Abstract

Three methods of preserving simulated specimens of urine were studied with six test strains of bacteria. Viable counts were measured by a surface viable count and by the filter-paper-strip method during a holding period of 72 hours. Refrigeration at approximately 4 degrees C was effective and reliable. Boric acid (1-8%) at room temperature was toxic for the strain of Escherichia coli at a density of 10(7) cfu/ml but this may not be significant at the higher concentration of bacterial cells often found in clinical specimens. NaCl-polyvinylpyrrolidone (PVP) solutions containing PVP of mol. wt 44 000 or 700 000 were not effective; they were toxic for the Gram-negative strains and did not retard the growth of Micrococcus subgroup 3. The two methods of measuring viable counts were compared for specimens held under different conditions; the specificity of the filter-paper-strip method was high but the sensitivity was low when many of the specimens contained approximately 10(5) cfu/ml.