Inulin Biscuit.-Put 50 gm. (11⁄2 oz.) of inulin in a large porcelain basin, place this over a hot-water bath, and with 30 c.c. (1 oz.) of milk and as much hot water as may be necessary, rub up into a smooth dough, into which the yolks of four eggs and a little salt have been mixed. To this add the whites of the four having first beaten them to a foam, and working them in carefully. Bake in tin molds smeared with butter. The taste of the biscuit may be improved by adding vanilla or other flavoring extract. Inulin is too expensive to be used by the average patient.

eggs,

Peanut Flour.-This contains about 25 per cent. of carbohydrates. The peanut kernels should be boiled in water for half an hour to extract a portion of the oil which they contain. They should then be dried, and rolled into fine particles with a rolling-pin. Place the kernels in boiling water acidulated with tartaric acid or vinegar, in order (1) to extract saccharin elements; (2) overcome the taste and odor of the peanut; (3) to prevent emulsification of the remaining oil. When they have been thoroughly boiled in acidulated water, the ground kernels should be subjected to dry heat and then rolled into a fine flour. This flour may be made into a form of porridge with milk; bread and biscuits may also be baked from it; and it may be made into the form of a German pancake.-(Stern.)

Home-made Substitute for Bread.-Beat up thoroughly six eggs; add a teaspoonful of baking-powder or its chemical equivalent, and one-quarter of a teaspoonful of salt, and beat again. Pour this mixture into hot waffle-irons smeared with butter, and bake in a very hot oven. By way of variety almonds (powdered) may be added. These biscuits may be eaten hot with butter and cheese.

Sugar-free Milk for Diabetic Feeding.—Take 1 liter of skim milk, heat to a temperature of 30° C., and add 10 c.c. of glacial acetic acid, diluted with 100 c.c. of water. Mix, and allow the mixture to stand for about fifteen minutes. Collect the separated casein, and let it drain on very fine muslin, using no pressure. Remove the casein to a mortar, rub into a smooth paste, add liter of distilled water, and strain as before. Repeat this washing of the casein twice. Transfer to a mortar, rub until quite smooth, and add 2 gm. of potassium hydrate dissolved in 100 c.c. of water (or as much of the potassium hydrate as is necessary to make the product just alkaline to phenolphthalein). Add 100 gm. of ordinary Devonshire clotted cream, 5 gm. of gelatin, previously dissolved, 0.06 gm. (1 gr.) of saccharin, and water, at about 38° C., up to 1 liter. Lastly, strain through fine muslin.-(Hutchinson.)

THE CHEMICAL COMPOSITION OF AMERICAN FOOD MATERIALS.

THE material in this section has been taken from the revised edition of Bulletin No. 28 of the Experiment Stations of the Department of Agriculture of the United States. This very valuable bulletin was prepared by W. O. Atwater and A. P. Bryant, and represents the best compilation of analyses of American food materials down to 1899. Only the averages have been abstracted from the tables; for ordinary purposes these will be found to be sufficient; for the complete tables the reader should refer to the original bulletin.

The earliest quantitative food analyses were made in 1795 by Pearson, in England, who analyzed potatoes. In 1805 Einhoff analyzed potatoes and rye. Later other workers gave various accounts of their work, but the great impetus to the study of food materials was given by Liebig and his followers, whose work was done chiefly in the period between 1840 and 1865. About 1864 Henneberg and his associates elaborated the so-called Weende method for proximate analysis. This method, with slight alterations, is used to-day wherever food analyses are made. "The methods followed in different countries agree so closely that for the last twenty years it has been possible to accept analyses by chemists in different parts of the world and compare them with one another without hesitation" (Atwater and Bryant). Since the establishment of the experiment stations an enormous amount of work has been done. The results given in the tables (on pp. 689-715) show the averages of thousands of analyses; these, together with the accompanying list, have been taken directly from Atwater and Bryant's publication.

EXPLANATION OF TERMS.1

The terms used in reporting analyses of foods and feedingstuffs need some explanation. Some of these terms have a technical meaning which is well recognized and understood by 1 These definitions are quoted from Atwater and Bryant.

scientists, although the dictionaries and similar books of reference have not yet included these uses in their definitions. In other cases the same word has been used by scientists in different ways. The more usual terms are defined and explained below in the sense in which they are employed in the following table and the publications of the Experiment Stations of the United States Department of Agriculture.

COMPOSITION OF FOOD MATERIALS.

Ordinary food materials, such as meat, fish, eggs, potatoes, wheat, etc., consist of:

Refuse. As the bones of meat and fish, shells of shellfish, skin of potatoes, bran of wheat, etc.

Edible Portion.-As the flesh of meat and fish, the white and yolk of eggs, wheat flour, etc. This edible portion consists of water (usually incorporated in the tissue and not visible as such), and nutritive ingredients or nutrients.

The principal kinds of nutritive ingredients are protein, fats, carbohydrates, and ash or mineral matters.

