
COMPARISON OF BACTERIA IN MILK. a5 
The study of the figures given in this table shows only a few 
points of especial interest, but particular attention may be called 
to the following points: 
1. When the averages of the strained and unstrained samples . 
are compared together, it will be seen that there is very little 
difference in the time of curdling. In these experiments the 
strained samples curdled actually, on an average, four hours 
earlier than the unstrained samples; a difference which is cer- 
tainly noticeable, but is hardly enough to be regarded as of any 
especial importance. 
2. It will be seen that the rapidity of curdling varies very 
widely, the shortest being 42 hours and the longest roq hours. 
The samples, it will be remembered, were all retained at the 
same temperature of 70°, and this difference in time of curdling 
is somewhat surprising. It would be anticipated that the dif- 
ference in the time of curdling under such conditions would be 
found due to the difference in numbers of bacteria that were 
present in the original milk. An examination of the table, 
however, shows that this was not the case. The sample which 
curdled in the shortest time, namely No. 4, had actually the 
smallest number of bacteria of any of the samples tested, only 
1,625 bacteria per cubic centimeter; whereas the sample in 
experiment No. 10 that curdled in 104 hours had 28,000 bac- 
teria, which is very nearly the largest number that was present 
in the whole series. It will be noticed also that in experiment 
No. 10, whereas the numbers of bacteria in the strained and in 
the unstrained samples were not very different, there was a 
- difference of about 27 hours in the time of curdling, a fact for 
which no explanation can at present be given. 
3. The average numbers of bacteria in the strained and in 
the unstrained samples are practically the same, indicating that 
‘straining has no especial influence in removing the bacteria. In 
“comparing the actual experiments in the table it will be found 
that sometimes the strained samples and sometimes the un- 
strained gave the larger numbers, and that there was a great 
irregularity in this respect. 
4. From the column giving the percentage of acid bacteria it 
would appear that a larger percentage of acid bacteria is pres- 
ent in the unstrained than in the strained samples, suggesting 
