The sensitivity of bacteria to antibiotics may be measured by a rapid method in which the criterion of sensitivity is inhibition of microcolony formation on agar which contains antibiotic. As this method allows a report on the antibiotic sensitivity of a pathogenic bacterium four hours after the test is set up, a trial of the method was carried out in a diagnostic laboratory. Two thousand five hundred and four strains of fast-growing bacteria were tested against 10 antibiotics. The overall correlation rate between the four-hour microscopic readings and the subsequent overnight readings on the same cultures was 98.1%. Seven per cent of the microscopical readings attempted gave indeterminate results, and reports on these tests were withheld until the following day.The time spent in making microscopical readings was considered fully justified by the fact that results of a high proportion of antibiotic sensitivity tests were available one day earlier than is usual with established methods.