The experiments in 1905,1906, and 1907, included in this research on the effect upon the metabolism of the ingestion of food, were made at Wesleyan University, Middletown, Connecticut, and those subsequent to 1907 in the Nutrition Laboratory in Boston. Only the respiration calorimeter was used in the Middletown experiments. In the Boston experiments not only the chair and bed respiration calorimeters were employed, but also two forms of respiration apparatus - i. e., the universal respiration apparatus and the Tissot respiration apparatus. With the calorimeters, the carbon-dioxide production, the oxygen consumption, and the heat production were determined; with the respiration apparatus, determinations were made only of the respiratory exchange, the heat production being obtained by indirect calorimetry. The several apparatus have been fully described elsewhere; brief descriptions are included in this publication.1 The general plan of experimenting has been given in the preceding section; the routine with the various apparatus has also been outlined with more or less detail in the discussion of the results of the experiments.

In the Middletown calorimeter considerable freedom of movement was possible. In the 24-hour experiments with this apparatus the subject was able to carry out the ordinary routine of a day, restricted only by the confines of the chamber and the experimental requirements for a minimum amount of activity. In the waking hours he sat in a chair; in the sleeping periods he lay on the bed. During the 8-hour experiments with the same apparatus he sat quietly in a chair. In the chair calorimeter in Boston the subject also sat in a chair for the entire experimental period and was instructed to reduce all movement to the minimum. In all these apparatus the water and urine vessels and telephone were placed conveniently near the subject, so that they could be used with the least activity possible. In the bed calorimeter the subject lay on an air mattress with few or no major changes of position. In the experiments with the two respiration apparatus the subject lay on a couch, practically without movement, during the periods.

In the Middletown calorimeter experiments the individual periods were usually 2 hours in length; in the Boston calorimeter experiments they were shortened to 1 hour and in a few instances to 45 minutes. With the universal respiration apparatus the periods ordinarily approximated 15 minutes in length; with the Tissot apparatus the periods were generally shorter. The observations in the experiments with the two respiration apparatus were not continuous, there being intermissions varying in length according to the conditions of experimenting, food material used, etc. It was usual to make a 15-minute observation every half hour during the experimental period, but the intermissions were often much longer. The intermissions with the Tissot apparatus were usually brief. As a rule, the subject lay on the couch during the intermissions with both apparatus and was more or less quiet.

1For a description of the Middletown calorimeter, see Atwater and Benedict, Carnegie Inst. Wash. Pub. No. 42, 1905. For the chair and bed calorimeters, see Benedict and Carpenter, Carnegie Inst. Wash. Pub. No. 123, 1910. For the universal respiration apparatus, see Benedict, Deutsch. Arch. f. klin. Med., 1912, 107, p. 156; Carpenter, Carnegie Inst. Wash. Pub. No. 216, 1915. For the Tissot respiration apparatus, see Tissot, Journ. de physiol. et de path, gen., 1904, 6, p. 688, and Carpenter, Carnegie Inst. Wash. Pub. No. 216, 1915, p. 61.

In the Middletown experiments the movements made by the subject were recorded by an observer, but in nearly all of the Boston experiments the degree of muscular repose was shown by some form of graphic record. In many of the calorimeter experiments observations were made of the body-temperature; the data recorded were secured per rectum, by means of an electrical-resistance thermometer.

Records of the pulse rate were obtained with the Fitz pneumograph in the Middletown experiments, but in the calorimeter experiments in Boston and in the experiments with respiration apparatus the Bowles stethoscope was used, the counts being made by a special observer.

Records of the respiration rate were secured in practically all of the Middletown and Boston calorimeter experiments and in the experiments with the tension-equalizer form of the universal respiration apparatus by means of the Fitz pneumograph. In the experiments with the respiration apparatus the pneumograph was connected with a kymograph, thus giving graphic records of the respiration. In the observations with the spirometer form of the universal respiration apparatus, the graphic record of the respiration was provided by a mechanical device attached to the spirometer instead of by the Fitz pneumograph.

In several groups of the Boston experiments, i. e., those with water, coffee, and beef tea, records of the blood pressure were obtained with the Erlanger sphygmomanometer.