524 
and which was also noticed in the nitration of 1.3.6 dinitro 
dimethylaniline. (See below). Moreover it appeared to be important 
not to allow the action of the nitric acid to be unduly prolonged. 
If for the nitration nitric acid is used which is completely free 
from nitrous acid, it is necessary to use a fairly concentrated acid 
(Sp. Gr. 1.37—1.40) in order to avoid prolonging the reaction 
unduly. To prevent excessive rise of temperature a moderate excess 
of nitric acid should be used. 
As a rule 10—12 ec. of nitric acid was added to 1 gram of 
the amine. If necessary the reaction may be accelerated by the 
addition of a small portion of sodium nitrite. In these conditions 
the two isomeric trinitro-derivatives are probably the only products. 
If, however, nitric acid is used containing much nitrous acid, or 
if the reaction is allowed to proceed for a long time, or, again, if 
the temperature rises appreciably above 20° C., then the action of 
the nitrous acid becomes manifest, and nitroso-compounds are easily 
formed. The isolation of the required isomerides is thus rendered 
extremely difficult, and, of course, the yield is reduced. In these 
conditions two light yellow substances of melting points 108° —109° C. 
and 201° C. respectively, were isolated. Both could be erystallised 
from alcohol and are obtained in the form of fine light yellow 
needles with a greenish reflection. The first can be obtained, as 
appeared later, by the action of nitrous acid on 1.3.4 dinitrodime- 
thylaniline and is transformed on treatment with nitric acid 
(Sp. Gr. 1.41) into the other. The latter is obtained by the action 
of nitrous acid on 1.3.4.6 trinitrodimethylaniline. 
In all probability these substances are therefore 1.3.4 dinitro- 
phenylmethylnitrosamine (M.p. 108°—109° C.) and 1.3.4.6 trinitro- 
phenylmethylnitrosamine (M.p. 201° C). 
The presence of the product of the reaction of nitrous acid on 
the 1.2.3.4 trinitro-compound, also a yellow substance with greenish 
reflection, which melts at 96°—97° C. after recrystallisation from 
alcohol, could not be detected. This is not surprising seeing that 
only a minute quantity of the 1.2.3.4 isomeride is produced in the reaction. 
Direct nitration of m-nitrodimethylaniline (M.p. 60° C.) with 
nitric acid (Sp. Gr. 1.4) did not lead to a better result. Here also 
the chief product was the 1.3.4.6 compound. 
The nitration of the 1.3.6 dinitrodimethylaniline which, like the 
corresponding 1.3.4 isomeride, is formed from dimethylaniline in 
presence of excess of sulphuric acid’), was also investigated. 
1) Rec. VI, 253 [1887]. 
