Standard plate counts (SPC) and oval tube counts (OTC) were compared and analyzed by the statistical method on 547 samples of raw milk from the Wichita milk shed. 389 of these milks were paired as routine two-dilution standard plates vs. single tube estimates of viable bacteria; 127 were set in duplicate by each method; and 31 samples were set in replicates of five for the purpose of two-factor variance analysis. Using the criterion of the IAMFES Committee on Applied Laboratory Methods, the experimental results indicated that the comparison odds were approximately even for equivalent counts between methods (48.9% of samples); two out of five random samples showed that the oval tube counts were higher (39.6% of samples) and that one out of ten of the standard plate counts was superior (11.5% of samples). Also, revised data of differences between two-dilution SPC's and single OTC's that passed the chi-square test for goodness of fit for a normal distribution gave a significant difference of means at the 1% level (89 samples). However, with another trial run, two-factor analysis did not yield a significant difference between methods (33 samples × 5 replicates), and the F ratio of the means of the pooled variance data showed no difference in precision between methods; the interaction between samples and methods was highly significant. A comparison of the results from these experiments with data obtained in five other laboratories over a span of 25 years is shown in Table form. It was concluded that there is no significant difference between the two methods where paired estimates are based on replication, but the oval tube is preferred where time and economy are important.

This content is only available as a PDF.