ed
in suitable culture media, such as skim milk, which was rendered
more or less perfectly sterile by pasteurization or sterilization.
Under such conditions the addition of a selected ferment could be
made to the fresh cream, and so control the type of fermentation
which occurred therein. An essential requisite in any organism used
for this purpose must be the ability to produce relatively large
amounts of acid rapidly at ordinary ripening temperatures, and also
to form sufficient quantities of the proper flavor-producing
substances to impart a suitable flavor to the butter fat. Such
starters are known as pure culture or commercial starters, and are
prepared in both liquid and dry form. At present they are used to a
greater or less extent in all of the leading dairy districts.

Liquid starters consist of a mass of sterile nutrient medium, milk
or beef broth, inoculated with the pure culture. The dry starters
are made by adding liquid cultures, containing the growing bacteria,
to some absorbing material, such as milk sugar, milk powder, or
starch, the whole mass being dried at low temperatures, so as not to
injure the bacteria. Under such conditions the bacteria, exist in a
dormant state, and are protected from their own by-products, to
which they would be exposed if maintained in liquid cultures. The
keeping quality, therefore, of dry cultures, is much better than
that of liquid cultures.

By the use of the pure-culture starters, the butter maker is able to
add to his cream the same kind of bacteria from day to day, and the
butter will be more uniform than when the less constant home-made
starter is employed. In cream to which the starter is added, there
are present a greater or less number of acid-forming bacteria,
depending upon the age of the cream, and upon the condition under
which it was produced. These will grow during the ripening process,
and the flavor of the product will be the result of the mixture of
the bacteria in the cream. The maker can not, therefore, be certain
that the addition of a pure culture to raw cream will effectively
control the type of fermentation. This can be secured only by first
destroying the existing bacteria in the cream, before the selected
culture is added. Heating the cream accomplishes this; and in cream
thus freed from the various kinds of bacteria, the butter maker can
insure the dominance of the desirable types, contained in the
pure-culture starter. If the cream can be obta