The water and refuse of various foods and the salt of salted meat and fish are called non-nutrients. In comparing the values of different food materials for nourishment they are left out of

account.

Protein. This term is used to include nominally the total nitrogenous substance of animal and vegetable food materials, exclusive of the so-called nitrogenous fats. Actually it is employed, in common usage, to designate the product of the total nitrogen by an empirical factor, generally 6.25.

This total nitrogenous substance consists of a great variety of chemical compounds, which are conveniently divided into two principal classes, proteids and non-proteids.

The term proteid, as here employed, includes: (1) The simple proteids, e. g., albuminoids, globulins, and their derivatives, such as acid and alkali albumins, coagulated proteids, proteoses, and peptones; (2) the so-called combined or compound proteids; and (3) the so-called gelatinoids (sometimes called "glutinoids ") which are characteristic of animal connective tissue.

The term albuminoids has long been used by European and American chemists and physiologists as a collective designation for the substances of the first two groups, though many apply it to all three of these groups. Of late a number of investi

gators and writers have employed it as a special designation for compounds of the third class."

The term non-proteid is here used synonymously with nonalbuminoid, and includes nitrogenous animal and vegetable compounds of simpler constitution than the proteids. The most important animal compounds of this class are the so-called "nitrogenous extractives" of muscular and connective tissue, such as creatin, creatinin, xanthin, hypoxanthin, and allied cleavage products of the proteids. To some of these the term "meat bases" has been applied. The latter, with certain mineral salts (potassium phosphates, etc.), are the most important constituents of beef-tea and many commercial "meat extracts."

The non-proteid nitrogenous compounds in vegetable foods consist of amids and amido acids, of which asparagin and aspartic acid are familiar examples.

The ideal method of analysis of food materials would involve quantitative determinations of the amounts of each of the several kinds or groups of nitrogenous compounds. This, however, is seldom attempted. The common practice is to multiply the percentage of nitrogen by the factor 6.25 and take the product as representing the total nitrogenous substance. For many materials, animal and vegetable, this factor would be nearly correct for the proteids, which contain, on the average, not far from 16 per cent. of nitrogen, although the nitrogen content of the individual proteids is quite varied. The variations in the nitrogen of the non-proteids are wider, and they contain, on the average, more than 16 per cent. of nitrogen. It is evident, therefore, that the computation of the total nitrogenous substance in this way is by no means correct. In the flesh of meats and fish, which contain very little of carbohydrates, the nitrogenous substance is frequently estimated by difference—i. e., by subtracting the ether extract and ash from the total waterfree substance. While this method is not always correct, it is oftentimes more nearly so than the determination by use of the usual factor.

The distinction between protein and proteids is thus very sharp. The latter are definite chemical compounds, while the former is an entirely arbitrary term used to designate a group which is commonly assumed to include all of the nitrogenous matter of the food except the nitrogenous fats.

1 United States Department of Agriculture, Office of Experiment Stations, Bulletin 65, p. 118.

In the tables herewith the common usage is followed, by which the protein is given as estimated by factor-i. e., total nitrogen multiplied by 6.25. In the analyses of meats and fish, however, the figures for protein "by difference" are also given. Where the proteid and non-proteid nitrogenous matter have been estimated in a food material the proportions are indicated in a footnote.

Fats. Under fats is included the total ether extract. Familiar examples of fat are fat of meat, fat of milk (butter), oil of corn, olive oil, etc. The ingredients of the "ether extract" of animal and vegetable foods and feeding-stuffs, which it is customary to group roughly as fats, include with the true fats various other substances, as fatty acids, lecithins (nitrogenous fats), and chlorophylls.

difference.

Carbohydrates.-Carbohydrates are usually determined by They include sugars, starches, cellulose, gums, woody fiber, etc. In many instances separate determinations of one or more of these groups have been made. The determinations of "fiber" in vegetable foods-i. e., substances allied to carbohydrates but insoluble in dilute acid and alkali, and somewhat similar to woody fiber-are given in a separate column.

The figures in parentheses in the crude-fiber column show the number of analyses in which the fiber was determined. The figures for "total carbohydrates" include the fiber, as well as sugars, starches, etc. Where the sugars or starches have been determined separately, footnotes are added giving the average results.

Ash or Mineral Matters.-Under this head are included phosphates, sulphates, chlorids, and other salts of potassium, sodium, magnesium, and other metallic elements. Where analyses of the mineral matters have been found they are added in the form of footnotes. These results usually give the percentage composition of the ash as produced by incineration rather than the proportions in which the different mineral ingredients occur in the food material.

Fuel-value.-By fuel-value is meant the number of calories of heat equivalent to the energy which it is assumed the body would be able to obtain from one pound of a given food material, provided the nutrients of the latter were completely digested. The fuel values of the different food materials are calculated by use of the factors of Rubner, which allow 4.1 calories for a gram of protein, the same for a gram of carbohydrates, and 9.3 calories per gram of fats. These amounts correspond to 18.6