There is, perhaps, no other organ in the body the problems with regard to which seem to be so nearly plain questions of hydrostatics. It is easy to make a model of a urinary tubule and its blood-supply. If such a model were shown to a sanitary engineer, and he were asked to explain the working of the drainage system of the body, and especially to answer the two questions which we have propounded, he would say that there could be no doubt as to the part of it through which most water enters the tube, the glomerulus. He could give no opinion as to whether urea, uric acid, and other substances of a like nature, accompany the water until he had tried the experiment of separating blood from water containing the inorganic salts of urine by a permeable membrane—the blood being at such a pressure as the physiologist told him he might expect it to have in renal arterioles, the water at such a pressure as he might expect it to have at the upper end of a urinary tubule. He would find that urea, and still more uric acid, is very reluctant to pass through the membrane. Again, when asked whether water, in which urea and other things were dissolved, would leave the tubule—say from the loop of Henle—to pass back into the blood, he would repeat his experiment with a membrane. This time he would allow the urine and the blood to be at the same pressure (or, possibly, would assign a higher pressure to the former), and he would dilute the urine to make the conditions agree with those which Ludwig supposed to exist; but his experiment would prove to him that, unless the urine were very dilute indeed, water would still tend to pass into it from the blood, and not _vice versa_. And here it may be remarked that the results of these experiments might have been predicted by calculation. When Ludwig advanced his theory, osmosis was a mysterious phenomenon. Its laws have since been accurately ascertained. Given the molecular weights of bodies in solution and their degree of concentration, the direction in which they will pass through a membrane can be predicted. The force with which water will tend to pass from one solution to another can be calculated. Urine as secreted contains far more urea, sodic chloride, and other salts than blood. It has a much higher degree of concentration. The concentration of blood is 0·55; that of urine, 1·85. Water passes from a less concentrated to a more concentrated solution, not _vice versa_. As a solution of a problem in hydrostatics Ludwig’s hypothesis is untenable.