640 [June 18, 



agents upon dinitrobenzol. Phenylene-diamine, the last product of 

 this reaction, is preceded by the formation of an intermediate com- 

 pound, nitraniline, a substance discovered many years ago by Dr. 

 Muspratt and myself *, 



C 6 H 6 C 6 H 4 N0 2 C 6 H 4 N0 2 C 6 H 4 NH 2 

 N0 2 NH 2 NH 2 . 



Benzol. Dinitrobenzol. Nitraniline. Phenylene-diamine. 



Nitraniline, as might have been expected, was found to be readily 

 convertible into phenylene-diamine. 



By the researches of M. Arppef , chemists have become acquainted 

 with a second nitraniline, which is obtained by the action of fuming 

 nitric acid upon phenyl-pyrotartramide and subsequent treatment of 

 the nitro-compound with potassa. This substance, which, as I after- 

 wards found, may be more readily prepared by a similar treatment 

 of other less difficultly obtainable phenylamides, such as phenyl- 

 acetamide or phenyl-succinamide, is isomeric with ordinary nitraniline, 

 but differs from the latter compound both in its physical and che- 

 mical properties, so as to leave no doubt regarding the individuality 

 of the two compounds, which have accordingly been distinguished as 

 alp ha-nitr aniline and beta-nitr -aniline. This singular isomerism, 

 which has been traced also in other phenyl-derivatives, remains un- 

 explained ;. Whilst engaged with the examination of phenylene- 

 dkirine, the idea naturally suggested itself, to ascertain whether 

 beta-nitraniline, when submitted to reducing agents, would yield a 

 body isomeric but differing from the phenylene-diamine obtained from 

 dinitrobenzol and alpha-nitraniline. 



Beta-nitraniline is readily reduced by a mixture of iron and acetic 



* Chem. Soc. Mem. vol. iii. p. 112. 



t Chem. Soc. Journ. vol. viii. p. 1 75. 



J Among the various attempts I have made to decipher this isomerism, I may 

 mention the treatment of the two nitranilines with the iodides of methyl and 

 ethyl. But these substances are not acted upon by the reagents in question, and 

 I take this opportunity of correcting an error which has crept into my paper on 

 the molecular constitution of the volatile organic bases (Phil. Trans. 1850, vol. i. 

 p. 93). In this paper I state that the action of iodide of ethyl on nitraniline gives 

 rise to the formation of hydriodate of ethyl-nitraniline. This statement is based 

 upon a single platinum determination. The platinum-salt of nitraniline con- 

 tains 28'66 per cent, of platinum, that of ethyl-nitraniline 26-53 per cent. Ana- 

 lysis had furnished me 26-23 per cent. I have since satisfied myself that the salt 

 was the imperfectly purified platinum-salt of nitraniline. 



