302 PHYSIOLOGY CHAP. 



shows that the solution contains very little sugar. On adding 

 saliva or pancreatic extract, as also on boiling with dilute mineral 

 acids, the opalescence and the iodine reaction disappear, while 

 Tromuier's test shows the presence of much sugar. 



Briicke obtained glycogen in the pure state by perfecting this 

 method. 



Estimation of Glycoyen (Pniiger). The liver of an animal is minced and 

 boiled for 2-3 hours in the water-bath with 30 per cent caustic potash. The 

 solution is then cooled, with the addition of 2 parts water and 4 \r.\i\ > 

 alcohol (96 per cent). The mixture is allowed to settle, filtered in a porcelain 

 filter, washed with a mixture of one volume 15 per cent potash and alcohol 

 at 96, and then with alcohol alone at 96. The precipitate is dissolved in 

 boiling water. The filtrate is boiled separately to extract the residue of 

 glycogen. The solution is then neutralised, and refiltered, after which 

 hydrochloric acid is added till the concentration is 2'2 per cent. It is then 

 boiled for 3 hours ; when cooled the liquid is neutralised and filtered. The 

 glucose is then estimated : 1 grin, glucose corresponds to 0'927 glycogen. 



After the discovery of glycogen, Bernard modified his original 

 view of the direct formation of sugar in the liver. The glyco- 

 genesis is indirect, i.e. is a function consisting of two quite distinct 

 processes, the first being the formation of the glycogenic substance 

 in the living hepatic tissue (amylogenesis), the second the conversion 

 of glycogen into sugar by a ferment, which is probably contained 

 in the blood (glycogenesis proper}. 



To Bernard's statement that the liver normally, i.e. during 

 life, manufactures sugar from glycogen, and discharges it by 

 internal secretion into the hepatic veins, Pavy (1861) objected that 

 glycogenesis is not a normal function of the liver, but a post-mortem 

 phenomenon. On drawing off the blood of a living animal from 

 the right heart with a catheter, he found that it contained hardly 

 a trace of sugar, while the same blood, collected after the death of 

 the animal, usually contained a considerable amount. On excising 

 fragments of liver from a live dog, and plunging them directly 

 into boiling water, he found mere traces of sugar in the extracts. 



Patter, MacDonell, Schiff, Lussana (1862-66) repeated Pavy's 

 experiments with variations, but always obtained the same results. 

 The living normal liver either contains no sugar or a small 



o o 



quantity only. They concluded that the liver only forms sugar 

 when the hepatic cells are injured by ueuroparalytic hyperaemia, 

 by heterogeneous substances injected into the vessels, post-mortem 

 decomposition, etc. The formation of glycogen is an essentially 

 physiological process; its conversion into glucose is an abnormal 

 or dissiniulatory post-mortem phenomenon. 



As against this conclusion and in favour of Bernard's theory 

 are the conclusions of Daltou (1871), who found that the living 

 liver contains Q'2-0'4 per cent sugar, a quantity in excess of that 

 usually found in the blood. Eecent analyses, moreover, have 



