaemia in this series died, but both had other unfavourable features). Thirdly, the claim of VOUREKA [this Bulletin, 1948, v. 23, 439] is discussed that penicillin-resistant staphylococci can be rendered sensitive by growth in the presence of other bacteria. The authors have not been able to confirm this, and in the present series of cases, resistant strains were decidedly commoner in mixed than in pure cultures, which is contrary to expectation if such a change occurred in vitro. LP Garroã.