It is obvious that one kind of food cannot take the place of a totally different kind. For example, while one saline may, to some degree replace another in osmotic phenomena, and while various strong acids may be substituted for hydrochloric in gastric digestion, sodium cannot entirely supplant potassium, nor on account of its greater toxicity can potassium take the place of sodium to even so great a degree. The insolubility of calcium salts renders them necessary in the formation of bone, and while magnesium salts are to some extent associated with calcium salts they cannot be substituted to any great degree. It is unnecessary to duplicate these statements for the various inorganic constituents of the body.

Provided that the various essential food ingredients are administered in sufficient amount, and in forms available for assimilation, it is theoretically immaterial in just what natural combination they are employed. For example, water may be entirely, though not advisably, replaced by various beverages and semi-solid or solid foods containing it. Indeed, about 200 c.c. of water is regularly produced in the system by the oxidation of hydrogen in organic combination, especially in carbohydrates, but also in fats and proteins. It is scarcely necessary to state that no one natural food stuff, or group of food stuffs is indispensable, or even particularly necessary for good nutrition, though perhaps an exception should be made of milk in the case of infants.

However, the foregoing statements require qualification in both directions, depending on possibilities of transmutation and on various limitations of theoretic knowledge and on empiricism. For instance, while vegetarianism is possible for protracted, if not indefinite periods, the entire absence of animal proteins, or even of meats, does not conduce to the best state of health, nor can this fact be entirely ascribed to the lack of the iron of meats. Whether it is due to some essential difference between animal and vegetable protein - and chemic differences are easily demonstrable, even in the course of digestion - or to the presence in meat of salts or other substances not present in adequate amount or proper proportion in vegetable foods, is not known. We even note empirically that different individuals seem to reach their optimum on different proportions of animal and vegetable protein, and even on different proportionate use of different groups of flesh, as of quadrupeds, fish and fowl, or of different species of meat, as beef, mutton, pork, etc.

As has been stated, fat is theoretically and, to a large degree, practically replaceable by protein and carbohydrate. Thus, obviously, no one kind of fat, as olein, palmitin or stearin is indispensable, and these different kinds seem to be entirely vicarious, except as the absorbability depends upon the melting point, so that relatively more stearin is required to replace the others, progressively. To a considerable degree the exact chemic constitution of the deposited fat of an animal depends upon its diet, and may be modified by artificial limitation of diet. To what degree the characteristics of the fat of any animal depend upon essentially specific modifications of its physiologic chemistry, and to what degree it is accidental, depending upon the habitual diet, is not yet determined, but it is not improbable that the latter factor is relatively of great importance, and it may even be that the yellow color of human fat, as contrasted with the white fat of most quadrupeds, is due to the large consumption of fowls and of butter.

It should be stated that the caloric value of the different varieties of fat and of carbohydrates differs slightly, though not enough to be considered even in quite exact studies of calorimetry, owing to the preponderance of other sources of error.

As is shown by a study of physiology, all assimilable carbohydrates ultimately produce simple hexoses, and unless the conditions are exceptional, dextrose. For the first year of life the human being normally uses only lactose, and for about the first six months cannot digest starches. In adult life we find very different individual appetites for starches and sugars respectively, and it is difficult to say how far these are physiologically interchangeable, much less to state how far any one sugar is available for nutrition. Both by practical experience and animal experiment, we know that the various sugars are not desirable substitutes for starch. Whether this fact is to be explained solely on the ground of irritation of mucous membrane, liability to fermentation and difficulty of preventing the passage of too large amounts of sugar into the blood, on account of overtaxing the glycogenic function of the liver, or whether there are more recondite reasons is not known.

Generally speaking, the appetite for sugars and for starch is complementary, the individual who craves cereals, breadstuffs, potatoes, etc., desiring little sugar and vice versa, subject to the qualification that the individual appetite for carbohydrates generally is inverse to that for fats and proteins respectively and together, and that the appetite for all foods varies. It does not appear that the craving for sugars is entirely a luxurious appetite, and it may even be that it depends upon a relative - but still not abnormal - weakness of ferments, or even upon relative shortness of the intestine. The consumption of sugar has increased enormously of late years, commensurate with the development of the cane and beet sugar industries (these two sugars being identical with the pure sugar of the maple), and also with the development of the culinary and confectioner's art. It may be considered perfectly normal for an individual to take 100 grams a day of carbohydrate in the form of sugar, and certain persons double this quantity.

