The following simple method enables the factory operator to test the period of exposure in the machine: Start the machine full of water, and after the same has become heated to the proper temperature, change the inflow to full-cream milk, continuing at the same rate. Note the exact time of change and also when first evidence of milkiness begins to appear at outflow. If samples are taken from first appearance of milky condition and thereafter at different intervals for several minutes, it is possible, by determining the amount of butter-fat in the same, to calculate with exactness how long it takes for the milk to entirely replace the water. Tests made by the writer[143] on the Miller pasteurizer showed, when fed at the rate of 1,700 pounds per hour, the minimum period of exposure to be 15 seconds, and the maximum about 60-70 seconds, while about two-thirds of the milk passed the machine in 40-50 seconds. This manifest variation in the rate of flow of the milk through the machine is undoubtedly the reason why the results of this type of treatment are subject to so much variation. Naturally, even a fatal temperature to bacterial life can be reduced to a point where actual destruction of even vegetating cells does not occur. ~Bacterial efficiency of reservoir pasteurizers.~ The bacterial content of pasteurized milk and cream will depend somewhat on the number of organisms originally present in the same. Naturally, if mixed milk brought to a creamery is pasteurized, the number of organisms remaining after treatment would be greater than if the raw material was fresh and produced on a single farm. An examination of milk and cream pasteurized on a commercial scale in the Russell vat at the Wisconsin Dairy school showed that over 99.8 per cent of the bacterial life in raw milk or cream was destroyed by the heat employed, i. e., 155 deg. F. for twenty minutes duration.[144] In nearly one-half of the samples of milk, the germ content in the pasteurized sample fell below 1,000 bacteria per cc., and the average of twenty-five samples contained 6,140 bacteria per cc. In cream the germ content was higher, averaging about 25,000 bacteria per cc. This milk was taken from the general creamery supply, which was high in organisms, containing on an average 3,675,000 bacteria per cc. De Schweinitz[145] has reported the germ content of a supply furnished in Washington which was treated at 158 deg. to 160 deg. F. for fifteen minutes. This supply came from a single source. Figures reported were from 48-hour-old agar plates. Undoubtedly these would have been higher if a longer period of incubation had been maintained. The average of 82 samples, taken for the period of one year, showed 325 bacteria per cc. [Illustration: FIG. 26. Effect of pasteurizing on germ content of milk. Black square represents bacteria of raw milk; small white square, those remaining after pasteurization.]