The increase m metabolism is, therefore, 2 per cent, above the true normal, but 11 per cent, above the normal controls selected by Benedict and Joslin. This selection may have been justified, for in order to choose individuals who were like the diabetic patients, emaciated controls were indicated and such show a subnormal basal metabolism. Allen and DuBois have pointed out that herein lies the fundamental cause of the divergency in viewpoint. These authors have published a summary of the carefully investigated cases of diabetes mellitus, 26 in all. They found that the basal metabolisms of one-half of these were within the normal range of ± 10 per cent.; 9 cases showed increases of 11 to 23 per cent, above the normal basal. Of these cases which manifested a higher metabolism, 3 had severe, 2 marked, 1 moderate, and 3 little or no acidosis. The patient who showed the greatest rise in metabolism had very slight acidosis. Four patients showed metabolisms of between 14 and 19 per cent, below the normal. Of the 17 patients whose basal metabolisms were normal or below the normal, 8 manifested very severe acidosis. The conclusion follows that acidosis cannot be the immediate cause of the increased metabolism when this is found in diabetes.

1 Allen and DuBois: Loc. cit.

In their analysis Allen and DuBois considered for the first time the factor of emaciation. They call attention to Magnus-Levy's1 description of a neurasthenic youth who had partially starved himself for a year or more. On entrance to the hospital he was "skin and bones." During the first experimental period he was given his former dietary containing 700 to 800 calories daily and was then given abundant food.

These valuable data indicate that if a man whose metabolism is normal has been reduced in weight by 30 per cent., his metabolism when he is fed with a low dietary may be reduced 44 per cent, below the actual normal level, or 33 per cent, below, as measured by the normal per square meter of surface. If a liberal diet be given at this juncture, the total heat production is 26 per cent, below the actual normal, or 17 per cent, below if recorded on the basis of surface area. This leads to the query whether basal metabolism of the emaciated diabetic is not really below that of a normal man.

Among the cases cited by Allen and DuBois there are 6 who show an emaciation of over 20 per cent., as follows:

* Nervous individual. 1 Magnus-Levy: "Zeitschrift fur klin. Med.," 1906, lx, 177.

It is evident that if the metabolism of a normal man, who through emaciation has lost 30 per cent, of his body weight, is 33 per cent, (or when well nourished 17 per cent.) below the normal level of metabolism, then the emaciated diabetic has in reality a higher heat production than he would have had if he had been free from diabetes. This is emphatically shown in the case of C. K., soon to be more fully described. This individual, who had diabetes in a severe form, subsequently became entirely free from sugar and manifested a high degree of tolerance for carbohydrate. The metabolism of the different periods may thus be summarized:

It is evident that in this emaciated individual the metabolism would have been lower in the first instance had he not been diabetic. The high protein metabolism would sufficiently account for the increased total heat production in this patient, although in some other instances of increased metabolism in diabetes this factor does not apparently always suffice to explain the increase. The frequent presence of lipemia (see p. 252) may explain in part the increased metabolism. The onset of diabetes in this case was very rapid. Nowhere in the literature is the protein metabolism in diabetes mentioned as being so high, and in no other case are the results of metabolism experiments so nearly akin to those obtained in experimental animals.

A preliminary report of the patient C. K. of Geyelin and DuBois1 gives details of his metabolism in a table which is reproduced on page 478.

1Geyelin and DuBois: "Journal of the American Medical Association," 1916, lxvi, 1532.

Clinical And Experimental Data In Case Of C. K

1 Average basal normal 39.7 calories per square meter per hour. 2 Transferred to Bellevue Hospital.

3 After meals which might cause increase in metabolism of 5 to 10 per cent, above basal.

4 Returned from Bellevue Hospital.

5 Transferred to Bellevue Hospital.