In the light of present knowledge, it would appear as a subject for severe criticism that an investigation on the influence of the ingestion of food upon metabolism which continued for a decade should show such relatively slight positive evidence contributing towards an explanation of the various phenomena observed. It was hoped that, as the research developed, definite information as to the cause or causes of the increase in the metabolism would be accumulated. Thus, in the earlier part of our research, impressed by the strength of the argument presented by Zuntz and his associates upon the influence of roughage or crude fiber in the diet, we included experiments with popcorn in our study on the influence of pure carbohydrates, on the supposition that the starch of the popcorn and the crude fiber of the hull would give roughage. As the research continued, however, it was found impossible to plan experiments, save under special conditions, for studying the cause of the increased heat production following the ingestion of food. Consequently our data represent for the most part only faithful records of a large number of experiments in which foodstuffs were given, either singly or combined, and the energy transformations subsequently measured. A careful search in our data for conclusive evidence as to the cause of this rise in the metabolism is, however, unsuccessful.

At the present time three explanations are offered of the increases noted with the ingestion of food. Zuntz and his associates, influenced largely by their extended experience with domestic animals, particularly with ruminants which consume considerable roughage and bulky food materials that remain for a long time in the intestine and require considerable digestive activity expressible in forms of muscular activity, maintained that the increase was due to the work of digestion, or Verdauungsarbeit. Rubner, as a result of his critical series of experiments on dogs, particularly the experiments with protein, was not inclined to attribute any share of the increase to the work of digestion, but explained the increase upon the theory that each foodstuff exhibited a specific dynamic action, believing that the elaboration of food materials preparatory to absorption and oxidation, particularly the cleavage and elaboration of the protein molecule, accounted for the excess heat production. More recently the hypothesis of Friedrich Mueller1 has been revived,2 in that it has been maintained that the increase in the heat production is due to a stimulus to the cells as the result of products obtained from the food materials ingested or elaborated from them. That these products are in all probability of an acid nature is evidenced by experiments from this laboratory; the influence of amino-acids has been definitely proved by Lusk.3

1Attention should here be called to a recent study on the basal metabolism of dwarfs and legless men (Aub and E. F. Du Bois, Arch. Intern. Med., 1917,19, p. 864), in which the authors say that "following the ingestion of large quantities of meat, the excretion of urinary nitrogen during the earlier hours is not an accurate index of the protein metabolism. The sulphur excretion is more rapid than the nitrogen excretion".

Although practically none of our experiments were ideally planned to determine definitely the cause of this increase, certain phases of the work should be considered as an attempt to find if the phenomena agree with any of these explanations. Our experimental plan included, first, the establishment of a base-fine, and second, a post-absorptive condition for the subject in each experiment, i. e., that the subject should have been without food for at least 12 hours. It was assumed that comparison with such a base-line would give a true measure of the increase in metabolism due to food. The various factors affecting the basal metabolism have been considered in detail elsewhere4 and likewise in our chapter on basal metabolism. (See page 47.) It is of interest to point out here, however, that even after the active digestion of food has ceased, Gigon concludes that there is considerable internal work which is characterized by Zuntz as Verdau-ungsarbeit. Indeed, Gigon ingeniously ascribes a depression found by him in the metabolism following the ingestion of 50 grams of olive oil as being due to the fact that the presence of the oil caused an abatement of the Verdauungsarbeit which had persisted during the experimental period. Furthermore, X-ray studies have definitely proved5 that even during relatively prolonged fasting the motility of the stomach and the intestines does not entirely cease; this was likewise found by Boldireff.6 In discussing the influence of the ingestion of food, it is especially necessary to bear in mind this activity of the digestive organs during the absence of food, including the movements of the alimentary tract, the secretion of the various digestive juices, and similar movements, for the ingestion of food may be supposed to increase the activity of all these factors.

1Mueller, Volkmann's Sammlung klin. Vortrage, May, 1900 (N. F. No. 272), p. 17.

