On the discovery of diabetes, especially if sugar and nitrogenous elimination are high, the case should be placed for a few days on a carbohydrate-free diet. The following conforms nearly to the requirements:

Meats, including poultry, fish and shell fish may be used freely except that the liver, and other glandular viscera are objectionable as containing glycogen.

Fat is best given in the form of butter, cream (unless absolute exclusion of lactose is necessary), salt pork, bacon, olive oil, mayonnaise, salads with oil etc., gravy etc.

Whipped cream adds to the palatability of various desserts.

Nuts are rich in fat and protein. The only common nut to be avoided is the chestnut, on account of its content of starch. Peanut butter and various meals and cakes made of nuts may be used.

Broths, meat teas, bouillon and clear soups may be used. Flour, rice, barley, noodles, vermicelli and milk are to be avoided in making soups. The innutritious vegetables mentioned subsequently may be added.

Saccharin may be used to sweeten beverages, jellies, etc. It is, however, irritating, and in considerable quantities it inhibits digestion. It is permissible in small quantities if not used too continuously. Unfortunately, the appetite for sweets is a genuine one for sugar and not merely the craving of a taste.

Gelatin may be used to prepare meat jellies or, variously flavored, and sweetened with saccharin, to prepare desserts. While not a protein, gelatin may be used to replace protein, to the extent of 25 grams a day or thereabouts, provided the minimum of protein (about 50 or 60 grams) is also given.

Beverages. Tea, coffee, cocoa-nibs but preferably not cocoa, may usually be given with cream or even milk, but not with sugar. Carbonated and mineral waters, Rhine and still Moselle wines, claret etc., but not sweet wines, malt liquors etc., may be used. By using essential oils, citric acid or acid phosphates and saccharin, various fruit beverages may be imitated.

Condiments may be used to give taste and variety to foods.

Gluten, bran, aleuronat, almond flours etc., are available only to a limited degree. Gluten breads rarely contain less than 35% of carbohydrates and while, in theory the reduction of carbohydrate from 50% or 60% is considerable, patients will usually prefer the use of real bread stuffs in correspondingly less amount.

A nearly carbohydrate-free milk may be prepared as follows: To 1 liter of skimmed milk, add 100 c.c. of 10% glacial acetic acid. After 15 minutes, filter through fine muslin without pressure. Rub up the curds to a fine paste and wash twice through the filter, using 1/2 liter of distilled water each time. This leaves the casein, free from lactose. Dissolve the curds in liquor potassae, using about 100 c.c. of a 2.5% solution, testing separate portions with phenolphthalein, so as to stop just short of complete neutralization. Add 100 grams of clotted cream, to bring the fats to the normal. Add 5 grams of gelatin, previously dissolved to give the proper viscosity. A little saccharin may be added to give the normal sweetness of milk. Make up the dissolved curds to the original volume, 1 liter. This has almost the normal nutritive value of milk, less the lactose, and can scarcely be distinguished by the taste.

Innutritious vegetables which may be used to give bulk and variety to the diet and to facilitate the use of meat, salad dressing etc., but which contain scarcely any digestible carbohydrate are as follows: cabbage, cauliflower, artichokes, Brussels sprouts, broccoli, green string beans, mushrooms, tomatoes, lettuce, endive, coleslaw, olives (which contain some oil), cucumbers fresh or pickled, including gherkins, young onions, cress, mustard, and the green ends or tops of asparagus, spinach, dandelion, turnips, celery etc.

Eggs may be used in a great variety of ways, including custards sweetened with saccharin, and cakes, provided there is no objection to the use of small quantities of carbohydrate in milk, flour, etc.

Fruits except those obviously starchy, as bananas, or quite sweet, may usually be given in moderation. Often diabetics can take considerable quantities of fruit sugar (laevulose) which is found in many fruits.

Vegetables to be particularly avoided on account of their richness in starch are: potatoes, Irish and sweet, beets, carrots, turnips, parsnips, peas and beans, all cereals and, hence, bread stuffs.

Remember that all lists of what the diabetic may and may not take, are arbitrary. Very few natural food stuffs, especially vegetable, are entirely free from carbohydrates. Malt liquors and sweet beverages, especially cider and wines, seem to have a worse effect than can be explained by their content of sugar. Otherwise, and allowing for idiosyncrasies, it makes very little difference what food the diabetic takes, providing the total ingestion of carbohydrate is within the limits of his power to oxidize sugar in the blood.

Most diabetics can and should take moderate quantities of foods containing carbohydrates. Oefele states that 88% can take 35 grams of cane sugar a day (6 ordinary lumps). It is safer not to give cane sugar, malt sugar or dextrose but as much as a liter of milk, containing about 40 grams of lactose may usually be taken in a day. 100 grams of Irish potato (one ordinary sized potato) contain 2 grams of protein and 20 of starch. 100 grams of bread stuff (about 2 ordinary slices of bread, 20 small round crackers; contain about 50 - 60 grams of carbohydrate. Oatmeal and other cereals contain about the same percentage and an ordinary serving for breakfast, 30 grams, may usually be given.

Remember that there is no special virtue in the milk cure, oatmeal cure, potato cure etc., as applied to diabetes. Any or all of these and other foods may be used, provided the carbohydrate does not exceed the amount which can be assimilated without too great glycosuria.

Remember that it is better for the diabetic to have too much sugar in the urine than too much acetone, and its congeners in the blood. The diabetic who seems to require a continued, carbohydrate-free diet will die anyway.

Learn to know the composition of the ordinary foods, at least approximately, and to have some idea of the actual weight, and, hence content in fat, protein and carbohydrate, of ordinary servings.

He sure that the diabetic has his food well cooked and neatly served. Give him as great a variety as possible, especially by using different flavorings and methods of serving the standard diet. There is no other disease in which it is so important to cater to the appetite of the patient while insisting on the principles of dietetics. If this precaution is not followed, the psychic hardship will react upon the diseased process or else the patient will break loose from restraint and will kill himself by dietetic excesses.

Remember that neither the dietetic nor other treatment of diabetes can be intelligently carried out, unless we have an approximate knowledge of what the patient ingests and what he excretes. This does not mean that all of his food and excreta must be measured and analysed daily, as in performing metabolism experiments, but it does mean that the approximate clinical estimations must be made frequently.

It is senseless to let a patient take carbohydrates and then administer yeast to decompose them. It would be more rational to let him eat and then make him vomit.

Remember that the diabetic may present special contraindications to any of the foods allowed in diabetes, just as if he were free from the underlying disease. If this contraindication is important, as in complicating nephritis, hepatic sclerosis or essential digestive failure, a compromise must be effected according to the merits of the individual case.