358 PROCEEDINGS OF THE AMERICAN ACADEMY 



before evaporation contained bromic acid, as was proved by the follow- 

 ing experiments. Some of the filtrate containing no free bromine gave 

 no precipitate when argentic nitrate was added, but upon boiling with 

 alcohol a heavy precipitate of argentic bromide was formed ; to another 

 portion of this filtrate potassic iodide was added, when iodine was set 

 free in large amount. The reaction must run as follows : — 



CeH3(S03Ag)3 + 3 Br, + 3 H.O = 



3 AgBr + CeH3(S03n)3 + HBrOg + 2 HBr. 



Iodine acts on the argentic benzoltrisulphonate in a similar way. 

 We have tried also to obtain a bromine compound by heating the potas- 

 sium salt with bromine to 140° for sixteen hours, and also by heating it 

 dissolved in strong sulphuric acid with bromine ; but in neither case 

 was hydrobromic acid given off, the trisulphonic acid remaining unal- 

 tered. It is possible, however, that longer heating with bromine might 

 produce the desired result. 



Our attempts to obtain a nitro acid were no more successful than 

 those to make the bromine acid, as the potassic benzoltrisulphonate 

 can be boiled with fuming nitric acid, or even with a mixture of fum- 

 ing nitric acid and sulphuric acid, for over an hour without undergoing 

 any alteration. When, however, the chloride is heated to 140° for 

 sixteen hours with a large excess of fuming nitric acid, it is converted 

 into a new substance with the most uninviting properties, which we 

 have not studied further because sulphuric acid is eliminated in the 

 process, and we propose to confine' our attention to the ti'isulpho 

 compounds. 



Benzoltrisulphochloi'ide, CgH3(S02C])3. 



This substance is prepared by heating the potassic benzoltrisulpho- 

 nate, previously dried at 150°, with a little more than the calculated 

 amount of phosphoric pentachloride (10 parts of the salt to 15 of the 

 pentachloride) in a test-tube, or flask to 150° in an air-bath. If the 

 substances are heated in a dish, there is little or no action. After 

 the phosphoric oxychloride has distilled off, the product is treated with 

 water, and finally warmed with it, but this warming should not be long 

 continued, as the trisidphochloride is much more easily attacked by 

 hot water than the benzolmetadisulphochloride. After washing the 

 chloride thoroughly with warm water, it is purified by crystallization 

 from boiling chloroform. The yield was between 80 and 90 jier cent 

 of the theoretical. Samjjles contaminated with lower sulphochlorides 



