THE METABOLISM OF CARBOHYDRATES 801 



immediately by an excessive secretion of urine and the appearance of sugar 

 in this fluid. The glycosuria may last from twenty-four to thirty hours. 

 If at the end of this time the animal be killed, the liver is found to be free 

 from glycogen. A sample of blood taken during the height of glycosuria may 

 contain from 3 to 4 parts of sugar per 1000. In order that the experiment 

 may succeed it is important that the animal be previously well fed. If the 

 puncture or ' piqure ' be carried out on an animal that has been starved or 

 whose liver has been freed by any means from glycogen, no glycosuria 

 is produced. It is evident that the effect of the puncture has been to cause 

 a rapid conversion of the glycogen previously stored up in the liver into 

 glucose. The glucose so formed escapes into the blood, raising the sugar 

 content of this fluid above the normal, and the excess is immediately excreted 

 by the kidneys together with an increased amount of water. A similar 

 temporary hyperglycsemia and glycosuria may be brought about by fright, 

 struggling or the administration of anaesthetics ; but the effect is absent, 

 if both splanchnic nerves have been previously divided above the supra- 

 renals. It has been shown (Elliott, Cannon) that all these conditions are 

 associated with an increased discharge of adrenalin from the medulla of 

 the suprarenals into the circulation. Since the injection of adrenaline itself 

 causes a condition of diabetes similar in all its limitations and aspects to 

 ' puncture diabetes,' it is now generally believed that the two conditions are 

 identical, and that the diabetic puncture acts through the splanchnic nerves 

 on the suprarenals, setting free adrenalin, which passing to the liver causes 

 a rapid ' mobilisation ' of the stored-up glycogen, and a consequent hyper- 

 glycaemia and glycosuria, lasting as long as the glycogen store holds out. 

 H [ (3) PHLORIDZIN DIABETES. Phloridzin is a glucoside extracted from 

 the root cortex of the apple-tree. It may be decomposed into a sugar and 

 phloretin. When phloridzin or phloretin is administered by the mouth or 

 subcutaneously it gives rise to glycosuria, unaccompanied, at first at any rate, 

 by any other symptom. The urine may contain from 5 to 15 per cent, of 

 glucose. The glycosuria induced in this way differs from the forms already 

 described in the fact .that it is not due to hyperglycsemia. Analysis of the 

 blood shows that the sugar is slightly diminished rather than increased. The 

 excretion of glucose seems to be due to a specific effect of the drug upon the 

 kidneys. If a cannula be placed in the two ureters so as to collect the urine 

 from each kidney separately and a small dose of phloridzin be then injected 

 by a hypodermic syringe into the left renal artery, the urine flowing from 

 the left ureter will in two minutes be found to contain sugar, while the 

 urine from the right kidney will not contain any sugar for another five or ten 

 minutes. The effect therefore is rapidly to drain off sugar from the blood. 

 In order to maintain the sugar content of the blood at its normal height the 

 liver must rapidly manufacture fresh sugar to take the place of that lost by 

 the kidneys. In the first instance the liver will utilise its 'stored- up glycogen 

 for this purpose. If a dose of phloridzin be given to each of two animals and 

 one animal killed as soon as the excretion of sugar is coming to an end, the 

 liver will be found free from glycogen. If now a second dose of phloridzin 



26 



