14 BULLETIN 1152, U. S. DEPARTMENT OF AGRICULTURE. 
condemnation of the product for poor physical qualities. Samples 
of shoyu-koji from one manufacturer gave a bacterial count of 
11,600,000,000 per gram for koji well covered with the yellow-green 
fruit of the mold ferment, but not matted together with vegetative 
growth of the mold. This koji was dry and mature when received. 
Examination of the process as carried on by the factory producing • 
the koji substantiated the impression that such a high count of bac- 
teria must have been preceded by poor conditions in the immature 
koji. Mature koji showing little or no fruiting of the mold ferment, 
but instead a cobwebby growth of a Mucor,' had a bacterial count of 
14,300,000,000 per gram. In both instances the organisms belonged 
to the Bacillus mesentericus or B. vulgatus group. When koji is 
overgrown with Mucor and is moist and sodden with slime from 
bacteria, it has a very filthy appearance, as well as objectionable 
odors contingent upon rapid proteolytic changes. 
It is evident that a koji made of practically sterile ingredients 
may develop ubiquitous bacteria, because of exposure necessary in 
manipulation. Several experiments indicate that the bacterial count 
in an acceptable product may be approximately two and one-half to 
three times as great as the mold count. The mold is present, how- 
ever, in much greater quantity than figures would indicate. It is 
obviously the conspicuous organism. Furthermore, the structural 
difference between a mold and a bacterium permits a far more com- 
plete count of bacteria when using bacteriological methods. The 
bacterial colony can break apart at every segment, whereas the 
vegetative structure of mold can only be torn apart. Therefore, the 
mold count is really a spore count, not a count of the whole mass of 
mold material, as in the case of the bacteria. 
A koji with a reasonably low bacterial count may be secured by 
using as a criterion for an acceptable product certain specific appear- 
ances, odors, etc. If viscosity, the odor of ammonia, putrefaction, 
poor mycelial or vegetative development of the mold, high ultimate 
temperature (over 40° C. at any time) are encountered, uncountable 
numbers of bacteria will be found. Koji of this character should be 
destroyed in such a way as to kill the microorganisms. Its retention 
around the workrooms is a menace to the clean koji. 
SHOYU-MOROMI. 
The shoyu-koji when mature is emptied into a tub of brine. In 
the experimental work conducted in cooperation with Dr. T. Taka- 
hashi one-half of a 65-gallon wooden cask was used for the tub or 
vat. The brine was a solution of commercial salt at a strength of 
20° to 22° B. The lot of mature koji ripening each day was 
emptied into the brine. With each lot was added fresh brine of the 
strength mentioned on page 17. The mash thus produced constitutes 
the shoyu moromi. 
Each tub of moromi was inoculated on the first day with four 
flasks (450 cubic centimeters of wort extract broth and 5 per cent 
sodium chlorid in each) of a yeast culture, Zygosaccharomyces sp. 
Each day a new batch of shoyu-koji was started and a mature batch 
was emptied into the mash vat, along with eight quarts of the brine. 
Mature koji, mixed with a brine solution, forms the shoyu-moromi, 
the ripening of which follows the mold fermentation. Commercial 
