302 PROCEEDINGS OF THE AMERICAN ACADEMY 



Action of Sodic Propylate on Tribromtrinitrobenzol. 



4 gr. of normal propyl alcohol were mixed with anhydrous ben- 

 zol, and treated with 1.5 gr. of metallic sodium until all the sodium 

 had disappeared; the reaction ran slowly, and was assisted by a 

 gentle heat. The sodic propylate thus obtained was carefully 

 cooled, and then a benzol solution of 10 gr. of tribromtrinitroben- 

 zol added in small portions at a time, shaking and cooling with 

 water after each addition. The proportions are about three atoms 

 of sodium and three molecules of propyl alcohol to each molecule of 

 tribromtrinitrobenzol. As the first portions of the solutions were 

 mixed the liquid turned blood-red, and this color, which at first faded 

 to yellow on shaking, became permanent as more of the solution of 

 tribromtrinitrobenzol was added. The mixture, which at no time 

 showed a rise of temperature, was allowed to stand over night in a 

 corked flask, and then the solution was filtered from a deposit of 

 solid matter, which, after drying off the benzol, gave tests for a 

 bromide and a nitrite. The benzol solution to which the solid not 

 used for the above tests had been added was treated with water, 

 and then acidified with dilute sulphuric acid, washed thoroughly, 

 and, after most of the benzol had been recovered by distillation, 

 the rest was distilled off in a current of steam. The residue thus 

 obtained was oily, but solidified after standing for some time; it 

 was then purified by crystallization from alcohol until it showed the 

 constant melting point 109-110°, when it was dried at about 70°, 

 and analyzed with the following results : — 



0.2366 gr. of the substance gave on combustion 0.4058 gr. of 

 carbonic dioxide and 0.1214 gr. of water. 



Calculated for 

 C 6 (0C 3 H 7 ) ;! (N0 2 ) 3 . Found. 



Carbon 46.51 46.78 



Hydrogen 5.43 5.70 



The substance gives no test for bromine when heated on a copper 

 wire. These results prove that it is the normal propylether of 

 trinitrophloroglucine. 



The aqueous solution, acidified with sulphuric acid, from which 

 the benzol had been separated gave on extraction with ether a 

 small amount of a yellow substance, which seemed to be trinitro- 

 phloroglucine, but was not present in sufficient amount for complete 

 identification. 



