1268 PHYSIOLOGY 



theories of urinary secretion both of Bowman and of Ludwig (1844), 

 theories which have furnished the basis of all our subsequent investiga- 

 tion of the subject. Both observers appreciated the great difference 

 between the membrane covering the glomerular loop and the lining 

 membrane of the tubule, and both drew attention to the difference in 

 the circulation in these two portions of the kidney. The glomerular 

 capillaries, supplied with blood through a short wide artery and drained 

 by an efferent vessel smaller than the afferent, would represent a region 

 of very high capillary blood pressure, whereas the pressure in the capil- 

 laries surrounding the tubules must be low and similar to that in other 

 capillary regions. Bowman therefore suggested that the urine consisted 

 of two parts, namely, one part containing the water and salts produced 

 by a process of filtration through the walls of the glomerular capillaries, 

 and another part, containing the specific urinary constituents urea, 

 uric acid, &c., secreted by the cells probably of the convoluted tubules. 

 To Ludwig, on the other hand, it seemed possible at first to account for 

 the whole process of formation of urine without the assumption of any 

 active intervention on the part of the cells of the tubules. He im- 

 agined that the whole of the urinary constituents passed from the 

 blood to the urinary tubule in the glomerulus by a process of filtration. 

 The glomerular transudate would represent therefore a very dilute 

 urine containing the crystalloids of the blood in the same concentra- 

 tion as in the blood and with no more urea than the blood itself con- 

 tained. The great difference in urea content between the blood 

 and the fully formed urine he ascribed to a process of concentration 

 taking place in the fluid in its passage through the tubules, in which 

 water and certain of the salts were reabsorbed, a process of re- 

 absorption conditioned by the difference in protein content between 

 the urine within the tubules and the lymph under low pressure on 

 the outside of the tubules. We know now that in its original form 

 the theory of Ludwig is untenable. If a process of concentration 

 takes place within the tubules it must involve the performance of 

 work by the cells lining these tubules, and could not take place as 

 a result of mere differences of colloid content between the two fluids. 

 It was shown long ago by Hoppe-Seyler that, if urine be dialysed 

 against serum, a passage of water takes place, not from urine to serum, 

 but from serum to urine, i.e. the latter is much more concentrated than 

 the former. The movement of water from one fluid to another 

 through a colloid membrane depends on the relative osmotic pressures 

 of the two fluids, and this in turn is determined by the molecular con- 

 centration of the two fluids. It is easy to estimate the molecular con- 

 centration of any sample of serum or urine. The method which is 

 most convenient is to determine the depression of freezing-point in the 

 two fluids. Whereas serum- ordinarily freezes at 0-56 C. to 0-59 C , 



