UNLABELLED The efficacy of a disinfectant can only be determined exactly, when quantitative as well as qualitative tests are being performed. Aim of this study was to ascertain whether the earlier published method of a germ carrier test could be applied also to the testing of disinfectants for instruments. Rubber tubes of 1 cm length were used as germ carriers, Staph. aureus, Pseud. aerug., Proteus mir., E. coli and Cand. alb. as test organisms. The culture media were CASO-agar and -broth, latter containing 20% defibrinated sheep blood in some experiments. Two disinfectants on aldehyde-basis and one disinfectant on phenol-basis and on biguanide-basis, respectively, were used for the disinfectant tests. The rubber tubes were contaminated with bacterial suspension, dried upright on sterile filters and then put into 10 ml CASO-broth. After high speed vortexing of the germ carriers the bacterial count was determined. Following factors that were believed to probably having an influence on the recovery rate of bacteria were investigated: 1. the amount of fluid staying on the rubber tubes; 2. the drying time; 3. the vortexing time; 4. repeated vortexing; 5. variations of the initial concentration of bacteria on the rubber tubes; 6. addition of inactivating substances. RESULTS ad 1. The mean amount of fluid staying on the tubes was 0.0056 g. ad 2. Drying times longer than 30 to 60 min resulted partly in considerably lower bacterial counts. ad. 3. and 4. Neither extension of vortexing time nor repeated vortexing of the same germ carrier could raise the recovery rate of bacteria. ad 5. Higher initial concentration of bacteria on the rubber tubes could not be achieved. The recovery rate of bacteria was reliable even when the initial bacterial concentration was low. ad 6. Addition of inactivating substances to the CASO-broth had no influence on the recovery rate of bacteria. The test method was checked by testing four disinfectants for instruments and proved itself a good and reliable quantitative method.