182 PROCEEDINGS OF THE AMERICAN ACADEMY. 



These results, therefore, do not decide with certainty between the two 

 formulas, for, although the percentage of carbon speaks strongly in favor 

 of [C(5H2(N02)30CH3]2Ba02H2, the determinations of barium agree 

 better with the percentage calculated for the second formula. As the 

 most probable impurity in the substance was baric carbonate, we shall 

 adopt the formula containing baric hydrate provisionally. 



Proj)erties of the Addition Product of Trinitrcmisol and Baric Hy- 

 drate. — It forms needles of a fine red color verging toward scarlet. 

 When heated to 110°, it loses its water of crystallization, and turns a 

 beautiful dark crimson, but the dried salt absorbs water very eagerly 

 from the air, regaining its scarlet red color. This fact necessitates great 

 care in handling the dried salt during the process of analysis. Part, but 

 not all, of the water of crystallization is given off in vacuo. It is insolu- 

 ble in water. Acids decompose it, giving trinitranisol, but no alcoholic 

 odor was observed. 



Calcic hydrate added to a solution of picrylchloride in methyl alcohol 

 gave a red precipitate, but this decomposed almost immediately with 

 formation of trinitranisol. Dinitranisic acid dissolved in methyl alcohol 

 gave a magnificent purple coloration, when treated with aqueous baric 

 hydrate, but the substance formed was very unstable. It is mentioned 

 here, as it forms an additional argument for classing the trinitranisol 

 compounds described in this paper with the colored derivatives of dini- 

 tranisic acid discovered by Ittner and one of us. 



The work will be continued in this Laboratory. 



