178 Proceedings of the Royal Irish Academy. 



and its derivatives all react with alcoholic potash to give coloured compounds, 



none of which, however, they succeeded in isolating. By the simultaneous 



action of oxidizing agents, however, they obtained in all cases derivatives of 



either dinitro-dibenzyl or dinitro-stilbene, and in many cases obtained 



derivatives of both of these substances. The reactions were represented as 



follows : — 



ill. . i ,11, . NO CH . C 6 H, . NO 

 2CH 3 .CtU, . NOi -> -y 



< 11. . i Ml,. NO: CH . CH, . NO 



>J,1V,< \ IV b 



< II; . C 8 II. • N<». CH . C 6 H, . NOs 



I II 



CHi.CtHt.NOi CH.C 6 H, . N0= 



The coloured compounds which, as stated, were not isolated were supposed 

 to be alkaline salts of derivatives of the substances IV« or IV6, the colour 

 being due to a quinonoid structure. 



Will Ber. A. Dtaoh. Chem. Ges. xlvii (1914), pp. 704-717) states that 

 a-trinitrotoluene reai ts with alkali in alcoholic solution to give a deeply 

 coloured solution which on simultaneous oxidation yields hexanitrodibenzyl. 

 The identity of the product was demonstrated by the fact that it was also 

 obtained by niti i p' -dinitrodibenzyl, but not by nitrating tetranitro- 

 Btdlbene. Will also Btatea that 0- and -y-trinitrotoluene react with alkalies, 

 forming in each case a dinitro-metacresol, that from the /S-isomer being 



NOi 



A, "" 

 I I 

 \/\ 



i II 



and that from the y-isi'iiiei being 



NO, 



i i 



NO, 



CHi 



II, A<-' .\ ' Amines and oj Ammonia. 



Bepp Liebig's Annalen der Chemie ccxv, pp. 344-375 found that 

 Hrinitrobenzene and a-trinitrotoluene with aromatic amines formed additive 

 compounds p deep colours. Be 31 ited that 0-aml y-trinitro- 



toluene did nol form additive compounds with amines that aniline reaeted 

 with a hot alcoholic solution of -y-trinitrotoluene, forming a dinitrotolylpheny- 

 lamine melting at 14^ C, and that ^-trinitrotoluene yielded an isomeric 



