OF ARTS AND SCIENCES. 303 



Properties of the Normal Tripropylether of Trinitrophloroglucine, 

 C 6 (OC 3 H 7 ) 3 (N0 2 ) 3 . — This substance crystallizes in plates often 

 as much as three millimeters long and one broad, which under the 

 microscope are seen to be made up of a number of flat prisms with 

 square ends united by their longer sides, and usually much striated 

 x on lines parallel to both sets of edges. Its color is white with a 

 very slight yellowish tinge, and it turns yellowish brown on stand- 

 ing exposed to the air. It melts at 109-110°, and is soluble in 

 cold ethyl or methyl alcohol, more freely when hot ; freely solu- 

 ble in ether, benzol, chloroform, acetone, or carbonic clisulphide; 

 soluble in glacial acetic acid; slightly soluble in ligroine, or in 

 water whether hot or cold. Neither strong sulphuric, strong nitric, 

 nor strong hydrochloric acid has any apparent action on it, whether 

 cold or hot. The best solvent for it is boiling alcohol. 



In another experiment no benzol was used, but the tribromtrini- 

 trobenzol suspended in propyl alcohol was treated with sodic pro- 

 pylate. The product in this case was an oil which did not solidify 

 even after standing for three months. We accordingly tried to 

 purify it by treatment with a cold solution of sodic hydrate, and in 

 this way obtained an orange solution, from which trinitrophloro- 

 glucine was easily isolated, pointing to the presence of the tripropyl 

 ether of this body in the oil. After we could obtain no more of the 

 sodium salt of the trinitrophloroglucine by further action of sodic 

 hydrate, the oil was washed with water, dissolved in alcohol, pre- 

 cipitated again with water, and dried at 100°, when we hoped an 

 analysis might throw some light on its composition; but in this we 

 were disappointed, as a determination of the amount of bromine 

 gave 45.29 per cent, whereas the tribromnitroresorcine dipropyl- 

 ether, which we hoped might have been formed, should contain 

 50.42 per cent, of bromine. As we could find no better way of puri- 

 fying this substance, we have been forced to leave undecided the 

 nature of the product formed by the replacement of nitro groups 

 in tribromtrinitrobenzol by propoxy radicals. Fortunately, it is not 

 a point of great importance. 



Action of Sodic Isopropylate on Tribromtrinitrobenzol. 



4 gr. of isopropyl alcohol were converted into its sodium com- 

 pound by treatment with 1.5 gr. of sodium in anhydrous benzol, 

 when it was found that the isopropyl alcohol acts on sodium more 

 energetically than normal propyl alcohol, but both act less readily 

 than ethyl alcohol. The isopropylate thus obtained was allowed to 



