
BACTERIA IN THE DAIRY. 23 
churned. ‘he examination of the butter was made, in most 
cases, without salting, inasmuch as salting very commonly 
obscures the peculiar flavors developed during the ripening 
process. The testing was made for flavor, for acid, and for 
aroma. In the records that were kept it was difficult in many 
cases to know exactly how to describe the flavor or aroma of a 
lot of butter. ‘The flavors and odors that develop as a result 
of the ripening of different species of bacteria are highly vari- 
able, and the words in our language for the description of 
either flavors or odors are entirely inadequate to any consider- 
able accuracy. Inasmuch, however, as the problem was chiefly 
to determine the relation to butter-making, the record was 
made from this standpoint, and the butter was described either 
as possessing a good, a bad, or an indifferent flavor, or as hav- 
ing a typical, or an unusual aroma, or no aroma. It must be 
recognized also that tastes differ, and a flavor which has 
appeared to me to be good might not always so appear to 
others. While, therefore, the classification is not as accurate 
as might be scientifically desired, it is sufficiently accurate for 
determining the relation of the various bacteria to butter- 
making and to the normal flavor and aroma. 
It will be evident from this description of the method of 
experiment that the tests have always been made in very small 
lots of cream. There are some decided advantages and some 
decided disadvantages in this method. The disadvantage is in 
the fact that tests in such small lots cannot be relied upon to 
produce very good results or truly normal butter. First-class 
butter, as is well known, cannot be made without attending 
very strictly to all conditions, and manifestly when butter 1S 
made in lots of half a pound or so it is impossible to control 
the results satisfactorily. On the other hand, however, itisa 
far easier matter to compare results obtained by different spe- 
cies of bacteria if they can be directly compared with each 
other. When we have six or more samples of butter made 
from the same lot of cream, and under identical conditions, 
except that the species of bacteria used in ripening is different, 
a comparison between the flavors and aromas is more valuable 
than if these tests were made upon different days from differ- 
ent lots of cream. Inasmuch as these experiments were 
designed to test the general effects of many species rather than 
