Although the earlier experimenters made but few observations on the effect of drinking hot or cold water, we find a number of studies on the effect of taking tea and coffee. Of special interest is the series of experiments made by Bocker,1 who concludes that the taking of coffee decreases both extensively and intensively the respiratory processes. Edward Smith2 has a series of observations on drinking both tea and coffee, and concludes that tea is a powerful respiratory stimulus, coffee being but little less powerful. With the technique used by both these investigators, it was not possible to study the fine differences which in later times have been found to exist; hence their results can not be considered as conclusive.

Speck3 studied the effect of coffee-drinking in two experiments and found a small but visible rise in the carbon-dioxide production and oxygen consumption, indicating to his mind a distinct stimulus to the digestive activities. Lehmann and Rohrer4 found that the volatile constituents of tea and coffee did not cause any noticeable changes in the respiration frequency. A series of papers from the Russell Sage Institute of Pathology has just appeared which includes a paper by Means, Aub, and Du Bois,5 reporting the results of a study in which four normal subjects were given from 8 to 10 grains of caffein - i. e., 8.6 milligrams per kilogram of body-weight. The authors state that the basal metabolism was increased from 7.4 to 23.5 per cent, these values representing average "peak" effects. Of special significance is the fact that there was no material change in the pulse rate.

Statistics Of Experiments

Our own observations with coffee include two calorimeter experiments made in Middletown and six respiration experiments in Boston. The results are given in tables 80 to 87 and discussed in the accompanying text. They are also summarized in tables 67 and 68. (See pages 135 and 136).

In all of the experiments the coffee was taken hot; in the two calorimeter experiments a certain amount of sugar was also taken. The general plan of both series of experiments was similar to that of the water-drinking studies. The measurement of the gaseous metabolism in the respiration experiments began 6 to 32 minutes after the drinking of the coffee; the total time between the taking of coffee and the end of the last period ranged from 1 hour 31 minutes to 5 hours 29 minutes.

The method of determining the total increment in the respiration experiments was unlike that used in the chewing and water-drinking studies in that the increase was found here by measuring plotted areas superimposed upon the base-lines determined on the respective days. The average heat production (computed) per minute, plotted for the average time of the periods, supplied the points for defining the area of increment. Inasmuch as this method was followed for all of the respiration experiments except the water-drinking and chewing studies, it is described in detail here and two illustrative curves are given (figures 1 and 2).

1Bocker, Beitrage zur Heilkunde, 1849, 1, p. 200.

2Smith, Phil. Trans., 1859, 149, p. 715.

3Speck, Physiologic des menschlichen Athmens, 1892, p. 42.

4Lehmann and Rohrer, Arch. f. Hyg., 1902, 44, p. 203.

5Means, Aub, and Du Bois, Arch. Intern. Med., 1917, 19, p. 832.

The curve in figure 1 is that for the coffee experiment with L. E. E., March 23, 1911. (See table 84.) The basal value for this day was 1.15 calories; the time between the drinking of 325 grams of coffee and the end of the last period was 3 hours 50 minutes. The initial value for this curve is on the base-fine at the point indicated on the horizontal scale as 0, this being the time when the subject finished drinking the coffee - i. e., at 10h40m a. m. The point plotted 29 minutes later is for 1.28 calories at the average time of the first period, that is, at 10h39m a. m. Values have been similarly plotted at six other points - i. e., 1.21 calories at 58 minutes, 1.28 calories at 1 hour 26 minutes, 1.29 calories at 1 hour 58 minutes, 1.28 calories at 2 hours 29 minutes, 1.26 calories at 3 hours 4 minutes, and 1.28 calories at 3 hours 43 minutes after taking the coffee. The curve has been extended to reach a perpendicular dropped to the base-line at the point of time corresponding to the end of the last period of the experiment. The area thus inclosed by the curve and base-line is considered to represent the total increment for the period of observation following the drinking of the coffee. With a planimeter this area measured 4.25 units, and since each unit of area represents a value of 6 calories, the total increment (6 X 4.25) was therefore 26 calories. The basal value corresponding to the period of 3 hours 50 minutes or 230 minutes (230 X 1.15) would be 265 calories. The percentage increase in the metabolism (26./.265) was therefore 10 per cent.1

The planimeter method used in the coffee and beef-tea experiments for determining the increment in the heat output was also used in the respiration experiments with other food materials for computing the percentage obtained by a comparison of the increment in the heat output with the fuel value of the food material ingested - i. e., the "cost of digestion."2 To illustrate this method as employed in experiments with food materials having a high energy value and consequently a great effect on the metabolism, the curve for the beefsteak experiment with Dr. S. on June 30, 1911,1 is given in figure 2. In this experiment 177 grams of beefsteak were taken by the subject. The last experimental period was completed 6 hours 35 minutes after he had finished eating. The points in the curve were plotted and area of increment defined as for the curve in figure 1. The total area, as measured by the planimeter, was 9.32 units, corresponding to 56 calories. The fuel value of the beefsteak ingested was 298 calories; the increment (56 ./.298) was therefore 19 per cent of the fuel value. (See table 215, page 284.) Statistical data not included in the tables or in the discussion are given in the following paragraphs for all of the experiments. Whenever the basal values were determined immediately before the coffee-drinking, the times given include both basal and coffee periods.

Fig. 1. - Curve showing increment of heat production following ingestion of 325 c.c. of coffee in experiment with L. E. E., March 23, 1911.

1See table 68, p. 136.

2See discussion of these values, p. 335 et seq.

Fig. 2. - Curve showing increment of heat production following ingestion of 177 grams of beefsteak in experiment with Dr. S., June 30, 1911.