The standard food analysis regarded of so great importance, had a number of shortcomings, as a method of determining the nutritive value of a food. By means of it the great differences in the make-up of the proteins, which give them very unlike value in nutrition, were not revealed. Protein from one source was assumed to be as good as that from another. Beans and peas, for example, contain about 23 per cent of protein, and compare favorably in this constituent with lean meat. These legume seeds were regarded for many years as essentially the equivalent of meat as a source of protein. They were sometimes spoken of as "the poor man's meat." We now know that the proteins of these seeds have peculiarities in their composition which make them of relatively low value in nutrition when they serve as the sole source of protein in the diet, or when they are combined with some of our most important food grains. Meat proteins are decidedly superior to them as supplements to most of the proteins of our vegetable foods. This was only brought to light by later investigations.

The analysis of foods for their carbohydrate content was not effective as a means of arriving at a decision as to the energy value of this portion of their substance. Man and animals can use starch and sugars as a source of energy, but not cellulose or hemicelluloses. The best method for differentiating between the several types or carbohydrates was that based on the ease with which they are converted into glucose or other simple sugars by the action of acid of certain concentration. The true celluloses, of which paper is made, are not dissolved by the reagents used in food analysis, and could be filtered off, washed, dried and weighed. There are many kinds of vegetable foods which contain considerable amounts of hemicelluloses, a class of carbohydrates which are easily converted by the action of acid into simple, soluble sugars in the process of analysis. Accordingly, these were estimated as part of the fraction called "nitrogen-free extract," and were, in the absence of specific data as to their digestibility, estimated in calculations of dietaries, as the equivalent of starch or sugar.

Hemicelluloses are not acted upon by the digestive secretions of man or animals. They have, therefore, no food value except that imparted to them through the agency of certain microorganisms inhabiting the digestive tract. These organisms may bring about their fermentation with the production of such organic substances as acids and alcohols as intermediary products. These may be absorbed and utilized by the body as a source of energy for mechanical work or for heat production.

The method for the estimation of fat in food-stuffs was fairly satisfactory, but the materials weighed as fat always contain waxes, cholesterol, chlorophyll, etc. These, with the possible exception of chlorophyll, have no food value, and are inert and chance components of the diet. The fats and carbohydrates were regarded as essentially sources of energy for work or for heat production, and were rightly considered as being able to replace each other in the diet in isodynamic quantities. One gram of carbohydrate or of protein has a caloric value of about 4.1, while 1 gram of fat has about twice this fuel value.

The mineral content of a food-stuff was determined in the standard method of analysis by burning a weighed sample in a dish of known weight. After the organic matter was all destroyed, the dish was cooled in a dry atmosphere and reweighed. The difference between the weight of the dish and the dish plus the ash gave the weight of the mineral matter. Little significance was attached to the inorganic content of foods until recent years, for it was assumed that, since all the elements which are required by the body are found in all foods, the amounts might vary considerably and still meet the needs of the body. In special cases the deficiency of a food in a particular element was so pronounced that it attracted attention. Thus milk is especially poor in iron, and since this element is a constituent of the hemoglobin of the blood, it was early recognized that a young child should not be too long confined to an exclusive milk diet, but should be given some food which would supplement milk in this respect.