30 BULLETIN" 420, U. S. DEPARTMENT OF AGRICULTURE. 
milk was low, the reduction caused by slow cooling was not so marked, 
although the bacterial count was usually less. 
From the foregoing experiments, conducted on a laboratory and 
also on a small commercial scale, it seems evident that there is no 
bacterial increase in pasteurized milk which is cooled slowly; on 
the contrary, there is often a decrease, especially when there are 
large numbers of bacteria in the milk before pasteurizing. 
These results are of particular value, for they indicate that it is 
possible to cool hot pasteurized milk slowly by means of forced-air 
circulation. From the results of our experiments we believe this 
is possible without any bacterial increase in the milk, provided the 
cooling to 50° F. (10° C.) does not require a longer period than five 
hours. This period of time should not be exceeded. It should be 
remembered that even if a period of three hours is required to cool 
niilk from 145° F. (62.8° C.) to 50° F. (10° C), the temperature is 
within the growing range of bacteria for only a portion of that 
period. With the process of cooling by forced-air circulation on a 
commercial scale milk can be cooled to 50° F. (10° C.) within three 
hours, which is well within the 5-hour limit, and in order to have a 
wide margin of safety we recommend that hot-bottled pasteurized 
milk be cooled within three hours after it is bottled. 
We wish to emphasize the fact that the bacteriological results 
which have been discussed apply only to the slow cooling of hot 
pasteurized milk during the process described in this bulletin. Pas- 
teurized or raw milk must be held, after the cooling process, at low 
temperatures in order to check bacterial development. 
EFFECT OF THE PROCESS ON THE CREAM LINE AND FLAVOR OF MILK. 
In the consideration Of the process of bottling hot pasteurized 
milk followed by slow cooling, it is of practical importance to know 
what effect such a process will have on the cream line and flavor of 
milk. Several laboratory experiments were first made to determine 
this matter. Milk was pasteurized, and hot 500 c. c. graduated 
cylinders were filled with hot milk up to the 500 c. c. mark. Together 
with the cylinder of hot pasteurized milk one cylinder was filled 
with raw milk and one with pasteurized milk which had been cooled 
to 50° F. (10° C.) in 15 seconds hf running through a coil immersed 
in brine. The method of cooling the hot-bottled pasteurized milk 
was varied considerably, as may be seen in Table V. After holding 
the milk for 24 hours-at 45° F. (7.2° C.) the numbers of cubic centi- 
meters of cream were read off directly from the graduations on the 
cylinder. This method, of course, gave a very reliable means of 
determining the effect that heating and cooling had on the cream 
line; in fact, it was too accurate, since the marked differences in the 
cream line seen in the cylinders were not apparent in the bottled milk. 
