As the coffee was taken hot in both of the calorimeter experiments, the same difficulties exist in making the proper correction for the changes in body-temperature that were indicated in the water-drinking experiments. With the subject A. H. M., in the calorimeter experiment of April 19, 1907, a positive increase in the heat production amounting to 11 per cent was noted. (See table 81.) Increments as great, if not greater, were shown in the carbon-dioxide production and oxygen consumption, thus confirming the fact that there was a true increase in the metabolism. In this case a relatively large amount of coffee, 1,011 grams, with 62 grams of sugar, was taken. That some of the increase in the metabolism may properly be ascribed to the sugar is clear from the data shown subsequently for the experiments in which the effect of ingesting cane sugar was studied.

In the experiment with A. W. W. on April 12, 1907, the amount of coffee taken was much smaller, only 271 grams; in addition, 23 grams of sugar were given. According to the data in table 80, there was an actual lowering of the metabolism (3 per cent), with practically no change in either the carbon-dioxide production or the oxygen consumption. To a certain extent, then, this experiment is similar to the experiments with hot water in which no material effect was observed on the metabolism. In fact, neither of these two experiments can be taken as giving positive evidence of an increase in the metabolism due to the ingestion of coffee.

The series of six respiration experiments with five subjects, all made in March 1911, gave more convincing results. (See tables 82 to 87.) In these experiments approximately 325 grams of coffee were taken with a temperature, so far as known, of 50° to 60° C. The increments averaged approximately 8 per cent, with a maximum of 13 per cent and a minimum of 2 per cent. These positive increments in the metabolism are distinctly at variance with the results of the calorimeter experiment with A.W.W., in which a slight decrease appeared. But the general trend is clear, and one may properly state that approximately 325 grams of coffee infusion at a temperature of about 60° C. will produce an increment in the metabolism of 8 to 9 per cent.

A careful analysis of the detailed data for these experiments shows that in practically all instances the increment was by no means at an end at the conclusion of the experiment; thus these figures probably represent low rather than high values. For example, in the experiment with J. J. C. on March 9,1911, a basal metabolism of 222 c.c. of oxygen consumed was recorded. Over 5 hours later, at the end of the experiment, the oxygen consumption was 268 c.c. A similar long-continued effect was noted with the same subject on March 21, 1911. It is thus clear that the ingestion of coffee produces a positive increment in the metabolism which must not be neglected in the interpretation of experiments in which it has been taken. It is conceivable that in the earlier experiments with diabetics reported from this laboratory by Benedict and Joslin the small amount of coffee taken by the subject one or two hours before the experiment may have been responsible for a part of the increase noted in the metabolism, although it was at that time specifically stated that the coffee could have no influence.1 The amount of coffee taken by the diabetics was, however, less than half of the amount given in these experiments, and it was usually taken some time prior to the beginning of the observations. Since November 1914, no coffee has been used by the diabetic subjects on the morning of the experiment.

An examination of the pulse-rate data obtained in the coffee experiments shows slight increases after the ingestion of coffee for nearly all of the experiments, with usually a subsequent rapid fall to its previous level. In one experiment, that with L. E. E., March 23, 1911, the rate was lower after the coffee was taken.

The systolic blood pressure was higher in most instances after the coffee was drunk. The maximum rise was about 20 mm. mercury in the experiment with J. J. C. on March 9, 1911.

Experiments made by Edsall and Means2 and Higgins and Means3 on the effect of caffein have an interest in this connection, as they show clearly the influence upon the metabolism of this constituent of coffee. Two experiments were made by Edsall and Means in the Massachusetts General Hospital, both of which indicated a definite although not very great rise in the metabolism. Those made by Higgins and Means, and published from this laboratory, show that with one of the subjects, J. H. M., the gaseous metabolism was markedly increased. With H. L. H. there was also an increase in the metabolism, although this was slight.

We may conclude, therefore, that coffee, owing probably to its caffein content, acts as a stimulus to the metabolism, the increment with 325 grams of coffee infusion amounting on an average to 8 per cent for several hours. Experiments with caffein-free coffee would therefore have special interest.

1Benedict and Joslin, Carnegie Inst. Wash. Pub. No. 136, 1910, p. 216.

2Edsall and Means, Arch. Intern. Med., 1914, 14, p. 897.

3Higgins and Means, Journ. Pharm. and Exp. Therapeutics, 1915, 7, p. 1.