A randomized prospective study was undertaken to determine the rate of punctures of the inner glove when techniques of double-gloving were employed for orthopaedic operations. Group I consisted of twenty-five procedures in which double latex gloves were used, and Group II comprised another twenty-five in which latex inner gloves and cloth outer gloves were worn. As the outer glove was expected to act as a barrier against puncture of the inner glove, only inner gloves were tested. All glove exchanges done intraoperatively for presumed tears of the inner glove were noted, as were punctures of the skin. The type and length of the procedure were also documented. Group I (double latex gloves) had a significantly higher rate of total inner-glove perforations per procedure (p less than 0.0001) than Group II (cloth and latex gloves). Furthermore, unrecognized perforations were significantly higher in Group I than in Group II (p less than 0.01). In Group I, the number of punctures increased with the duration of the operation; a puncture was found in all procedures in which the operation lasted longer than three hours (p less than 0.01). In Group II, only two tears of the inner glove occurred. Both were unrecognized and were independent of the duration of the operation.