EFFECT OF DILUTION BOTTLE MIXING METHODS ON PLATE COUNTS OF RAW-MILK BACTERIA

Forty-six raw milk samples were analyzed for plate counts at 32 C by eight different laboratories; each using three mixing procedures for the initial dilution. These were: five inversions in a 5 sec period, 15 inversions in a 15 sec period, and the “standard” method of twenty-five, 1 ft long, vertical cycles in a 7 sec period. The standard method gave the highest bacterial counts (71.1 × 103/ml average) the 15-15 method was second highest (60.4 × 103/ml average) and the 5-5 method was lowest (57.8 × 103/ml average). The standard method gave significantly higher (P<0.01) bacterial counts than the other two. The inversion methods were not significantly different from each other. Tests of reproducibility (pooled average variances for each method) did not show any significant differences between mixing methods. There were significant differences in reproducibility between laboratories. There was evidence of interaction between mixing methods by samples and mixing methods by investigators.