OF ARTS AND SCIENCES. 287 



Properties of the Dietliylether of Trinitrophloroglucine. — This 

 substance crystallizes in square-ended rather short prisms, or in 

 longer flat needles, also usually with square ends, although some- 

 times instead of this they are very sharp; both forms are fre- 

 quently arranged in fan-like groups. It has a straw-yellow color, 

 and melts at 89° . Very freely soluble in ethyl or methyl alcohol, 

 ether, benzol, glacial acetic acid, chloroform, or acetone; soluble in 

 carbonic disulphide; very slightly soluble in ligroine; slightly 

 soluble in cold water, more soluble in hot; dilute alcohol is the best 

 solvent for it. The three strong acids have no apparent action on 

 it, hot or cold. Sodic hydrate dissolves it, forming an orange-red 

 sodium salt, which crystallizes in fine needles. Amnionic hydrate 

 dissolves it rather slowly, forming a yellow solution. Solutions of 

 sodic or acid sodic carbonate also dissolve it with evolution of car- 

 bonic dioxide forming yellow solutions; it is evident, therefore, 

 that the substance possesses strongly acid properties. 



The filtrate from the diethylether of trinitrophloroglucine, which 

 had changed from red to yellow on acidification, after concentration 

 on the water bath deposited upon cooling rather ragged thick yellow 

 needles; but if the evaporation was carried on spontaneously, yellow 

 hexagonal prisms were obtained, which dissolved in alkalies more 

 easily than in water, and gave, according to the amount of alkali 

 used, red or yellow salts, crystallizing in hair-like needles, proper- 

 ties which indicated that the substance was a polyatomic nitrophe- 

 nol, probably the trinitrophloroglucine. This substance has been 

 described by Benedikt * as crystallizing from water in hexagonal 

 prisms, and melting at 158° ; we accordingly purified our substance 

 by crystallization from water, but found that it melted as high as 

 167°. In spite of the want of agreement between this melting 

 point and that given by Benedikt, the crystalline form and other 

 properties indicated that the substance was trinitrophloroglucine, 

 and this was proved to be the case by the following analyses. As 

 Benedikt found that trinitrophloroglucine contains one molecule of 

 water of crystallization, we examined our substance for this. 



I. 0.7721 gr. of the air-dried substance when heated to a tem- 

 perature of 105° lost 0.0477 gr. 



Calculated for Found. 



C 6 (0H).,(N0,) 3 H 2 O. I. 



Water 6.45 6.18 



* Ber d. ch. Ges., XI. 1376. 



