46 BULLETIN 642, U. S. DEPARTMENT OF AGRICULTURE. 
therefore, that three different grades of milk, based on their bacterial 
counts, were considered. 
The figures presented in the tables show that with two samples of 
approximately the same initial bacterial count the increase was 
not always at exactly the same rate. In some determinations there 
were occasional counts far above normal at given periods, which 
can. be explained by the fact that there may have been a slight 
variation in bacterial types, although the initial count was approxi- 
mately the same. One type may have grown more rapidly- at a 
certain temperature than the types in other samples. It is believed, 
however, that in this work the variations above the normal were 
due to inability to control absolutely the temperatures in the hold- 
ing boxes. In general, however, the variations had little or no 
effect upon the average counts for the entire series of samples 
studied. | 
In Table 26 the average results have been grouped in two differ- 
ent ways; first, to show the growth of bacteria in a series of sam- 
ples produced under the three conditions and having different 
initial counts, when held at 4.4°, 10°, and 15.5° C. (40°, 50°, and 
60° F.), respectively; second, to show the growth of bacteria in 
each grade of milk when held at the same temperatures. In the 
lower part of the table are shown the average ratios of bacterial 
growth arranged to correspond to the counts in the upper portion. 
The ratios were obtained by dividing the count, after each succes- 
sive period of 24 hours, by the initial count. 
In the sample held at 4.4° C. (40° F.) there was a relatively 
small growth of bacteria during the period of 96 hours. A most in- 
teresting effect of temperature on the growth of bacteria is shown 
by the samples of milk produced under the third condition and held 
at 10° C. (50° F.). It will be seen that when milk with an average 
count of 4,295 was held for 72 hours, the average count was but 
little higher than that of milk with an original count of 136,553 
per cubic centimeter held 24 hours at the same temperature. <A 
similar condition was found also among samples of low-count milk 
held 48 hours at 15.5° C. (60° F.), which showed a count of ap- 
proximately 33,000,000 bacteria, while high-count milk reached ap- 
proximately 24,000,000 in 24 hours. At the end of 96 hours the 
bacterial growth reaches a point where the counts are so high as 
to be approximately the same for all grades cf milk. 
It is evident that the development of bacteria in different grades 
of milk at 10° C. (50° F.) has a direct practical bearing. For ex- 
ample, if milk were produced on a farm under conditions? which 
when fresh averaged approximately 4,000 bacteria per cubic centi- 
meter, when held at 10° C. (50° F.) for 48 hours it would contain 
an average of approximately 127,000. If milk were produced under 
1 Morning’s milk. 
