I. In normal animals the injection of caffeine slightly diminishes the absorption of fluid from the peritoneal cavity, in spite of the fact that the amount of fluid and sodium chloride eliminated through the kidneys is markedly increased. The lessened absorption of fluid is due to a slight lowering of the osmotic pressure of the blood.
II. In nephrectomized animals caffeine increases the absorption of fluid from the peritoneal cavity; the increase in absorption is greater in nephrectomized animals which received caffeine than in nephrectomized animals which did not receive this substance, and it is due to additive increase in the osmotic pressure of the blood. In a similar manner, caffeine increases the absorption of fluid from the peritoneal cavity in animals in which, instead of nephrectomy, other operations, not directly affecting the kidneys, had been performed. In this case also the increase in absorption is presumably preceded by and due to an increase in the osmotic pressure of the blood.
III. In animals injected with uranium nitrate three days previously, caffeine diminishes the absorption of fluid from the peritoneal cavity, notwithstanding the high osmotic pressure of the blood which we observe in such animals. This agrees with the results of our previous experiments in which we found that in animals injected with uranium nitrate the absorption of fluid is not increased in spite of the rise of the osmotic pressure of the blood.
IV. At the time of the conclusion of the absorption experiments, the amount of fluid retained in the vessels was found to be diminished in each series in which caffeine was used. Only in certain cases can this be due to the increased amount of fluid leaving the blood vessels through the kidneys; in other cases it can only be due to a movement of water from the blood vessels into the tissues caused by the injection of caffeine.
V. In normal animals, in nephrectomized animals and in animals in which an operation not directly affecting the kidneys had been performed, caffeine causes an absolute and relative increase in the elimination of sodium chloride from the peritoneal fluid, as a result of which the remaining peritoneal fluid shows a lessened content of sodium chloride. Caffeine causes also a decrease in the sodium chloride content of the blood. We see, therefore, that under the influence of caffeine a greater amount of sodium chloride is eliminated from the body fluids into the tissues or through the kidneys. The factors which cause the sodium chloride to leave the body fluids are probably primarily responsible for the diuresis which takes place after administration of caffeine. In the case of caffeine and other similar substances the diuresis is, therefore, in all probability not due primarily to a specific action of the kidney, but to conditions which affect the distribution of sodium chloride in the body.
VI. The distribution coefficient of other osmotically active substances differs from that of sodium chloride. These other substances have a tendency to move into the body fluids in increased quantities under the influence of caffeine.
VII. Summarizing all experiments in which we studied the absorption from the peritoneal cavity, we may state that changes in the osmotic pressure of the blood represent the principal factor in explaining the variations in the rate of absorption of fluid from the peritoneal cavity.
VIII. There exists no direct relation between an increase in the rate of absorption of fluid from the peritoneal cavity and an increase in the amount of urine secreted. If it should be found that even at a period following the injection of caffeine, later than that at which we have studied the absorption, a rise of the osmotic pressure of the blood does not appear, then we may state that the diminution in the amount of edema in the body cavities resulting from the administration of caffeine is entirely due to an inhibition of the production of edema and not to an increased absorption of fluid from the serous body cavities.