and while an active circulation coursed through the liver, only a bare trace of sugar might really be present in the organ, or even none at all, if its vascular apparatus and their contents. should be excluded from the abstract concept, or scheme, of true hepatic tissue, while nevertheless, as a whole, the liver should be actively engaged in the production of sugar. Not knowing how very short a time is required after death for the accumulation of a notable quantity of sugar in portions of excised liver, he has calculated his estimates altogether too high, by failing to dose his excised bits of liver soon enough after excision; neglecting to appreciate the retrospective sig nificance of observations he had himself made respecting a point on which Pavy afterward based a highly-plausible though unsound hypothesis. In consequence of this omission, Bernard's earlier and best known dosages of sugar in the liver do not indicate the quantity present during life, but relate almost entirely to the faculty the organ enjoys of generating sugar, after death, by a continuation of the chemical processes active within it while a part of the living body.

Indeed, it must be held that the transformations observed in the exsected liver, especially if the organ be maintained at the temperature proper for its functional activity during life, do not vary at all in chemical character from those normally occurring within that period which precedes the establishment of putrefaction; differences observed thus far have been conclusively shown to be of a purely quantitative kind.

Not accepting this doctrine, Pavy very unexpectedly assailed the glycogenetic theory, in a paper read before the Royal Society in June, 1858. He injected a strong solution of potassa into the liver immediately after death, with the proposed object of arresting the function upon which glycogenesis depended, but his method was altogether faulty. By such a procedure he not only washed out from the parenchyma of the organ all traces of sugar presumably present, but, if any escaped evection in this way, inadvertently effected its prompt conversion by a well-known chemical reaction (see Bernard's "Leçons," 1854-55, pp. 32, 33) into substances devoid of reducing power on the salts of copper. Thus unintentionally did the experimenter himself remove and destroy

the very substance in question, from whose absence he proposed to deduce a denial of the glycogenetic function. By a legitimate method, however, he determined an exceedingly valuable fact—namely, having suddenly sliced off a piece of the liver of the dog just killed, he threw it instantly upon a freezing mixture of ice and salt; when frozen, it was ground into a pulp and a decoction made of it; the absence of sugar was almost complete. This experiment, the first of its kind, must be regarded as a cardinal one on this subject. Pavy, therefore, claimed that, as usually encountered in the hepatic tissue and blood after death, sugar must be entirely of postmortem formation.

Although Pavy's injection-experiments were presently shown by Thudichum and Harley in England to be irrelevant, a close critical attention was at once directed to the point at issue, and observers were astonished to find glucose nearly or quite absent from bits of liver tested immediately after exsection, or after freezing, as Pavy had affirmed. Many physiologists repeated Pavy's experiments, and adopted his conclusions, so that, a variety of considerations still cogent being temporarily lost sight of, faith in Bernard's doctrine was profoundly shaken. The liver, during life, did not seem either to contain or to form sugar, notwithstanding all that had up to this time been so authoritatively taught.

It began to be suspected, after a time, that the difficulty might lie in the normal presence of a far smaller quantity of saccharine matter in the hepatic substance than Bernard, C. Schmidt, Lehmann, Poisseuille and Lefort, and others, had as yet imagined, and that during life this quantity might be so extremely small as to escape detection, at least by tests hitherto employed. Experiments, nevertheless, even under this preconception, by Meissner and Jäger entirely corroborated Pavy's affirmations. Meissner and Ritter sliced out portions of liver from living rabbits, which they immediately plunged into boiling water; they observed no sugar-reaction with Trommer's test. So also, in 1866, Schiff attained similar results. In 1868, by a more careful application of his tests, Eulenberg was able to establish the existence of traces of sugar in the watery decoctions of fresh healthy livers; but,

fearing that even the short interval elapsing between the excision and comminution of the bits of tissue and their elevation to 212° may have allowed a post-mortem generation of sugar, he ground the bits of liver-substance in a mortar with pounded glass and alcohol immediately after exsection; nevertheless, on testing extracts so obtained, there was no evidence whatsoever of the presence of sugar by the mode of testing he employed, which does not seem to have been applied with sufficient delicacy.

Thus, during a period of at least eight years, the glycogenetic theory remained more or less invalidated by the results of such authoritative, and, to all appearance, conclusive experimentation.

