The remedy is equally clear, viz., to increase the supply of carbohydrates, a simple enough process in certain circumstances, but one requiring considerable caution in diabetes. In the mild form of this disease, as ascertained by estimating the amount of sugar in the urine and the absence of acetone bodies, a rigid diet as free from carbohydrates as possible should be adopted. If this checks the excretion of sugar, then a little ordinary bread is gradually added to the diet, and the quantity consumed just before the reappearance of the glycosuria is a measure of the carbohydrate food which can be tolerated by the patient. In the severe forms of the disease, too strict a limitation of the carbohydrates, especially when a diabetic is in bed, is a risky procedure, so that a few days must be allowed to elapse before changing from a mixed to a rigid diet. The ideal diet for such cases is nitrogenous food, and fatty food with a small quantity of carbohydrate, saccharine food being entirely avoided.

The rationale of all the dietetic systems for diabetics, such as the oatmeal cure, the preference of potatoes over bread, etc, can easily be comprehended from a consideration of the above facts. In diabetes an important object of quest is the discovery of less harmful energy-producing substances to replace carbohydrates, and lsevulose, oatmeal or potato starch, citric, lactic, glutinic, and glyconic acids can all do this to a certain extent. A great desideratum is a practical substitute for sugar and sugar-yielding foods, and glycerine, glycerine aldehyde, and alcohol have all been recommended, but it should be noted that the last mentioned is a poison. It is believed by many that the carbohydrates are tolerated in the following order of acceptability: laevulose, lactose, starch, cane sugar, grape sugar, and galactose, the last three being especially badly borne. Clinical experience, however, hardly corroborates this testimony, and it is well known that in some cases the sugar formed from albumin is more detrimental than carbohydrate sugar. For this reason diabetics should never eat more than 150 grams of protein per day. Amongst vegetarians, both in India and elsewhere, it is customary to use cocoanut and other oleaginous seeds as a substitute for animal proteins, and they claim that the treatment of diabetes in mixed feeders is much more successful when meat, alcohol, tea, coffee, and cocoa are withdrawn from the diet.

The lower fatty acids - butyric, iso-valeric, caproic and perhaps acetic acid - increase the acetonuria, and butyric acid may be changed directly into β oxybutyric acid. In very severe cases therefore butter, which contains so many of the lower fatty acids either free or as triglycerides, should be used fresh and after washing, never old.

When there is a threatening of coma, dextrin may be injected per rectum and alkalis administered. The most convenient form of alkali is a combination of bicarbonate of soda with half as much bicarbonate of potash and the addition of one-twentieth as much of the carbonates of calcium and magnesium. The quantity of bicarbonate of soda, in excess of 2 drachms, necessary to render the urine alkaline is a good measure of the degree of acidosis, and in extremities it should be remembered that subcutaneous or intravenous injections are methods of greater certainty.

To prevent the occurrence of acidosis after anaesthesia, especially in susceptible subjects, it is requisite to see that the patient does not proceed to operation with the stomach empty, but a fairly substantial meal, always containing carbohydrates, such as a plate of rice pudding, should be supplied shortly before the appointed time. In children the urine should be tested for acetone by Legal's method, and when present the operation should be delayed until, by the administration of glucose and bicarbonate of soda in drachm doses every two hours during the day, the acetonuria disappears.

In urgent cases glucose can be administered by the rectum, or advantage can be taken of the glucose tubes which can now be purchased and are employed for making solutions for use by hypodermoclysis.

It is instructive to contrast the two dietetic procedures described in this chapter - the one insisting on the limitation of the carbohydrates, the other deprecating too strict a limitation of the same food substances and enlarging on the danger incident to such a course in certain circumstances. It is an example of the ease with which the metabolic equilibrium may be disturbed, a warning not too rashly to ignore the nutritive requirements of the body, and an apt illustration of the necessity for a well-balanced ration suited to the indications of each individual case.