12 BULLETIN 1152, U. S. DEPARTMENT OF AGRICULTURE. 
Color. 
Mold-ripened soy beans and wheat are usually clear yellow when 
prepared by the Japanese method for shoyu-koji. This coloration is 
due to pigments of the spores or seedlike bodies of the mold. Occa- 
sionally the color of the mass of beans and wheat is yellowish green. 
The reason for this variation is not easily denned. Temperature, 
humidity, heaviness of seeding, and rate of growth undoubtedly 
affect the color development. A lack of green in the fruit of the 
Aspergillus flavus-oryzae group is invariable when the mold is grow- 
ing on material free from carbohydrates under culture conditions. 
The ingredients of soy sauce afford wheat starch in a quantity 
presumed to be sufficient only to stimulate mold growth. In the 
work conducted in the Bureau of Chemistry, microscopical evidence 
is against this supposition, since, in this work at least, the supply of 
wheat starch was not always exhausted. Even in the brined mash or 
moromi. innumerable starch grains as yet showing no signs of cor- 
rosion were found. The color variations in the Aspergillus flavus- 
oryzae group are more easily comprehended in the consideration of 
pure laboratory cultures than in that of the more unstable mold- 
fermented beans and wheat (&£)• Environmental as well as nutri- 
ent factors need to be investigated in order to understand the clear 
yellow color of shoyu-koji. 
Bacterial Count. 
Condemnation of batches of koji may be readily based on the 
presence or absence of an ammoniacal odor or evidence of putre- 
faction. Any odor of ammonia or of putrefaction, any sticky con- 
dition, or any appearance of rottenness in koji is due to bacterial 
and not to mold activity. The bacteria involved in the ripening of 
shoyu-koji in these experiments belong to the Bacillus mesentericus 
and B. vulgatus groups (2). 
The opportunities for bacterial contamination from the soy beans 
and wheat were considered. Soy beans as they came from the field 
were studied in plain broth. The organisms recovered were club- 
shaped rods (4.8 by 7{jl), free spores (lpi in diameter), short rods (2.5 
to 5 by 0.5[x), short thick rods (4.5 by 1^), and germinating spores. 
Soy beans soaked for 24 hours in cold running water and then cooked 
for 30 minutes at 15 pounds pressure showed no growth of micro- 
organisms on culturing. Soy beans soaked for the same period in 
unchanged water having a temperature of 20° to 22° C. were cooked 
in a similar manner. Cultures developed short, thick rods. 
Whole wheat as it came from the market, when cultured in a 
routine manner, developed spore-bearing rods with rounded ends, 
measuring 1 to 1.5 by 3 to 5jjl. The same wheat when roasted and 
exposed in a room for cooling developed only mold. When the wheat 
sample was taken directly from the roaster no mold developed. The 
mold which developed undoubtedly came from the air of the labora- 
tory which was thoroughly infected with molds from frequent hand- 
ling of mold cultures and the making of large quantities of shoyu- 
koji. All bacteria discussed here are aerobic spore formers belong- 
ing to the Bacillus mesentericus and B. vulgatus groups. 
