16 DEPARTMENT BULLETIN 319, U. S. DEPT. OP AGRICULTURE 
isms were of the lactic-acid bacteria type but differed from them in 
forming gas in lactose media. The most interesting of the organisms 
described is a long, slender bacillus corresponding to one described 
by Kern as Dispora caucasia and to which Freudenreich gave the 
name Bacillus caucasicus. In morphology, in failure to grow on 
ordinary laboratory media, and in high acid production in milk, this 
bacillus resembles very closely the bacillus mentioned later, in con- 
nection with yogurt, as Lactobacillus bulgaricus. If Freudenreich's 
description is accurate, B. caucasicus differs from L. bulgaricus in 
forming gas from lactose and in being feebly motile. Gas was 
formed slowly at 35° C. and still more slowly at 22°. No one of 
these organisms alone produced kefir, but when the four together 
were grown in milk typical kefir was produced on the first or second 
transfer. 
According to the investigations of Nikolaiewa (66), three organ- 
isms are always present in the fermented milk. One of these, Bac- 
terium caucasicum, which forms the filament of the grain, is evi- 
dently identical with Freudenreich's Bacillus causcasicus. This in- 
vestigator considers this bacterium, with a torula yeast fermenting 
lactose, dextrose, and cane sugar, as essential to the production of 
kefir. Other bacteria and yeasts are found in the grains and the 
fermented milk, but they are looked upon as contamination. 
It is probable that kefir is produced under different circumstances 
by different organisms. Any combination of bacteria or of bacteria 
and yeasts that will produce a lactic-acid and a mild alcoholic fer- 
mentation in milk will make kefir, although to obtain the most 
desirable flavor certain organisms may be essential. 
Hammarsten (3Jp ) shows in the following table (Table 1) the 
changes brought about in cow's milk by this fermentation : 
Table 1. — Chemical analysis of kefir 
Item 
2 days 
old 
4 days 
old 
6 days 
old 
Item 
2 days 
old 
4 days 
old 
6 days 
old 
Casein 
Per cent 
2.570 
.425 
.071 
3.700 
Per cent 
2.586 
.405 
.089 
2.238 
Per cent 
2.564 
.390 
.120 
1.670 
Fat 
Per cent 
3.619 
.641 
.665 
.230 
Per cent 
3.630 
.624 
.832 
.810 
Per cent 
3.626 
Lactalbumin 
Ash_.___.____ 
.630 
.900 
Alcohol _. 
1.100 
It will be observed that the changes were confined almost entirely 
to the lactose and its by-products. The casein remained unchanged, 
and the increase in the peptones was insignificant. The lactalbumin 
decreased slightly. The casein of kefir is, according to this chemist, 
not especially soluble, but may be more easily digestible because of its 
finely divided condition. The lactose diminished appreciably, and 
there was a corresponding augmentation of alcohol and lactic acid. 
A certain part of the lactose is consumed in the formation of carbon- 
dioxide gas not included in this analysis. 
The following directions are given for making kefir when the 
grains are available: The dry grains are softened by soaking in 
warm water, which should be changed several times. When the 
grains rise to the surface and become white and gelatinous they are 
ready for use. One part of these grains is used fro three parts of milk 