2Benedict, Trans. 15th Int. Cong. Hygiene and Demography, 1913, 2 (2), p. 394.

3Lusk, Journ. Biol. Chem., 1915, 20, p. 555.

4Benedict, Emmes, Roth, and Smith, Journ. Biol. Chem., 1914, 18, p. 139; Benedict and Roth, Journ. Biol. Chem., 1915, 20, p. 231; Benedict and Smith, Journ. Biol. Chem., 1915, 20, p. 243; Benedict and Emmes, Journ. Biol. Chem., 1915, 20, p. 253; Benedict, Journ. Biol. Chem., 1915, 20, p. 263.

5Cannon, The mechanical factors of digestion, 1911.

6Boldireff, Arch. d. Sci. Biol., 1905, 11, p. 1.

Any evidence bearing upon the possibility of intensive peristalsis or digestive action, or any series of experiments in which such stimulating agencies may be present, are of special interest in considering the cause for increased metabolism following food. Thus it is conceivable that the starchy foods would be more slowly acted upon than sugars, and yet an examination of the results of our experiments with such foods shows that they produced nearly as great an increment as the sugars did. (See tables 123,124,249, and 250, pages 196, 199, 336, and 338.) This is indeed surprising and might logically be taken as evidence in favor of the Verdauungsarbeit theory. While the dry starch of the popcorn could reasonably be considered as requiring a large amount of digestive work, it is hardly possible that bananas would contain material sufficiently irritating to the intestinal canal to have a great effect upon peristalsis or segmentation. At least two series of experiments carried out in this laboratory indicate that intestinal activity, as exemplified by the action of smooth muscle, does not measurably affect the metabolism. In one series the effect of purgatives and agar-agar was studied,1 and in the second a study was made of the metabolism of dogs with ablated pancreas and consequently deficient digestibility.2

In view of the results obtained in these two series of experiments, we find it unconvincing to explain any portion of the increase subsequent to the ingestion of food as being due to Verdauungsarbeit in the sense in which Zuntz uses the term.

All writers who discuss the cause of the increase in heat production following the ingestion of food are at once confronted by the problem of giving a concrete explanation of the term "specific dynamic action," first used by Rubner. Perhaps no worker has considered this subject more in detail than Lusk, who has written one of the best expositions of Rubner's views that has ever been published.3 Lusk proposes to compare the increase in heat production with the increased protein katabolized as a measure of the so-called "specific dynamic action," a process which is radically different from that originally employed by Rubner.4

Great difficulty is immediately experienced when we attempt to consider our experimental evidence in accordance with the prevailing views as to the cause of the increased heat production following food. Our experience with diabetics and with normal persons with a normally induced acidosis on a carbohydrate-free diet, as well as our experiments with unoxidizable material in the intestinal tract, lead us to favor more strongly the theory of acid-body stimuli, but it would be clearly a misuse of this present series of experiments to attempt to use them as experimental evidence for any of the three current theories. It is of significance that popcorn and bananas, with their large content of fiber material, increase the metabolism, a fact which tends to support the Verdauungsarbeit theory. The well-known increases in peristalsis subsequent to the ingestion of pure sugars, especially levulose, would also probably be considered by the advocates of the Verdauungsarbeit theory as sufficient explanation of the increment noted with sugars. On the other hand, the results of the two studies previously referred to, in one of which excessive peristalsis was induced by the administration of Glauber salts and agar-agar to man, and in the other a study was made of the metabolism of dogs having defective assimilation due to ablated pancreas, strongly disprove the Verdauungsarbeit theory.

1Benedict and Emmes, Am. Journ. Physiol., 1912, 30, p. 197. 2Benedict and Pratt, Journ. Biol. Chem., 1913, 15, p. 1. 3Lusk, Science of Nutrition, 3d ed., 1917, p. 232, et seq. 4Williams, Riche, and Lusk, Journ. Biol. Chem., 1912, 12, p. 349.