OP ARTS AND SCIENCES. 285 



It contained no bromine. There can be no doubt, therefore, that 

 the product was the triethylether of trinitrophloroglucine, and that 

 the formation of this substance produced the sodic bromide which 

 was always obtained by the action of sodic ethylate on tribromtri- 

 nitrobenzol. 



Properties of Trinitrophloroglucine Triethylether. — This sub- 

 stance crystallizes in long slender plates, terminated usually by 

 one plane at a very sharp angle to the sides. These crystals are 

 often united longitudinally into broader plates, which frequently 

 have serrated ends, and may reach more than a centimeter in 

 length and a millimeter in breadth. The color is white, but it 

 turns brownish on exposure to the air. It melts at 119-120° ; is 

 soluble in cold ethyl or methyl alcohol, freely soluble in either of 

 these solvents when hot ; very freely soluble in benzol, chloroform, 

 or acetone; freely soluble in glacial acetic acid; soluble in carbonic 

 disulphide; slightly soluble in ligroine; essentially insoluble in 

 water, whether cold or hot. The three strong acids have no per- 

 ceptible action on it hot or cold, except that it is dissolved by hot 

 strong nitric acid, but on dilution it is precipitated unchanged, 

 to judge by the melting point. We have tried no experiments on 

 its saponification by the long continued action of acids. A cold 

 aqueous solution of sodic hydrate had no action upon it, and only a 

 slight action when heated, but in alcoholic solution sodic hydrate 

 decomposes it even in the cold, more rapidly when hot, forming a 

 reddish yellow solution of the sodium salt of a phenol. 



In addition to the tribromnitroresorcine diethylether and trini- 

 trophloroglucine triethylether, we have always obtained from the 

 crystallization a considerable amount of a substance which accumu- 

 lated in the mother liquors and separated from its alcoholic solu- 

 tion as an oil solidifying after standing for some weeks. We have 

 devoted a great deal of time to the study of this substance, because 

 it appeared in such quantity that we thought it must be a principal 

 product of our reaction; but this work has shown that it consists 

 principally of the products already described, mixed with ven' 

 little of an oily substance which prevents them from crystallizing 

 as usual. In a preparation from SO gr. of tribromtrinitrobenzol, 

 the mother liquors gave 4.3 gr. of the more or less liquid fraction, 

 but this amount was reduced to 1.2 gr. after removing the trini- 

 trophloroglucine triethylether completely by treatment with alco- 

 holic sodic hydrate, and the tribromnitroresorcine diethylether as 

 nearly as possible by crystallization from alcohol of the mixture 



