Summary A-plate loop count method for determining viable counts of pasteurized milk products is described. The technique uses a 0.01 ml calibrated platinum alloy loop attached to a continuous pipetting syringe for rinsing the 0.01 ml of sample into a petri dish with 1 ml of sterile buffered dilution water prior to pouring with agar. Results with this method indicate that accuracy within ± 15 percent is attainable on samples of skim, 2 percent, homogenized, whole milk, and half and half. Measurements of chocolate milk were not accurate.

Summary Standard plate and psychrophilic bacterial counts, on 312 samples of milk collected from farm bulk tanks and milk cans, were compared with methylene blue reduction times using the standard procedure or preliminary incubation of samples at 55 F for 18 hr prior to testing. When psychrophilic bacterial counts were less than 100,000/ml, according to standard plate counts a high proportion of methylene blue reduction times were properly classified. However, psychrophilic bacterial counts greater than 100,000/ml did not significantly decrease reduction times, thus a high proportion of such milks was not properly classified by methylene blue reduction times. The proportion of misclassification of samples having standard plate counts in excess of 200,000/ml and methylene blue reduction times of 5 and 6 hr was determined.

Summary Donnelly, C. B. (Robert A. Taft Sanitary Engineering Center, Cincinnati, Ohio), James T. Peeler, and Luther A. Black. Evaluation of state central milk laboratories by statistical analyses of standard plate counts.—Standard plate counts reported by analysts in 54 state central laboratories for “normal” milk samples sent out in 1961, 1962, and 1963 revealed that an incubation temperature of 32 C produced more colonies than 35 C, and that the overall variance, s 2 , of counts for any sample should not greatly exceed 0.012. Based on this variance, s 2 = 0.012, 80% of the analysts should agree within 1.3 standard deviations of the logarithmic mean of 75% of the normal samples. A method for detecting outliers and judging the performance of individual analysts is described.

When statistically analyzed, the results of comparative, examinations of raw milk samples revealed that bacterial counts by oval tube and microplate methods were essentially equivalent to standard plate counts. The oval tube provides an economical and accurate method for determining the count of producers' milk and may be used in the laboratory or on the platform. The microplate provides an equally accurate but more rapid method than either the oval tube or standard plate since counts be obtained in 20 hours.

A plate loop count method for determining viable counts of raw milk has been described. The technique uses a 0.001 ml calibrated loop attached to a continuous volume syringe for rinsing the sample into a standard petridish prior to pouring with agar. Results with this method compare closely with the standard plate count performed simultaneously on the same sample.

Analyses of 469 series of plate counts from milksheds in almost every region of the country showed clearly that the 3-out-of-4 method of grading bacterial quality of milk by plate counts is considerably more stringent than the log-average method. In practice a 3-out-of-5 rule yields about the same proportion of violations and degradings as the log-average: These findings held for a grade limit of 100,000 as well as 200,000 colonies per ml. Wide, roughly periodic fluctuations in plate counts were often observed, but did not appear to be related to season of the year.

Methods are described for packing unfrozen split milk samples which hold their temperature within acceptable limits of 32–40°F for 24–30 hours when stored at ordinary room or higher temperature. The samples may be packed in a wrapped vacuum bottle, surrounded by a refreezant, and placed in cork or Styrofoam shippers or may be packed in nested Styrofoam boxes. Plate counts of split samples, shipped in the nested boxes to participating laboratories for examination, agreed well with each other and with the control laboratory.