THE DIGESTIBILITY OF MILLETS. 3 
It was found that a bread resembling corn cake, but with molasses 
and a little ginger added to give flavor, was satisfactory for experi- 
mental purposes. The ingredients and proportions used in prepar- 
ing the millet bread were as follows: Fifteen cups of meal, 3| tea- 
spoons of salt, 3| teaspoons of soda, 5 teaspoons of ginger, 1J cups 
of molasses, 1 scant cup of shortening (lard), and 2 quarts of hot 
water. The ingredients were thoroughly mixed and baked for 
1J hours in a moderate oven. The bread prepared according to this 
method was largely crumb, having only a thin but very hard crust. 
The basal ration was so chosen as to contain a minimum amount of 
protein in order that the larger proportion of this constituent would 
be derived from the millet. As in earlier tests, it consisted of boiled 
potato, fruit (orange), and sugar. The subjects were allowed to 
drink tea or coffee without milk or cream, if they wished, and, of 
course, all the water desired. The bread was baked each day and 
accordingly was always served fresh. A quantity of potato suf- 
ficient to supply all the subjects for the entire experimental period 
was boiled and mashed. Sometimes the subjects warmed the potato 
before eating it and sometimes not. 
ANALYTICAL METHODS. 
Samples of the bread were analyzed. The composition of the 
potato, of the fruit (which furnished a very small proportion of 
the total protein), and of the sugar was computed from average 
figures. 1 The feces resulting from the test periods was freed from 
water by drying at 95° C, then weighed, pulverized, thoroughly 
mixed, and sampled. The analytical methods followed were those 
recommended by the Association of Official Agricultural Chemists. 2 
DETAILS OF THE DIGESTION EXPERIMENTS. 
The subjects were urged to eat liberally of the bread and were 
allowed to eat of the accessory foods served, as they desired. How- 
ever, the amount of potato served was small, in order that only a 
relatively small amount of potato protein would be consumed. No 
attempt was made to have all the subjects eat equal amounts; their 
preferences varied in some instances quite widely. The food eaten 
by each was weighed in separate portions, and after each meal any 
which remained uneaten was also weighed, the difference between 
these two representing the amount eaten. 
Five young men (medical and dental students), who had gained 
experience in other investigations of like character and had shown 
themselves trustworthy, served as subjects in this investigation. All 
were in good health and reasonably active and, so far as could be 
iU. S. Dept. Agr., Office Expt. Stas. Bui. 28 (1906). 
2 U. S. Dept. Agr., Bur. Chem. Bui. 107 (1907). 
