216 Proceedings of the Royal Society of Victoria. 



Smolka {Monatsh. Chem. 8, p. 1, and Chem. Soc. abstracts, 

 1887, p. 566) finds that when Dextrose is boiled with excess of 

 KMnO^ it is completely oxidised thus : — 



G, Hi.Oe + 8 KM nO, = 2 KH., M n, Ojo + SK., CO., + 3 CO^ + 3H,, O 

 the MnOa being in combination with some of the K2O to form a 

 potassium hydro-majiganite (K._,0-3H20'8Mn02). 



The amount of oxidation varies with the temperature and 

 concentration of the solution and if there be not excess of 

 KMn04 present, oxalates, formates and unaltered dextrose, 

 result; whilst the KMnO^ is reduced partly to mangan/V and 

 partly to manganf??^^' oxides. 



It is obvious that cane sugar would be largely converted into 

 dextrose and laevulose by heating in dilute acid solutions. 



Morawski and Stingl {Journ. der prakt. Chem. [2] 18, p. 78) 

 find that KMnO^ on reduction in neutral or alkaline solutions 

 with oxalic acid among other reducing agents produces a brown 

 precipitate of KHgMn^On,, i.e. a salt of tetramanganic acid. 

 They also give a theory of the constitution of KMn04 and its 

 reduction products. 



Feilitzen and Tollens {Berichte, 1897, p. 2581) took strong 

 solutions of KMn04 and sugar containing about 5 equivalents of 

 KMn04 to one of sugar and left them after mixing until the 

 next day. 



The precipitate obtained was thoroughly washed (with alcohol 

 towards the end of the operation), dried, and then analysed with 

 the following result : — 



100-79 



The amount of carbon was found to vary from 8-5% to 1-6% 

 to 0-6% according to the amount of washing, and was probably 

 due to the presence of some unoxidised sugar. 



It will be noted that the precipitate was found to be ]Mn.,03 

 and not MnOj. 



