Page:The Harveian oration delivered at the Royal College of Physicians June 26, 1889 (IA b22361285).pdf/29

 effusions which we cannot attribute to obstructed circulation, such as the effusions in Bright’s disease.

hen we remember how urinary salts and urea (which may be regarded as the crystalloids of the urine) foul the blood in albuminuria, causing a change in the qualities of the liquor sanguinis, by the introduction of an excess of substances capable of permeating membrane with facility, we at once see how a knowledge of Graham’s law may possibly assist us in explaining the dropsical state.

The loss of the extractive matters of the blood may also be cited as tending to alter the relation of the liquor sanguinis to the membranes with which it comes in contact. In Asiatic cholera, when the disease is characterised by the usual flow of serous fluid from the intestinal surface, conditions are present suggesting the probability that retention of crystalloids in the circulation may have much to do with bringing about the result; for, if the secretions generally, and especially the secretion of urine, be suddenly stopped, the blood will become rich in crystalloids—a condition which we know greatly favours the diffusion of fluids through membrane.