—— ee 
on the nature of 
PRODUCTION OF MILK OF LOW BACTERIAL CONTENT. 5 
METHOD OF SAMPLING AND MAKING THE BACTERIAL COUNT. 
After milking, the cans of milk were promptly carried from the 
barn to the milk room, a distance of a few hundred feet, situated in 
the same building as the laboratory. -Long-handled white-agate 
dippers, cleaned between milkings and sterilized just before using, 
were used to stir the milk and to take the sample. A separate dipper 
was used for each sample, and the milk was poured carefully into 
sterile pint bottles, which were immediately closed with sterile caps. 
The fact that the 
laboratory was 
in the same build- 
ing made it pos- 
sible to pour the 
plates within one 
hour after milk- 
ing. Thesame 
routine was fol- 
lowed at both 
morning and eve- 
ning milkings, 
regardless of how 
the other factors 
varied. Since 
the interpreta- 
tion of results 
depends partly 
the medium and 
on its subsequent 
incubation it is 
necessary to state Fic. 2.—The small-top pail used in the experiments. 
exactly how the counts were obtained. For the sake of uniformity 
plain-extract agar, prepared according to the revised recommenda- 
tions of the Committee on Standard Methods of Bacterial Milk 
Analysis,t was used. The plates were incubated for five days at 
30° C. (86° F.) and counted with the aid of a hand glass of three 
and one-half diameters magnification. 
THE EXPERIMENTAL WORK. 
In the experiments the plan was to begin with conditions in which 
the barn and cows were as filthy as possible. When those conditions 
1 Rochester meeting of American Public Health Association, September, 1915 (Commit- 
tee on Standard Methods). 