Our accomplished fellow-countryman, Flint, seems first to have advanced anew to meet the question, and to endeavor to harmonize the apparently irreconcilable views of physiologists. In 1868 (NEW YORK MEDICAL JOURNAL, January, 1869) he published a short paper on this subject, with the details of three experiments on dogs. The liver-tissue was excised and thrown into boiling water, and the watery decoction made by further boiling was tested. When the time between excision and immersion was twenty-eight seconds, the presence of sugar by Trommer's test could not be affirmed. In Experiment II, the time was ten seconds; Trommer's test was again applied, but there was still no evidence of the presence of sugar. In Experiment III. the time was again ten seconds; after boiling for seventeen minutes, Fehling's liquid, the most delicate and reliable of all sugar-tests, still failed entirely to show the presence of sugar in the liver decoction, though revealing unmistakably the existence of a small quantity of that substance in a sample of the blood of the hepatic veins, obtained directly after excision of the bit of liver.

In neither of these experiments was Flint able to demonstrate the presence of sugar in the hepatic tissue; its existence in the blood of the hepatic veins, however, it must be recollected, was not in question, and had never been denied. Flint seems to have collected the blood more rapidly than it had ever been done before, not more than one minute being consumed in Experiment III. in the application of the ligatures.

Few operators have been rapid enough in their manipulation to execute the required procedures within anything like so short a time; usually from five to ten minutes are spent in the application of the abdominal and thoracic ligatures; speed, however, was essentially a condition of Flint's hypothesis. With a properly-prepared Fehling's liquid, no competent observer has ever failed, in a healthy animal, to detect sugar in the blood of the hepatic veins; its quantity is usually from one-half of one per cent. to one per cent. of the dry residue of such blood. I have often had occasion, while experimenting or lecturing on this subject, to observe the uniform presence of sugar in this locality, but I can remember one or two instances where I was unable to detect more than the feeblest traces of sugar in bits of liver simultaneously cut off and made into a decoction. This was especially so in an experiment in 1856, where a dog was rapidly opened, a bit of liver cut off quickly (with a haste, however, due to other considerations), and pulped, while the liver was torn out and thrown into a retort. When tested for sugar by Fehling's test, the bit of liver showed an exceedingly small quantity of that substance; and, as the animal was in perfect health, I was quite at a loss for an explanation of the matter, as this was two years before Pavy's experiments were reported.

Although his results did not strictly warrant his conclusions, Flint decided in favor of a normal hepatic glycogenesis. "During life," he says, "the liver contains only glycogenic matter, and no sugar, because the blood washes out the sugar as fast as it is formed; but after death, or interference with the circulation, sugar is not thus removed, and can be detected in the substance of the liver."

Following Flint, Lusk, of New York, in a very able critique of this subject in a paper on the "Origin of Diabetes" (see NEW YORK MEDICAL JOURNAL, July, 1870), detailed some careful and highly-satisfactory experiments on dogs, in which from a quarter to half a grain of sugar to the ounce of blood was found in the blood of the right ventricle of the heart, and by ocular admeasurement about one-fourth as much in the blood of the jugular vein; the method followed was catheterization of the ventricle as practised by Bernard. In his affirmation

respecting the presence of glucose in the systemic blood, Lusk substantiated previous determinations of Coze and Pavy. In experiments of this kind, Pavy had expressed surprise at the "exceedingly minute" quantity of sugar present in such blood, omitting to take into account the velocity of the blood-current. He thought more sugar was found in blood drawn from the ventricle through a fine incision.

The presence of sugar in the hepatic veins, and in the ascending cava between the orifices of the hepatic veins and the heart, and in the right ventricle also, is thus seen to have saved the glycogenic theory from annihilation; the existence of sugar in these localities persisting as a fact which could not be explained without recurrence to hepatic action as the source of the sugar, when the indisputable absence of this substance in the abdominal vena cava below the débouchement of the hepatic veins, and in the blood of the portal system, was taken into consideration.

In 1871 Dalton made the most careful and decisive experiments on the subject which had yet been instituted. He crushed his bits of exsected liver in a special appliance, working far more rapidly than the pestle and mortar, and threw them at once into alcohol or boiling water. Fehling's test was used with the greatest circumspection. The time consumed in exsecting the bits of liver, in twenty experiments, varied from three to thirteen seconds, averaging 6.2 seconds. In every case the "final watery solution gave a decided and perfectly unmistakable sugar-reaction," amply sufficient for the employment of volumetric methods of determination. Dalton found 2.4872 parts of sugar in 1000 of liver-tissue as a mean of ten experiments (maximum 4.3750, minimum 0.8040). In his tabulated results, I can trace no relation between the time consumed in extirpation or the period elapsing after the animals were fed and the quantity of sugar found. Dalton's conclusions were as follows:

1. Sugar is to be found in the liver at the earliest period at which it is possible to examine the organ after extirpation. 2. The average quantity at this time is at least 2.5 parts per 1000.

3. The sugar thus found does not belong to the arterial