The increase in the respiratory quotient subsequent to the ingestion of carbohydrate is in practically all instances due to a pronounced rise in the carbon-dioxide production rather than to a decrease in the oxygen consumption; the latter is also increased in the majority of instances. This increase in the carbon-dioxide production, which is the only factor measured in Johansson's experiments, certain experiments of Gigon, and a large number of Rubner's, does not indicate accurately the effect upon the metabolism itself as measured either directly in calorimeter experiments or by indirect calorimetry when both the carbon-dioxide and oxygen determinations are made. The increase in the carbon-dioxide production observed after carbohydrate ingestion may have three explanations:

As can be inferred from the average basal respiratory quotient, the katabolism during the post-absorptive period is a protein-fat-carbo-hydrate katabolism. When carbohydrate has been ingested, fat may be completely excluded from the katabolism, and we then have a protein-carbohydrate katabolism; under these conditions the proportion of carbon dioxide produced will be larger than that when fat is used in the production of a like amount of energy. Hence one explanation of the increase in carbon-dioxide production may be that it is due simply to a replacement of fat by carbohydrate in the metabolism.

Second, the increment in carbon-dioxide may be derived in appreciable amounts from a cleavage of carbon-dioxide from carbohydrate in the formation of fat. The formation of fat as a result of excessive carbohydrate feeding is no longer in question, for the experiments of Meissl1 and of Bleibtreu2 on swine and geese, to say nothing of the many experiments with man and other animals than swine and geese, have shown this conclusively. With the ingestion of 100 grams of pure carbohydrate, there is immediately made available 380 to 400 calories, while the basal requirement may not exceed 70 to 90 calories per hour. The sugar ingested would therefore logically suffice for the basal requirement of a period of 5 to 6 hours, during which time we may properly say the conditions are those of excessive carbohydrate feeding. Just what is meant by excessive carbohydrate feeding is, of course, in large part dependent upon the period between the feedings and the total amount ingested, but logically there is no reason why the above argument is not sound. It is fair to assume, therefore, that part of the carbon-dioxide may be derived from the cleavage of carbohydrate to form fat.

1Meissl. Zeitschr. f. Biol., 1886, 22, p. 63. 2Bleibtreu, Arch. f. d. ges. Physiol., 1901, 85, p. 345.

Third, an increased carbon-dioxide excretion may result from an actual increase in metabolism, during which process additional carbonaceous material is burned, with a resultant increase in the production of carbon dioxide. This fact is of prime importance, since the measurement of oxygen consumption, and particularly of heat production, will likewise indicate such an increase in metabolism.

If only the carbon-dioxide excretion is measured, it is impossible even to estimate the varying amounts due to each one of these three factors. On the other hand, when the oxygen consumption or heat production is determined, we have definite information as to the probable amount of excess carbon dioxide due to an increase in metabolism. The fact has already been clearly established by direct calorimetry, and is substantiated by indirect calorimetry, that the ingestion of carbohydrate in these experiments actually results in an increased heat production, that is, an increased metabolism entirely aside from intermediary transformations. Indeed, the heat production, as indicated in the data for the calorimeter experiments, is of such a magnitude as completely to preclude the assumption that the extra heat produced is due solely to hydration or simple cleavage. We may therefore properly consider that the ingestion of carbohydrate, particularly of sucrose and levulose, results in a direct stimulus to the total metabolism in the body.

The marked rise in the respiratory quotient also leads to the firm conviction that the fat combustion must, in large part, have been replaced by carbohydrate metabolism, at least in those experiments in which the respiratory quotient closely approaches unity. With respiratory quotients of 0.97, which by correction would result in non-respiratory quotients of unity, we may likewise assume that the non-protein metabolism is due to carbohydrate, an assumption which seems legitimate in view of the fact that quotients of this character frequently appear in our series. On the other hand, quotients considerably above unity also frequently appear, especially in the levulose and sucrose experiments. These distinctly imply, if not absolutely prove, the formation of fat from carbohydrate. It still remains a question as to whether this latter process, namely, the formation of fat from carbohydrate, may not proceed even when there is a somewhat lower respiratory quotient than that of unity. With the marked differences in the action of the several sugars on the total metabolism, and particularly on the respiratory quotient, it is conceivable that with the ingestion of sucrose or levulose the fat metabolism may not be completely suppressed and that we may have a very considerable formation of fat from carbohydrate with a slight fat combustion still progressing. The actual proof of this is, however, beyond the possibilities of existing technique.

