166 STORRS AGRICULTURAL EXPERIMENT STATION. 
This method involves two errors. One results from the 
imperfections of the current methods of chemical analysis, the 
other is due to the fact that the feces contain certain amounts 
of material other than undigested residue of the food—the so- 
called metabolic products. It is safe to assume, however, that 
the errors of analysis are not large. The metabolic products 
are mainly residues of the digestive juices, mucus, and the epi- 
thelium mechanically separated from the walls of the alimentary 
canal. While the quantities of these metabolic products are 
small they are, nevertheless, sufficient to make it desirable that 
allowance be made for them in accurate experimenting. No 
method has yet been devised for their exact determination, 
however, and it is customary not to take them into account 
but to regard them as belonging to the undigested residue of 
the food. As they represent material which is used for the 
purposes of digestion, and hence is not available to the body 
for the formation of tissue and the yielding of energy, this 
method of treating them as if they were undigested food 
involves practically no error so far as the value of the food for 
the principal purposes for which it is used, namely, to furnish 
the body with nourishment. 
Hxperimenters employ various methods for distinguishing 
between the undigested residue of the food, the digestibility of 
which is being tested, and the residues from the food eaten 
before and after the experiment. The method here followed 
involves the use of milk and charcoal. For the meal immedi- 
ately preceding the experiment—generally the supper of the 
day before the experiment begins—the subject drinks a mod- 
erate amount of milk. With this he takes a quantity of very 
finely divided charcoal, which is enclosed in gelatine capsules, 
and is easily swallowed. ‘The feces from this milk have a con- 
sistency and color which makes it possible to separate them 
from those of the food which is taken for the succeeding meal. 
In the same way milk and charcoal are taken for the meal fol- 
lowing the last one of the experiment. The separations by 
this method have proved quite satisfactory in our experience. 
COEFFICIENTS OF DIGESTIBILITY. 
The proportions of ingredients digested, when expressed in 
percentages, are commonly designated as coefficients of digesti- 
bility. Thus in experiment No. 4 (see table 49 beyond), the 

