274 Mr. W. H. Balmain on the Formation of Compounds of 



upon was obtained chiefly from an incomplete analysis of the 

 " boronitruret of potassium," by heating it with hydrate of 

 lime; the result being diffused through water, a stream of 

 carbonic acid passed through it, and the whole boiled, borate 

 of potass was in solution and carbonate of lime precipitated, 

 which, being acted upon by muriatic acid, yielded an imper- 

 fectly transparent liquid, and from this the thready substance 

 was deposited on long-continued ebullition. 



All attempts to decompose the " boronitruret of potassium," 

 so as to isolate the theoretical " boride of nitrogen," have 

 hitherto been unsuccessful ; each experiment adding its testi- 

 mony to the remarkable stability of the compound. It can- 

 not be done by means of oxidizing agents, for both the potas- 

 sium and the boron take oxygen at the same time, and either 

 boracic acid and potass are formed, or else the thready substance 

 alluded to above, as appeared to be the case when peroxide 

 of manganese with boracic or sulphuric acid was used as the 

 oxidizing agent, since, after diluting and acting upon the re- 

 sidue with a solution of oxalic and sulphuric acids to remove 

 boracic acid and any remaining peroxide of manganese, a 

 white solid was left which had the same appearance and, before 

 the blowpipe, the same characters as that substance. 



Finding that cyanogen passed over a mixture of boracic 

 acid and charcoal heated to redness gave me no result, I en- 

 deavoured, as a last resource, to obtain compounds of " boride 

 of nitrogen" with the common metals by heating their cy- 

 anides with boracic acid, fully expecting that these cyanides 

 would decompose at too low a temperature for the deoxidation 

 of the boron to take place, and I was agreeably surprised 

 when upon trial it appeared that the cyanide of copper 

 heated with boracic acid gave a result, which, after being 

 washed, yielded ammonia when heated with a mixture of hy- 

 drate of lime and carbonate of potass; and cyanide of lead, a 

 result which not only yielded ammonia, but produced a phos- 

 phorescence before the blowpipe which differed from that of 

 the " boronitruret of potassium " only in its colour, which 

 was more yellow and less green. 



Both of these results were so impure, owing to the presence 

 of oxides in the cyanides, that their characters could not be 

 taken as those of the compounds of the metals with " boride 

 of nitrogen," and they were only valuable as proving the 

 possibility of making those compounds by such a process. 

 The copper result gave a very fine green flame before the 

 blowpipe, but would not phosphoresce ; and after the metallic 

 copper had been removed by nitric acid a substance remained 

 which appeared more like the "thready compound" supposed 