The intermediary processes must be still further considered and the fact recognized that when the body is surcharged with carbohydrate, as it is after the ingestion of 100 grams of sugar, there may result a considerable deposit of glycogen. This process would be without action upon the respiratory quotient and one might suppose it to be without action upon the total metabolism. It is nevertheless a fact that in the experiments with an initial respiratory quotient so low as to suggest a glyeogen-poor reserve there was no evidence of a sufficient storage in the body of the ingested carbohydrate to produce a marked decrease in its effect upon the total metabolism.

One of the most striking illustrations of this fact was in the experiment with H. H. A. on January 2, 1912, in which 100 grams of sucrose were given (see table 176). The store of glycogen in the body of this subject was presumably very low, as evidenced by the basal respiratory quotient of 0.72. This was due to the fact that a few days previous he had been the subject of a series of experiments with a carbohydrate-free diet. If this subject had first replenished his glycogen store with carbohydrate before the ingested material was katabolized or before any portion of it was converted to fat, we should not expect an immediate increment of either the total metabolism or the respiratory quotient. As a matter of fact, the ingestion of 100 grams of sucrose in this particular case resulted in the maximum increment for the entire series with sucrose in both the heat and carbon-dioxide production and very nearly the maximum rise in the oxygen consumption. We have already observed (table 181) that inside of 40 minutes the quotient rose from 0.72 to 1.05 and remained at a rather high value for at least two subsequent observations, the quotients being 0.98 and 0.93. Still another illustration of this lack of evidence as to glycogen storage is supplied by the levulose experiment with L. E. E. on May 22, 1911 (see table 180). Although the post-absorptive quotient of 0.77 was the lowest obtained in this series, the ingestion of 100 grams levulose produced very nearly the largest excess carbon dioxide, namely, 23.2 grams, 3.1 grams excess oxygen, and 21 calories of excess heat (see table 175).

It is a source of regret that the series of experiments with carbohydrates did not include a larger number with both glycogen-poor and glycogen-rich subjects. As has been shown by previous tests in this laboratory,1 it is perfectly feasible to obtain a glycogen-poor condition by one or two days of carbohydrate-free diet. Durig, with his keen foresight, recognized the significance of this question and carried out one experiment in which an attempt was made to have the subject in a glycogen-poor condition, but the initial respiratory quotient of this subject, 0.799,1 did not indicate a much lower glycogen storage than that of his two previous experiments with quotients of 0.835 and 0.809, respectively.

1Hieeins, Peabody, and Fitz, Journ. Med. Research, 1916, 34, p. 263.

To sum up, the experiments upon the ingestion of carbohydrate show clearly that carbon-dioxide measurements have little significance if unaccompanied by measurements of either the oxygen consumption or the heat production. The increment in the carbon-dioxide production invariably noted may be caused by three different factors, all of which may be working together, but an actual increase of the heat production can only be shown by oxygen measurements or by direct calorimetric measurements. In considering the three causes for the increment in carbon dioxide, i. e., the replacement of fat by carbohydrate in the metabolism, the formation of fat from carbohydrate, or an actual increase in the total katabolism (all of which involve a destruction of carbohydrate) the disappearance of carbohydrate after ingestion due to possible glycogen storage in the body should not be lost sight of. Presumably this latter condition will be best favored by a depletion of the glycogen store in the body previous to the ingestion of the carbohydrate.

Since the protein katabolism in these experiments plays such a relatively small role, rarely over 15 per cent of the total heat production being derived from protein, we can practically neglect the intermediary transformations of protein in our calculations. Sufficient evidence has, however, been accumulated to show that the contention of Gigon, i. e., that in the nuchtern or post-absorptive condition there is constancy in both the nitrogen content of the urine and the character of the katabolism as indicated by the respiratory quotient, can not hold true. The data in this publication make clear the fact that respiratory quotients of the same individual may vary greatly. The careful series of experiments published by Togel, Brezina, and Durig,2 also show that with the same subject the basal post-absorptive respiratory quotient varied inside of a period of less than 4 months from 0.799 to 0.903. While, therefore, we are fully cognizant of the extremely suggestive and stimulating discussion by Gigon of the constancy of the basal cellular metabolism, established, as he thought, by his determinations of the nitrogen, carbon-dioxide excretion, and basal metabolism, yet we firmly believe that subsequent data can not confirm his assertion (see footnote 2, page 264). In our series of experiments it is wholly impossible to conceive of a constant nitrogen metabolism with a constant fat metabolism on which the carbohydrate metabolism is simply superimposed.

1Togel, Brezina, and Durig, Biochem. Zeitschr., 1913. 50, p. 311. 2Ibid., p. 296.