712 THE EXCRETION OF DEXTROSE 



is the case, but lie has not proved the point. It has been shown 

 that the rate of conversion of liver glycogen into dextrose is 

 retarded by calcium salts and favoured by the ions which cause 

 diuresis. Until we know that the glycosiiria in Brown's experi- 

 ments is not accompanied by, and therefore possibly due to hyper- 

 glycsemia, we cannot exclude the possibility that the glycosuria 

 may be of hepatic and not of renal origin. At present, however, 

 we may assume with Brown that this glycosuria is due to the effect 

 of ions on the kidney. This independence of the diuresis and 

 glycosuria suggests several possible explanations ; either that 

 the dextrose and water are secreted by different parts of the 

 kidney, and this is possible on Heidenhain's view ; or that salts 

 may influence separately the secretion by the glomerulus of water 

 and dextrose, which again is possible on Heidenhain's view, but 

 quite impossible on Ludwig's ; or that the same salt may increase 

 the absorption of dextrose by the tubule and not that of water. 

 The data which we have for deciding between these possibilities 

 is meagre. We do not know for certain the site in the kidney 

 of the excretion of dextrose. Nussbaum's experiments placed it 

 in the glomerulus, and although they are inconclusive, as we 

 shall see, they do afford strong evidence that this is so. In 

 phloridzin glycosuria we know that the kidney volume passively 

 follows the arterial blood pressure and shows no independent 

 expansion, that the diuresis is not associated with an increased 

 percentage excretion of chlorides, and that repeated injections 

 of phloridzin lead to changes in the cells of the convoluted 

 tubes, leaving the glomeruli unaltered. This evidence suggests 

 that the glycosuria of phloridzin at any rate is independent of 

 the glomerulus, and caused by an excretion of dextrose by the 

 tubules, and that the site of action of the calcium chloride is 

 the tubule. Of course it does not follow that dextrose is normally 

 excreted by the tubule, and that when calcium chloride stops the 

 glycosuria of a saline diuresis it is not acting upon the glomerulus. 

 The experiments of Mosberg and others, in fact, seem to show 

 that the path of excretion of sugar after phloridzin is abnormal. 

 For he found in frogs after ligature of the renal arteries that 

 injection of dextrose led to no secretion of urine, but that phloridzin 

 caused a flow of urine containing sugar. Thus in phloridzin 

 glycosuria we seem to have a strong indication that the tubule 

 can excrete and not simply absorb. The fact that calcium 



