PURDIE'S REACTION 219 



hydroxy-compounds of the ordinary type proceeds by the 

 formation of an oxonium additive compound with methyl 

 iodide, as methyl tartrate, a substance which is most readily 

 alkylated, shows no tendency to form such derivatives (Irvine 

 and Moodie, loc. cit.). The methylation of tetramethyl 

 glucose is undoubtedly brought about in this way by pre- 

 liminary addition of methyl iodide and subsequent removal 

 of the elements of hydrogen iodide from the oxonium com- 

 pound by silver oxide, but this is to be regarded as an altogether 

 exceptional instance. 



Lander (4 and 17) takes the view that the first step in the 

 alkylation of tautomeric compounds of the keto-enol character 

 is the formation of silver derivatives of both isomerides, that 

 is to say, OAg and CAg compounds ; these subsequently 

 undergo double decomposition with alkyl iodides, yielding 

 the corresponding 0- and C-alkyl derivatives. 



The alkylation of amides and substituted amides has been 

 supposed (4 and 17) to take place in a similar manner, by the 

 intermediate formation of silver derivatives, but the recently 

 published work of Matsui (64) appears to negative this view. 

 It is shown that in the alkylation of amides such as acetamide 

 and benzamide, silver oxide can be replaced by cuprous 

 oxide, lead oxide, or even anhydrous potassium carbonate. 

 These substances, and silver oxide also would therefore seem 

 to act simply by removing hydrogen iodide. 



It will thus be seen that the discovery of the silver oxide 

 reaction has opened many lines of research, and it has proved 

 to be of more immediate profit to pursue these lines of work 

 rather than to closely scrutinise the reaction itself. Further 

 discussion of the mechanism of the reaction must therefore 

 be postponed until the present evidence has been considerably 

 amplified. 



CHARLES ROBERT YOUNG 