While fats are theoretically replaceable by other organic foods, and while it even appears that if absolutely fat-free foods were available, no harm would result from the absolute suppression of the ingestion of fats, this statement cannot be made for carbohydrates. Even aside from the inevitable increase of nitrogenous waste products from the vicarious use of proteins, and the fact that it is physiologically impossible for the system to increase the digestion and absorption of fats beyond a pretty definite maximum, it appears that about 80 grams of carbohydrate are necessary to prevent catabolic disturbances of a fatal nature.

Proteins, unlike fats and carbohydrates, not only are employed to yield heat and energy but to replace tissues, and are absolutely non-replaceable below a pretty definite minimum, either by fats, carbohydrates, or even non-protein nitrogenous matter, such as gelatin and purins, free or combined. But, on the other hand, it should be borne in mind that slightly beyond the minimum of 50 to 80 grams a day (the exact amount not yet having been firmly established), proteins not only can but should be replaced by other foods, providing that there is no such obstacle as occurs in diabetes and obesitv.

Regarding the transmutability within the system of the different organic ingredients, it has been definitely proved that both proteins and carbohydrates may produce fat, and may be so deposited. It has also been definitely proved that dextrose and glycogen may be formed from fat or protein, and probably from both, and not only in severe grades of diabetes but in the normal body. Similar information regarding gelatin, etc., is not available.

There is a very general popular notion that a reserve of energy can be accumulated by an excess of ingestion; and this notion is of practical importance, partly because it is measurably true, but largely because it is not true in the crude sense and because it leads to overeating.

Voit has shown that when superalimentation is induced in a dog - barring, of course, gross waste from failure of digestion and saprophytic destruction of food - 91.5 per cent, of the carbohydrate that is in excess of the caloric demands is deposited as fat, and 95 per cent, of a similar excess of fat. Exact knowledge with regard to the deposition of fat from protein is not available.

Deposited fat is the only way in which a reserve of energy of notable quantitative importance can be accumulated. The human body may accumulate at least 50 kilograms of fat, or nearly 500,000 calories in potential energy, or enough, theoretically, to last the body for 200 days. Unfortunately, when there is a demand for this energy - which rarely happens in civilized life - the body may lack the power to utilize it, and it can never be used without concomitant wasting of protein tissues. Under no circumstances can the body reserve energy to last for much more than 40 days without renewal of organic nutriment.

Carbohydrates can be stored in the body in the form of dextrose in solution and glycogen in muscle and gland cells, especially in the liver, to the extent of about 300 grams, or barely enough to represent half a day's caloric demands.

A comparatively small amount of protein in excess of the daily demand for calories can be stored. Voit's experiments on dogs showed that the proportionate storage was about 5 per cent., at most 8.5 per cent., while vonNoorden estimated that the human being could thus store 10 per cent, of the excess protein. In other words, in any day the energy eliminated will amount to 90 per cent, of that ingested in the form of protein, however much protein is taken. A very small and not definitely known amount of the reserve protein occurs in soluble, or at least unfixed form, in the blood and body juices generally. The rest is stored in hypertrophic or newly formed cells. Thus, while protein may be deposited in forming muscle, and increasing the tissues generally and while there is obviouslv so much more tissue that may be available in starvation or relative inanition, there can be no important storage of protein in the body.

Certain eminently practical deductions can be drawn from these facts. 1. The protein ration should not be increased with the idea of establishing a reserve of force and tissue material, as such a course leads to increased labor of the emunctories and to an excess of possibly toxic products. It is not. however, established that the ideal diet should include only the minimum of protein required to prevent loss of nitrogen. 2. The proper normal nutrition of the body requires at least two meals a day 3. The storage of fat should never exceed the amount which represents the caloric demands of 40 days, about 11 kilograms beyond the normal minimum of fat required for mechanic packing (not the fat remaining in an extremely lean body) - about 6 kilograms for a man of 70 kilograms total weight. That is, a man of full stature should never carry more than a total of about 17 kilograms of fat, and obesity of more than about 11 kilograms beyond the standard weight can serve no useful purpose, even in the emergency of deprivation of food. 4. While by overeating it is comparatively easy to deposit 250 to 500 grams of fat a day (1-2 to 1 pound), even exposure to cold and muscular exertion in the entire absence of food, can scarcely utilize more than 300 grams of fat a day (a little over 1-2 pound), and while greater loss may occur in sickness, the physiologic reduction of weight on a low diet can never equal the rate at which fat can be deposited, unless for very brief periods.