238 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



measured by the calorimeter was 100.01 per cent of the amount generated by the electric current. 

 The averages of the results obtained in seventeen alcohol tests are summarized in the following 



table: 



Summary of results of tests in which alcohol was burned in the calorimeter. 



Carbon 



dioxide. 



Water. 



Grams. Grams. 



Amount required 19, 239. 8 1 2, 264. 4 



Amount found '19, 206. 9 i 12, 379. 1 



Per cent. Per a ni 



Ratio i >f amount found to amount required 99. 8 . a 100. 9 



Heat, 



Caloric?. 



64, 554. 1 

 64,513.3 

 /'< i c< nt. 

 99.9 



'After the completion of the later experiments a alight Irak was found in the "valve box" through which the 

 outgoinu- air current passed on its way to and from the "freezers," and by which water, condensed on the outside, 

 may have entered. There is every reason to believe that the quantity of water actually found was thus made too 

 large by a fraction of 1 per cent. In the average of the first nine experiments the amount of water found was 100.6 

 per cent of that required. As an alcohol check test was generally made between each two metabolism experiments 

 or series of experiments we have a means < if km iwing when the leak began to effect the results and the amount of the 

 error introduced. See Bulletin 109 of the Office of Experiment Stations, above referred to. 



The results thus indicate that the respiration calorimeter is an instrument of precision and 

 that the determinations of carbon dioxide, water, and heat produced within the chamber of the 

 respiration calorimeter are sufficiently accurate for experiments with the living subject. 



THE EXPERIMENTS. 



GENERAL PLAN. 



For the subjects of the experiments men were selected who were in good health, had appar- 

 ently normal digestion, and did not find the confinement in the chamber uncomfortable. A diet 

 was chosen which provided materials as palatable and in as much variety as was consistent with 

 convenient preparation, and with accurate sampling and analysis. The quantity and composi- 

 tion of the diet were generally such as to maintain the body nearly in nitrogen and carbon 

 equilibrium under the conditions of the experiment, whether cf work or of rest. In 13 of the 

 experiments the diet included alcohol. 



The alcohol amounted in general to about 72 grams (2£ ounces) a day, or as much as would be 

 contained in a bottle of claret or 3 or 4 glasses of whisky. In most cases pure (ethyl) alcohol, but 

 in some whisky or brandy was used. It was mixed with either water or coffee, and was given 

 in 6 small doses, 3 with meals and the rest at regular intervals between, in order to avoid as far 

 as possible any effect upon the nerves. The alcohol supplied not far from 500 calories of energy. 

 In the experiments without external muscular work, the total energy of the diet was about 2,500 

 calories, so that the alcohol furnished one-fifth of the total energy. In the experiments in which 

 the man was engaged in more or less active muscular work, the total energy of the food was 

 larger, averaging about 3,900 calories, so that the alcohol furnished between one-seventh and 

 one-eighth of the total energy of the diet. 



In order that the subject might become accustomed to the diet and reach approximate 

 nitrogen equilibrium with it before the experiment proper began, a preliminary digestion 

 experiment of at least 3 days immediately preceded the metabolism experiment. Any change 

 of diet found desirable or necessary was made during this period, and the preliminary experi- 

 ment was continued until nitrogen equilibrium was supposed to be more or less nearly reached. 

 In most cases the preliminary experiment continued i days. During this period the subject was, 

 in general, engaged in his customary occupation, but conformed his muscular activity more or less 

 to that of the coming experiment. Thus if it was to be a work experiment, he rode a bicycle or 

 walked a considerable distance each day. If it was to be a rest experiment, he avoided all 

 unnecessai'y exercise. For supper on the last daj r of this preliminary digestion experiment 



