256 PROCEEDINGS OP THE AMERICAN ACADEMY 



I, 0.4010 gr. of the substance gave on combustion 0,5575 gr. of 



carbonic dioxide and 0.0745 gr. of water. 

 II. 0.2273 gr. gave by the method of Carius 0.3448 gr. of the mix- 

 ture of argentic chloride and bromide. 



III. 0.4160 gr. gave by the same method 0.6278 gr. of the mixture of 



argentic chloride and bromide. 



IV. 0.3014 gr. gave, when boiled with a solution of pure potassic 



hydrate, after acidification with nitric acid and precipitation 

 with argentic nitrate, 0.1962 gr. of argentic chloride. 

 v. 0.4284 gr. gave by the same method 0.2773 gr. of argentic 

 chloride. 



Calculated for Found. 



IV. V. 



16.08 16.01 



Properties. — White needles with an odor similar to that of benzoyl- 

 chloride, but much less marked, owing to its higher boiling point ; melt- 

 ing point, 30°. It boils at 245-247°* (uncorr.) with slight decom- 

 position ; water does not dissolve it, and acts upon it very slowly, if 

 at all, in the cold, more rapidly when boiling; it is easily soluble in 

 ligroine or benzol, while alcohol dissolves it with conversion into the 

 ethylester, a viscous liquid with an odor similar to that of oil of 

 anise. 



Parahromhenzoic Anhydride, (CgH^BrCO)^O. — This substance was 

 prepared by heating the sodic parabrombenzoate with parabromben- 

 zoylchloride not in excess ; convenient proportions are 3 gr. of the 

 sodic salt to 2 gr. of the chloride. The mixture was heated in a flask 

 with an air-condenser for an hour to 200° by means of a parafline bath, 

 and the product purified by washing, first with ligroine, then with a 

 strong solution of sodic carbonate, and finally with benzol to remove 

 the tarry impurities. The yield was about 50 per cent. Owing to 

 its very slight solubility in all the common solvents, we did not attempt 

 to purify it by crystallization, but this was fortunately unnecessary as 

 the following analyses show that the substance purified as above by 

 washing is essentially pure. 



* It was not thought worth while to try to determine this temperature with 

 greater accuracy, because of the decomposition with whicli the boiling is accom- 

 panied. 



