572 



Prof. T. G. Brodie. 



to conclude that the idea of filtration at the glomerular surface must be 

 abandoned. , 



Perhaps the most striking piece of evidence is derived from the considera- 

 tion of the concentration and constitution of the urines obtained during 

 extremely free secretion. The evidence is quite clear that the main bulk of 

 the water secreted by the kidney undoubtedly comes from the glomeruli. 

 Hence the more rapid the flow of fluid from the kidney the more closely 

 must that fluid resemble in constitution the fluid discharged from the 

 glomeruli, since a much shorter time is then allowed to the cells of the 

 tubules to modify it by absorption or secretion, and if filtration is the active 

 process in the glomeruli this fluid ought to approximate more and more 

 closely in composition to the blood plasma so far as the salts, urea and all 

 constituents of the plasma other than proteins are concerned. But the dilute 

 urine secreted after drinking copious amounts of lager beer,* or of water ,f 

 shows a constitution in salts widely different from that of the blood. Con- 

 sidering only the total concentration, as estimated by the depression of the 

 freezing point, it is quite easy to obtain a urine with A = — - l° C, and one 

 as low as — 0*075° C. has been recorded. j To effect a change in concentration 

 so extensive as this denotes, by filtration through a semipermeable membrane, 

 would necessitate a pressure difference on the two sides of the membrane of 

 at least 4000 mm. Hg, a pressure difference utterly out of comparison with 

 the blood-pressure. Therefore to make such a result accord with the filtration 

 theory, it becomes necessary to assume a most extensive reabsorption of 

 the salts and other substances of small molecular size, a reabsorption on such 

 an extensive scale and at such a rate as is, I think, entirely out of the 

 question. 



If, in the second place, we investigate the correlation between the blood flow 

 and the rate of secretion, we find that while there is a general correspondence, 

 in that increased urine flow is usually accompanied by increased blood flow, 

 this is by no means a universal rule.§ I have frequently observed in kidneys 

 in which there was at the start a fairly free blood flow and but slow urine 

 secretion, a copious diuresis to come on without any change in the blood 

 flow. Indeed on no less than five occasions I have seen a distinct decrease 

 in the blood flow to occur as the diuresis commenced, and moreover in these 

 experiments the volume of the kidney actually increased. In every direction 

 we find that the urine flow does not vary strictly with the blood flow nor 



• * Dreser, 'Arch. Exp. Path.,' 1892, vol. 29, p. 303. 



t Maeallum and Benson, ' Journ. Biol. Chem.,' 1909, vol. 6, p. 87. 

 J Maeallum and Benson, loc. cit. 



§ Cf. Gottlieb and Magnus, ' Arch. Exp. Path.,' 1901, vol. 45, p. 223. 



