It is possible to live on meat alone, but the quantity required is very great, involving the digestive organs, the liver, and the kidneys in an excessive amount of work. On the other hand, it is possible to reduce the consumption of proteins to a minimum by substituting for them fats and carbohydrates. But, again, after the proper balance is disturbed, the substitution ceases to be a simple problem in arithmetic. The carbon-food has to be increased out of all proportion to the protein which it replaces. If a dog which is being fed on a diet natural to it—chiefly meat—is in a condition of nitrogenous equilibrium, carbohydrate may be substituted for some of the meat. But from the very beginning it is found that, if nitrogenous equilibrium is not to be disturbed (if the dog is not to be induced to consume its own tissues), a weight of carbohydrate must be given considerably greater than the weight of the protein withdrawn. The disproportion increases as the experiment proceeds, until perhaps 12 to 15 grammes of carbohydrate have to be substituted for every gramme of protein. The proteid food has now come down to 1·5 gramme per kilogramme of the animal’s weight. Owing to the increase of carbohydrate, the caloric value of the total food, nitrogenous and non-nitrogenous, is several times as great as the animal requires. The surplus is oxidized without any equivalent in work. At about this point the experiment is brought to an end, owing to the failure of the digestive organs to deal with so large a mass of food.