ATP measurement can be used as an indicator of biological mass, and the extreme sensitivity of the firefly ATP assay has led to its use in bacterial detection systems. Clinical specimens present problems not encountered with cultured isolates of known bacterial species. The lower limit of sensitivity for detecting bacteria using the firefly assay is 100,000 bacteria per ml. Non-bacterial ATP, which is probably present in all clinical specimens, produces false-positive results unless it is completely destroyed, and this destruction must be carried out under conditions that do not affect bacterial ATP. A cause of false-negative results is the presence in all urine specimens of unidentified materials that inhibit the luminescent enzymic reaction. These considerations indicate that application of the firefly ATP assay in bacterial detection systems for clinical specimens is feasible only if a preparatory step separates bacteria from interfering materials and from non-bacterial sources of ATP, and concentrates microorganisms to measurable levels. These limitations sharply curtain the applicability in diagnostic microbiology of this exotic chemical reaction.