Dr. Hofmann on the Poly-Ammonias. 533 



washings chiefly consist of hydrochlorate of aniline, which base 

 separates in oily globules on addition of potassa to the filtrate. By 

 testing the filtrate in this manner from time to time, it is found that 

 the basic body separated by addition of potassa gradually exhibits a 

 tendency to solidify, and ultimately falls as a yello\vish-white crystal- 

 hne precipitate. The residue upon the filter is now dissolved in 

 warm (not boiling) water, separated by a filter from a brown resinous 

 insoluble substance, and precipitated by ammonia or potassa. The 

 crystalline precipitate obtained in this manner is washed till free from 

 alkali, and repeatedly crystallized from weak spirit. It is diflficult 

 to obtain it perfectly white, a yellowish substance, which appears to 

 be partly formed during the process of solution, adhering with great 

 pertinacity. 



Thus obtained, the new base is a white crystalline powder ; fre- 

 quently it is obtained in minute scales, generally of a yellowish tint. 

 It is insoluble in water, but readily dissolves in alcohol and ether. 

 From the hot solution in these solvents it is precipitated by water as 

 a yellow oil, solidifying on cooling with crystalline structure. It is 

 easily dissolved by acids, with many of which it forms crystalline com- 

 pounds. From the saline solutions thus produced the base is repre- 

 cipitated by potassa and also by ammonia. The salts of the new base 

 are not very stable ; their solutions, especially when heated for some 

 time, inevitably contain more or less aniline, the crystalline base itself 

 undergoing changes which I have not yet sufficiently examined. 



The analysis of the new compound presents some difficulty. 

 Even after protracted exposure over sulphuric acid in the exsiccator, 

 it retains a small quantity of water, while a temperature of 100° is 

 apt to decompose it. 



The nature of the body was, however, readily established by the 

 examination of a perfectly stable hydrochlorate, and also of a very 

 definite platinum- salt. 



The results obtained in the analysis of these salts establish for the 

 new base the formula 



H J 



It is obviously formed by the substitution of the triatomic molecule 

 (C^II)'" for 3 equivalents of hydrogen in 2 molecules of aniline, 

 which thus coalesce into a diamine molecule. Accordingly the base 

 might be called diphenyl-formyl-diamine, that is, diammonia, in 

 which 2 equivalents of hydrogen are replaced by 2 molecules of 

 jihenyle, and 3 equivalents of hydrogen by 1 molecule of formyle, 

 1 equivalent of hydrogen remaining unreplaced. Its formation is 

 expressed by the equation 



4(C,, liy N) -h C2 HC1,= C.^ Ill, N, I-ICl + 2(Ci ., H; N, nCl ) 



Hydrochlorate of Hyilroclilorate of 



ili]ilicnyl-formyl- |)hcnylai)iinc. 



(liuminc. 



As seen from this equation, the new base, although uumistakeably 



