198 REPORT— 1902. 



to hydroxides R.NjOH, which are colourless unless they contain sub- 

 stituent nitro-groups, whilst the s^/w-isomerides furnish extremely explo- 

 sive yellow diazo-anhydrides. ^ These substances, which may sometimes 

 be obtained by treating a diazonium salt with an alkali s^/^ diazo-oxide, 

 slowly couple with the phenols, yield 0-diazo ethers with the alcohols, 

 and react explosively with benzene, yielding diphenyl-derivatives. 

 According to Bamberger's earlier papers the anhydrides should have the 

 formula 



R.N.O.N.R, 



N N 

 Hantzsch's stereo-chemical formula,^ 



N.R R.N 



II II 



N— O— N 



however, explains why the anhydrides readily yield s?/7i-diazo-cyanides on 

 treatment with hydrogen cyanide,^ but only very slowly dissolve in 

 hydrochloric acid to form diazonium chlorides. Bamberger has since 

 suggested another formula for these substances, which assumes that they 

 are salts of the basic diazonium hydroxide with the isomeric acidic diazo- 

 hydroxide 



R.N.O.N : NR 



N 



This configuration accounts for their formation from a diazonium chloride 

 and an alkali diazo-oxide. The first of these formulae is the least pro- 

 bable, but it is not possible at present to decide between the remaining 

 two. 



Cyclic Diazo-comj)ounds. — Diazotised sulphanilic acid and other cyclic 

 diazo-compounds are taken to be either diazonium salts, 



K 



'N^;N 

 SO.,/ 



or S2/«-diazo-anhydrides,'* 



r/Nn 



the introduction of negative substituent radicles increasing the tendency 

 to assume the si/w-configuration. The azimino-derivatives are assumed to 

 be syw-compounds, 



\NH/ 



\ Bamberger, Ser., 1896, 29, 446. « Ber., 1896, 29, 1067. 



» Ber., 1898, 31, 636. 



* Hantzsch, Ber., 1895, 28, 1734, and Ber., 1896, 29, 1622. 



