Turbidimetric methods for determining the potency of antibiotics are inherently more accurate and more precise than are comparable agar diffusion procedures, but assays conducted in liquid media are subject to degradation from less than ideal conditions to a much greater extent than are diffusion methods. The relationships between test organisms, antibiotics, and assay concentrations are discussed. A valid assay procedure must produce a linear response with an adequate slope (-0.4 to -1.2) by the test organism to increasing concentrations of drug; such linear response normally occurs over a limited range of concentrations. Criteria used to select photometers that offer the greatest advantages to analytical microbiologists are described, with guidelines for the most effective use of the chosen instrument.