The effect of water-drinking was studied in this research in five calorimeter experiments and six respiration experiments. The experiments with the respiration calorimeter at Middletown, Connecticut, consisted of two 8-hour observations in which large amounts of water, 1,584 grams and 3,935 grams, respectively, were taken by the two subjects. (See tables 69 and 70.) With the chair calorimeter in Boston three experiments were made in January 1911, in which approximately 1,800 grams were taken by each subject. (See tables 71 to 73.) In the shorter observations with the respiration apparatus six subjects were studied who were given from 200 to 500 grams of water, the temperature of the water in all but one case being somewhat over 50° C. (See tables 74 to 79.) The increments in the heat production are summarized in tables 67 and 68. (See pages 135 and 136).

1Benedict and Emmes, Am. Journ. Physiol., 1912, 30, p. 197.

2Benedict and Slack, Carnegie Inst. Wash. Pub. No. 155, 1911, p. 73.

3Rancken, Skand. Arch. f. Physiol., 1908, 21, p. 161.

4Stengel and Hopkins, Am. Journ. Med. Sci, 1917, 153, p. 101.

5Lusk, Journ. Biol. Chem., 1915, 20, p. 555; see especially pp. 558, 576, and 615.

The method used for computing the increment in the water experiments with the respiration apparatus was that employed in the experiments with chewing. (See page 127.) The basal value for the day was obtained in a series of 15-minute observations when the subject was without food or water. The second series began 5 to 34 minutes after the subject had taken water; the intervals between the periods were from 14 to 29 minutes. The average heat production for the basal periods and the heat output for each of the periods with water are given in tables 74 to 79. The increment in the metabolism was obtained by comparing the average value for the basal periods with the heat production for each period after the ingestion of water. It is seen by an inspection of the tabulated data for this group of experiments that these small and in some cases negative increments in the heat production represent the results of two to four periods of measurement. The sum total of time for these measurements, which are obviously not continuous, is 30 to 60 minutes, extending over periods of 50 minutes to 2 hours and 43 minutes following the drinking of water. In calculating the total increments it was assumed that the rate of increase for the time between the drinking of the water and the end of the last period was the same as that observed in the measured periods.

For example, in the experiment with J. J. C, February 7, 1911 (see page 146), the basal value for the heat production is 1.10 calories per minute and the values for the respective periods after the water was taken are 1.10, 1.14, and 1.11 calories per minute. It is thus seen that the average increase per period and per minute is equivalent to 0.02 calorie. The time between the drinking of the water and the end of the last period (llh05m a. m. to 12h31m p. m.) was 86 minutes. The total increment computed as heat (86 X 0.02) was thus 2 calories. The basal value for a corresponding length of time (86 X 1.10) was 95 calories; the percentage increase in metabolism (2 ./.95) was, therefore, 2 per cent.

Statistical data not included in the tables or in the discussion are given in the following paragraphs for all of the experiments. The times given include both basal and water-drinking periods, when the basal values were determined immediately before the water-drinking.