viii THE EXCKETION OF UKINE 427 



the whole of the crystalloid constituents of urine contained in the 

 blood plasma, were given off from the blood circulating through 

 them. Filtration is passively determined by the great difference 

 of pressure between the interior of the, glomerular capillaries and 

 the cavity of Bowman's capsule. The filtrate from the glomeruli 

 differs from that of the ordinary blood capillaries merely in con- 

 taining no protein, or traces only, because the walls of the 

 glomerular vessels are more resistant than those of the ordinary 

 capillaries, so long as they are under normal conditions. This 

 filtrate, then, has every characteristic of a very dilute urine. 

 During its passage through the uriniferous tubules it gradually 

 becomes concentrated by reabsorption of water into the lymph 

 that surrounds the tubules, which, having become poor in water 

 and rich in protein owing to the absorption of water in the 

 glomeruli, now takes up water by endosmosis. Thus, on Ludwig's 

 theory, the whole process of urinary secretion can be mechanically 

 explained as an effect of filtration and osmosis, without the inter- 

 vention of any specific secretory activity of the cells which 

 surround the glomeruli and line the tubules. 



We must now review the experimental data of the subject, 

 using them as tests of the value of either theory, by which the 

 ground will be cleared for the construction of another intermediate 

 explanation, which may harmonise better with the physiological 

 facts as a whole. 



III. The mechanism of urinary secretion can be experimentally 

 modified in two different ways : 



(a) Disturbance of the normal conditions of vascular circula- 

 tion in the kidneys, with various methods. 



(&) Alteration of the normal constitution of the blood, by 

 various means. 



The fact that the secretion of urine is intimately bound up 

 with the circulatory conditions in the kidneys has been established 

 by a series of convincing experiments. It is, however, necessary 

 to define exactly on what circulatory conditions increase of 

 secretion depends, and on what its decrease or total suspension. 



Various experiments on animals show that the amount of 

 urine flowing from the ureters increases or decreases with rise or 

 fall of arterial pressure. Thus on bleeding an animal copiously 

 the flow of urine into the ureters diminishes in proportion with 

 the fall of aortic pressure ; on re-injecting the blood that has 

 been lost, the flow of urine is accelerated, as the aortic pressure 

 regains its initial value. 



Goll (1854), in Ludwig's laboratory, obtained 8 - 9 grms. urine 

 from the ureters of a large dog in 30 min. before bleeding, 4'92 

 grms. after a loss of 530 grms. of blood, 7 - 66 grms. immediately 

 after transfusion of the blood that had been lost. 



Stimulation of the cervical vagus, by which the beats of the 



