This section is from the book "Chemistry Of Food And Nutrition", by Henry C. Sherman. Also available from Amazon: Chemistry of food and nutrition.
The most direct, and in some respects most convincing, way of ascertaining the energy metabolism is by the method of direct calorimetry. This consists in measuring the total energy expenditure of the body as heat or as heat and mechanical work by confining the subject in a chamber permitting of actual measurement of the heat produced. It was not until the development of the Atwater-Rosa-Benedict respiration calorimeter that complete and satisfactory data covering periods of one to several days were obtained. This apparatus consisted of an air-tight copper chamber, surrounded by zinc and wooden walls with air-spaces between, and was large enough for a man to live in without discomfort, being about 7 feet long, 4 feet wide, and 6½ feet high. An opening in the front of the apparatus, which was sealed during an experiment, serves as both door and window and admits sufficient light for reading and writing. A smaller opening, having tightly fitting caps on both ends, was used for passing food, drink, excreta, etc., into and out of the chamber. The chamber was furnished with a folding bed, chair, and table, and was ventilated by means of a current of air which passed usually at the rate of about 2½ cubic feet per minute. At first this ventilating air current was maintained and measured by means of a specially constructed meter pump which also automatically took samples of the air for analysis. Later the apparatus was so modified as to make use of the same air throughout an experiment, the carbon dioxide and water given off by the subject being removed by circulating the air through purifying vessels, and the oxygen which the subject uses being replaced by adding weighed amounts of oxygen to the air current as required.* By this means it is possible to carry out, in the calorimeter, metabolism experiments in which the oxygen and hydrogen as well as the carbon and nitrogen balances are determined, and from these data the gain or loss of carbohydrate as well as of protein and fat can be determined.
* Figure 8 indicates diagrammatically the ventilating system as applied in one of the later forms of apparatus.

Fig. 8. - Diagram of ventilation of respiration calorimeter. The air is taken out at lower right-hand corner and forced by the blower through the apparatus for absorbing water and carbon dioxide. It returns to the calorimeter at the top. Oxygen can be introduced into the chamber itself as need is shown by the tension equalizer. Courtesy of Dr. F. G. Benedict and the Carnegie Institution of Washington.
The ventilating air current is so regulated that it enters and leaves the calorimeter at the same temperature; and between the copper and zinc walls are placed a large number of thermo-electric junctions connected with a delicate galvanometer by means of which each wall is tested every four minutes, day and night, during the progress of an experiment, and the minute amounts of heat which may pass to or from the calorimeter through its walls are quickly detected and made to balance each other. Thus there is no gain or loss of heat either through the walls of the chamber or by the ventilating air current, and the heat given off by the subject can leave only by the means especially provided for carrying it out and measuring it. A part of the heat liberated is carried from the chamber in latent form by the water vapor in the outgoing air, which is accurately determined. The rest of the heat is brought away by means of a current of cold water circulating through a copper pipe coiled near the ceiling of the chamber. The quantity of water which passes through the pipe and the difference between the temperature at which it enters and that at which it leaves the coil are carefully determined and show how much heat is thus brought out of the chamber.
 
Continue to: