After 300 Charnley low-friction arthroplasties done in two hospitals by two surgical teams using clean-air enclosures but no preoperative, operative, or postoperative antibiotics, there were three deep wound infections, two caused by Staphylococcus albus, coagulase negative, and one by Klebsiella. Two of these infections occurred in hips previously operated on and one was in a hip with no previous surgery. These findings suggest that two operating-room environments using the same basic principles without antibiotics can achieve a 1 per cent rate of deep wound infection with no early deep wound infections after total hip replacement, a rate comparable to that reported in other series in which antibiotics were used. However, this study did not answer the questions: Is the environment the primary reason for the low infection rate or is it the discipline required by the environment? Will the rate of late (four to five years) wound infection after operations done in a clean-air enclosure be lower than that after procedures done in a "normal" operating-room environment using preoperative, operative, and postoperative antibiotics?