Cli. XLL] METABOLISM OF CARBOHYDRATES 615 



condensation of these two sugars to form lactose is a comparatively 

 simple problem ; there is no doubt that galactose is present in certain 

 foods, for instance it is contained in some vegetables, and it also 

 comes from the sugar of milk. Lactose in the mammary gland is, 

 however, not the only place where galactose is necessary, for the 

 galactosides of nervous tissue (see p. 437) also contain it. But it is not 

 necessary to assume that all the galactose necessary for the formation 

 of milk sugar and of the galactosides comes direct from the galactose 

 of the food. Dextrose, galactose, and laevulose are all isomeric, and 

 the intramolecular rearrangements that would be necessary to 

 convert one into another member of this group do not seem to be 

 beyond the power of the tissue cells, and there is a good deal of 

 evidence that such transformations actually occur in the body. The 

 other carbohydrate constituents of the body are pentoses found in 

 some nucleo-proteins, and glucosamine in the gluco-proteins, but 

 in relation to these our knowledge is highly speculative. 



It is further known that the hepatic glycogen may, under certain 

 circumstances, originate from proteins, and that many proteins con- 

 tain a carbohydrate radical ; some, such as mucin, yield a considerable 

 amount, but in the commoner proteins, the amount is in the neigh- 

 bourhood of 1 per cent, or less. The mucins do not participate to any 

 great degree in metabolism, and so the question arises whether the 

 small amount of carbohydrates in the ordinary proteins is sufficient 

 to account for the amount of glycogen formed. Arithmetic shows it 

 will not; moreover, a protein (such as casein) which contains no 

 carbohydrate radical at all, is equally efficacious as the others in 

 yielding sugar when administered to animals suffering from glycosuria 

 produced by phloridzin. We must therefore search for something 

 in the protein molecule, as the source of the carbohydrate, and 

 leucine, having like sugar 6 atoms of carbon, was naturally the first 

 substance to be examined. But all experiments on the admini- 

 stration of leucine led to negative or nearly negative results. Another 

 amino-acid, alanine (with its compounds tyrosine, phenyl-alanine, and 

 tryptophane), was found to be the much more probable source of 

 the carbohydrate. The relationship of alanine to the carbohydrates 

 is a near one, for if HO is substituted in its molecule for NH 2 , we 

 get lactic acid, and it was found that the administration of alanine 

 to rabbits led to glycogen formation in their livers, and to the 

 passage of lactic acid into their urine. Another cleavage product of 

 protein, namely, aspartic acid, may act in a similar way, and this also 

 is intelligible on chemical lines, for if aspartic acid loses carbon 

 dioxide, it is converted into lactic acid, and it is no great step from 

 this to sugar. Similar experiments have been performed with 

 glycine and other amino-acids, but the results, though in part positive, 

 are by no means go clear, nor is the chemical relationship between 



