ed yeast. The yeast ferments
the cane sugar while the acid-forming bacteria ferment the milk
sugar. There is thus obtained a drink that is similar in composition
to the real koumiss, in which both the acid and the alcohol come
from the fermentation of the milk sugar. In koumiss and kefir the
curd is very finely divided and will remain in suspension for a long
time as with butter milk.

=Determination of the cause of taints in milk.= It is often of the
greatest importance to be able to locate the cause of abnormal
odors or tastes in milk, since methods for overcoming the trouble
can be intelligently applied only when the actual cause is known. An
abnormal condition may be caused either by the direct absorption of
odors before or after the milk is drawn from the animal, or it may
be due to bacteria. If the milk appears bad-flavored when first
drawn, and if such taint becomes less pronounced as the milk becomes
older, it is likely that the trouble is due to some characteristic
of the feed. Certain feeds, like green rye, rape, cabbage, and
certain of the root crops, like turnips, impart a strong odor to
milk, if the same are fed shortly before milking. If the tainted
condition appears only some time after the milk is drawn, it may be
due to the direct absorption of taints from the surroundings in
which the milk is kept, or it may be caused by bacteria. These
causes can often be differentiated, by noting whether the taint
tends to increase in intensity with age. If such is the case, it is
likely that the cause is of germ origin, but if the reverse is true,
it cannot be ascribed with certainty to bacteria and recourse must
be had to other methods, such as the transfer of a small quantity of
the tainted milk to a sample of perfectly fresh milk, or preferably
to some milk that has been heated to the boiling point and then
cooled. In the case of an odor due to direct physical absorption, it
will not appear in the inoculated sample, since the small amount
transferred is not sufficient to be noted. If it is due to living
organisms, the inoculation of the smallest quantity into a fresh
sample is likely to reproduce the same change as originally noted.

=Tests for the bacteriological condition of milk.= Within certain
limits milk can be indirectly examined as to its bacterial content
without any special equipment. Milk when drawn from the cow has an
apparent acidity ranging from 0.16 to 0.18 per cent. By the use of
any of the met