, PRODUCTION OF MILK OF LOW BACTERIAL CONTENT, | 113) 
drawn directly from the udder. In all cases in which the bacteria 
in milk as it left the udder were counted the figures were low, be- 
cause the results represented milk drawn in the middle of milking. 
It is, of course, well known that the “ foremilk” contains more bac- 
teria per cubic centimeter than the “middle milk.” The difference 
between the bacterial count of the “ regular milk” and of milk drawn 
directly from the udder (“middle milk”) is designated as external 
‘contamination. As the foremilk is included in the regular milk this 
difference is probably greater than it should be. 
The figures show also the value of the small-top pail. The 
difference, however, is not fairly represented by the average, for 
the reason that two samples of milk from the open pail were much 
higher than the others, which raised the average. 
The most striking fact brought out by the experiment is that milk 
of a low bacterial count can be obtained, even under filthy condi- 
tions, if careful attention is given to three simple factors, namely, 
sterilized utensils, small-top pails, and clean udders and teats. 
Attention is called to the fact that the experiment was conducted 
under filthy conditions in order to emphasize the value of the three 
factors mentioned. Common decency alone should not permit the 
production of milk under such conditions, 
Fig. 7.—Condition of one of the cows during Experiment No. 3. 
