METABOLISM. 20 
body, be washed out by water injected through the portal vein till 
all sugar is removed, and if then, after standing for a time, the wash- 
ing be renewed, the first portions of water that pass contain sugar. 
The same process may be repeated several times. 
What is known as the glycogenic function of the liver was dis- 
covered by Claude Bernard in 1853, and has been the subject of a 
bewildering amount of discussion. and controversy, both as to the 
origin of glycogen, its final fate, and its relations to the production 
of dextrose by the liver. Certain facts, however, may be regarded 
as established with at least a high degree of probability: 
Firsi—The liver produces glycogen from dextrose and other 
(not all) carbohydrates, as above described. 
Second—The liver seems also to form glycogen from proteids, 
since this substance is found in considerable quantity in the livers 
of animals fed exclusively on meat. 
Third—Glycogen largely disappears from the liver during fast- 
ing, and to a considerable degree also in the absence of carbo- 
hydrates from the food. 
Fourth—The liver produces dextrose at an approximately con- 
stant rate, largely independent of the food-supply or the variations 
in the store of glycogen. 
These facts seem to point unmistakably to the sugar-producing 
function of the liver as the primary factor in the whole matter. The 
general metabolism of the body requires a constant proportion of 
dextrose in the blood, and as this dextrose is consumed the liver 
furnishes a fresh supply. This supply it manufactures from the 
materials brought to it by the blood of the portal vein. When 
carbohydrates are lacking in this blood, it apparently has the power 
of breaking down the proteids and perhaps the fats, thus supplying 
the needful dextrose. Some authorities claim that the same process 
goes on when carbohydrates are present, and it seems not unlikely 
that this is true, but when the food-supply consists so largely of 
carbohydrates as it does in the case of our domestic herbivorous 
animals, the conclusion seems unavoidable that at least a consider- 
able part of the dextrose consumed in the body must be derived 
from these substances. As already suggested, a very plausible view 
of the matter is to regard the resorbed nutrients of the portal blood 
as serving to feed the protoplasm of the hepatic cells and to look 
