142 PROCEEDINGS OF THE AMERICAN ACADEMY 



molecule of each, but we obtained from the mixed solutions only crys- 

 tals melting at 285° ; as, however, the melting-tubes contained a slight 

 sublimate, and the habit of the crystals was somewhat different from 

 that of tribromtrinitrobenzol, we thought it possible, although not 

 probable, that a compound might have been formed, which decomposed 

 before it melted, and have accordingly analyzed the crystals, which, 

 remembering the instability of Hepp's substance, were dried only by 

 pressing between filter-paper, in order to be certain that the substance 

 should not be decomposed. 



0.1200 grm. of the substance gave, according to the method of Carius, 

 0.1500 grm. of argentic bromide. 



Calculated for *•„„,,,» Calculated for 



C 6 Br 3 (N0 2 ) 3 . * ouna ' C 6 Br 3 (N0 2 ) 3 C 10 H 8 . 



Bromine 53.33 53.20 41.52 



It is evident, therefore, that the substance is only tribromtrinitro- 

 benzol, and that it does not combine with naphthaline under these con- 

 ditions. The same negative result was obtained when chloroform or 

 ether was substituted for benzol as the solvent. We may add, too, 

 that the tribromtrinitrobenzol shows no tendency to unite with benzol, 

 so far as we could find. 



Triamidotrinitrobenzol, C 6 (NH 2 ) 3 (N0 2 ) 3 . 



When tribromtrinitrobenzol is mixed with cold alcoholic ammonia, 

 an action sets in almost immediately, as shown by the appearance of 

 an orange color in the solution ; and, if the substances are allowed to 

 stand for twelve hours in a corked flask at ordinary temperatures, the 

 reaction proceeds further, but is not complete, as is shown by the 

 presence of white specks consisting of unaltered tribromtrinitrobenzol 

 in the undissolved solid. It is necessary, therefore, in order to bring 

 this small amount of unaltered substance into the reaction, to boil the 

 mixture in a flask with a return condenser for about half an hour, 

 adding more alcoholic ammonia as it is needed. The nearly insol- 

 uble triamidotrinitrobenzol is then filtered hot from the orange liquid, 

 which has the color of a strong solution of potassic dichromate,* and 

 the paler yellow solid purified by washing, first with water to remove 

 ammonic bromide, and afterward with alcohol to get rid of the organic 

 impurities. 



* Our study of the substances contained in this liquid is not complete as 

 yet, but a description of the results obtained up to this time will be found on 

 page 145. 



