In cases of perverted metabolism, an abnormal accumulation of acids can occur in two ways, namely, first, the excretion of acids may be reduced, or, second, the formation of acids may be increased, while at the same time their elimination is interfered with. Aside from those cases of pulmonary and cardio-vascular disease in which the elimination of carbonic acid through the lungs is reduced and aside from acute attacks of gout in which the excretion of uric acid is reduced, the first named possibility is of very small practical significance. At least, interference with the elimination of acid products alone never produces the clinical picture of acid autointoxication. Insufficient elimination, however, may occasionally be combined with over-production of acid, and in this way the danger of intoxication be increased.

The second possibility is pathologically much more important. The acids that must be considered in this connection are, among others, sarcolactic acid, carbaminic acid, aliphatic acids, oxalic acid, uric acid, aromatic oxy-acids, and above all, the acids that interest us here in particular, namely, B-oxybutyric acid, diacetic acid and acetone.

There can be no doubt that these three substances are closely related to one another. The three have been grouped by Gelmuyden and designated by the collective name of "acetone bodies." Until very recently, there was considerable diversity of opinion in regard to the true connection existing between these bodies. Von Jacksch assumed that the appearance of acetone in the organism produced symptoms of intox ication. If the tissue fluids contained too much acetone, the latter would combine with certain acids that are formed from the disassimilation of proteids, and in this way produce diacetic acid. In the same manner von Jacksch imagines the genesis of acetone from B-oxybutyric acid. The discoveries of the last few years, however, have taught us to abandon this conception, partly on account of certain chemical considerations, partly on account of certain experimental and clinical findings.

It can readily be shown that outside of the body, in the test tube, the transition of these three substances into one another occurs in a way that is exactly opposite to the teachings of the older chemists. Oxybutyric acid is an oxy acid with a secondary alcohol group. Like all acids of this character, it is readily converted into its keto acid, viz. diacetic acid, by oxidation. The latter can be considered as acetone in which one hydrogen atom is substituted by the carboxyl group. If diacetic acid is warmed or boiled, acetone and carbon dioxide are formed. The following formulas will illustrate my meaning:

B-Oxybutyric acid: CH3 - CHOH - CH2 - COOH.

Diacetic acid: CH - CO - CH2 - COOH.

Acetone: CH3 - CO - CH2.

The same process that occurs in vitro occurs in vivo if B-oxybutyric acid is administered to animals or human subjects. In healthy human subjects and in healthy animals B-oxybutyric acid is completely destroyed and the only evidence of its passage through the organism is a slight increase of the urinary acetone. In diabetic dogs, after extirpation of the pancreas, and in diabetic human subjects, diacetic acid appears in the urine in addition to a considerable quantity of acetone. Some authors even claim that the acid itself has been known to pass into the urine. The reverse process, namely, the appearance of B-oxybutyric acid or diacetic acid in the urine after the administration of acetone has never been observed, the only urinary change being an increase of the acetone. The fact that the introduction of B-oxybutyric acid is followed by different consequences in healthy and in diabetic animals; the fact, furthermore, that the subcutaneous injection of sodium oxybutyrate in animals which had been poisoned by carbon monoxyde (Araki) does not lead to the elimination of more CO2 but to the elimination in the urine of an unchanged portion of the injected acid and some diacetic acid and acetone, demonstrates that all three substances owe their origin to some common perversion of oxidation and that this perversion is due to different causes in the two cases quoted above, viz., carbon monoxyd poisoning in the one case (dog) and pancreatic diabetes in the other. In the former case the perversion of oxidation is due to the lack of oxygen; in experimental diabetes, on the other hand, and in diabetes in human subjects there is no lack of oxygen, for diabetic subjects as well as depancreatized dog with diabetes consume all the oxygen that is necessary to promote the combustion of the ingested pabulum. We cannot, therefore, in these cases, be dealing with a lack of oxygen but merely with some local interference with oxidation.