SECRETION 



337 



a state known as " hydrcemia " or "hydraemic plethora." The same result can be 

 brought about more simply by the injection of a quantity of isotonic saline. In 

 both cases a large rise in the rate of urinary secretion results. It was further 

 shown by Knowlton (1911) that if a colloid, such as gelatine or gum acacia, was 

 added to the saline, so that the osmotic pressure of the colloid added was equal to 

 that of the serum colloids, injection of such solutions caused scarcely any increase of 

 flow. Again, Barer oft and IStraub (1910), by the ingenious device of replacing a 

 great part of the blood plasma by Ringer's solution, retaining the blood corpuscles 

 to supply oxygen, were able to obtain a greatly increased rate of secretion without 

 rise of blood pressure. 



The experiment is of sufficient interest to be described in more detail. A rabbit with a 

 blood pressure of 95 mm. of mercury was secreting 0'05 c.c. of urine per minute; 22 c.c. of 

 blood were removed, and replaced by 25 c.c. of Ringer's solution. The secretion rose to 0'4 c.c. 

 per minute, with a blood pressure of 52 mm. of mercury. The blood which had been removed 

 was centrifuged, the corpuscles washed, and made up with Ringer's solution to the volume of 

 the blood withdrawn. This was then injected ; the blood pressure rose to 84 mm. of mercury, 

 nearly as high as it was originally, while the flow of urine rose to 2 '35 c.c. per minute, or 

 nearly fifty times as rapid as the original one, and six times as rapid as that after simple 

 saline injection to replace the volume of the blood taken out. The authors point out the 

 possibility of great variations in the necessary filtration pressure by comparatively small changes 

 in the osmotic pressure of the colloids. Thus, suppose the pressure in the glomerular blood 

 vessels to be 27 mm., and the osmotic pressure of the colloids to be 25 mm., the pressure 

 available for nitration would be 2 mm. Suppose the osmotic piessure of the colloids to be 

 reduced by one-fifth, so that it becomes 20 mm., then the filtration pressure becomes 7 mm., 

 or 3 "5 times as great as before. 



From these various experiments it is clear that the filtration hypothesis is 

 capable of accounting for the production of a urine which is equivalent to the, 

 blood plasma minus its colloids. 



If the glomerular process is merely a filtration, it is clear that what work is 

 required for it is afforded by the blood pressure, in other words by the heart, so 

 that the cells of the glomeruli take no part in the performance of work. Barcroft 

 and Straub (1910) have made observations on the oxygen consumption of the 

 kidney which confirm this point of view. No increase in the oxygen consumed 

 occurred when the increase of secretion was brought about merely by dilution of 

 the blood. We shall see presently that where work has to be done, increased 

 oxygen is consumed in order to give energy by oxidation. 



A subsidiary point of interest in this connection, which also gives support to the filtration 

 theory, is that, if there were some special function of the cells of the glomerulus in the nature 

 of selective secretion, the kidney would not continue to turn out an important salt, such 

 as sodium chloride, when the organism has been deprived of it in the food. Cohnheim's 

 experiments (1912, p. 80) show that the contrary is the case. On the filtration hypothesis, 

 sodium chloride must appear in the liquid leaving the glomerulus as long as there is any of it 

 free in the blood. We shall see, however, that it may be reabsorbed to a considerable extent 

 in the tubules, 



A certain difficulty must not be overlooked. The urine of the frog, and of 

 water animals in general, as it leaves the kidney, is of a lower osmotic pressure 

 than that of the serum minus its colloids. After copious drinking of watery fluids, 

 this may also happen in man. Burian (1910) calls attention to this fact, and 

 suggests that the explanation may lie in the possibility of the particular layer of 

 blood plasma in immediate contact with the filtering surface being, under certain 

 conditions, of a lower concentration than the rest. An adsorption process of some 

 kind suggests itself. It is to be remembered that we have no direct knowledge of 

 the composition of the solution as it leaves the glomerulus in such cases, and it is 

 known that the tubules are able to absorb valuable stuffs from the glomerular 

 filtrate as it passes over their cells ; although, when the flow is very rapid, there 

 does not seem to be much time given for the process. It has been held that the 

 tubules secrete water, but there does not seem to be any evidence for this. The 

 functions of the tubules will be discussed later. 



Brodie (1914) calculates that the pressure required to drive urine down along the tubules 

 at the rate of diuresis is practically identical with the blood pressure in the glomeruli, so that 

 only one or two millimetres at most would be available for a filtration pressure. The difficulty 

 of accepting the calculation rests on the fact that the fourth power of the radius of the 



22 



