THE ordinary method of figuring dietaries, using the tables of food by percentage composition, involves much tedious figuring so that such dietaries are very seldom calculated in practice. Although there is no settled "best diet" for human beings applicable to all conditions a scientific diet cannot be planned unless it is known definitely what people eat.

In one of the Bulletins of the School, Professor Irving Fisher's article "A Graphic Method in Practical Dietetics" was reviewed. The number of the Journal of the American Medical Association April 20, 1907, and the reprint of the article, both are exhausted. As his method of calculating food values is very valuable we are republishing in this Supplement the tables given in the original article.

Dr. Fisher's method of calculation is given in the article as follows: "Two methods have hitherto been used for computing proportions of proteids, fats, and carbohydrates. One consists in using the tables of percentages by weight of proteids, fats and carbohydrates; the other, Dr. J. H. Kellogg's, in using a table which gives the number of calories in the form of proteids, fats and carbohydrates per ounce of each kind of food. These may be described, respectively, as the method, of 'weight per cent' and the method of 'calories per ounce.' The method here suggested is different from either, and may be called the method of ' calories per cent'.

"Jt takes as its starting point not a unit of weight, but a unit of food value, called a ' standard portion' of each kind of food. A ' standard portion' is defined as that amount of food which contains 100 calories, or food units. A table is constructed which gives the weight in a ' standard portion' of each particular kind of food, and out of the 100 calories contained therein the number of calories in the form of proteids, fats and carbohydrates.

" In order to carry out this method food should be served at the table in ' standard portions,' or simply multiples thereof. The amount of milk served, instead of being a whole number of ounces should be (for average milk) 4.9 ounces - the amount that contains 100 calories. This 'standard portion' constitutes about two-thirds of an ordinary glass of milk. Of the 100 calories which it contains 19 will be in the form of proteid, 52 in fat, and 29 in carbohydrates. In other words, of the food value of milk, 19 per cent is proteid, 52 per cent fat, and 29 per cent carbohydrates.

"One advantage of this method is apparent at once. It enables us to make a true comparison between different foods as to the relative amounts of proteids, fat and carbohydrate. The other methods are misleading in this regard. For instance, though it is well recognized that milk is a higher proteid food than pecan nuts, yet, if we compare milk and the pecans on the basis of the method of weight per cent, we shall find that the pecans appear three times as rich in proteid, milk containing 3.3 per cent and pecans 11 per cent. But if we compare them on the basis of calories per cent we find that, while milk contains 19 calories of proteid out of each 100 of total calories, pecans contain only 6, milk showing three times as much proteid as pecans. * * *

Moreover by having the composition of foods in food units (calories) the fats are on the same basis as the proteids and carbohydrates. This is not the case in composition by weight, for one ounce of fat in the body produces 264 calories of heat and energy, while one ounce of carbohydrate or pro-teid produces only 116 calories. Or in grams, one gram fat gives 9.3 calories, one gram carbohydrate or proteid 4.1 calories. (A calorie is approximately the amount of heat required to raise the temperature of 1 pound of water, 4° F).