420 MM. Liebig and Gay-Lnssnc on [June, 



There being so great a difference between the results of calcu- 

 lation and experiment, we determined to examine the residuum of 

 the distillation, which was of a blackish-grey colour. We found 

 by treating a portion of it with water, that it was very alkaline, and 

 that which was not dissolved contained much chloride of silver. 

 Thus by the agency of the silver, and the oxygen contained in 

 the fulminate, a part of the chloride of potassium was decom- 

 posed, and converted into potash and chloride of silver, and the 

 former was combined with carbonic acid : this circumstance 

 explains sufficiently why so small a volume of gas was obtained. 



During the decomposition of the fulminate of silver mixed 

 with the chloride of potassium, a small quantity of carbonate of 

 ammonia was obtained, which was collected in one part of the 

 tube by surrounding it with a piece of paper moistened occa- 

 sionally with ether to cool it. For the purpose of determining 

 the quantity of carbonic acid combined with the ammonia, we 

 passed up a tube over mercury some muriatic acid, and a small 

 fragment of marble insufficient to neutralize it. We agitated it 

 thoroughly, in order to favour the solution of the carbonic acid 

 gas in the muriatic acid, and sometime after we introduced into 

 the solution the portion of the tube upon which the carbonate of 

 ammonia had condensed : effervescence was very perceptible ; 

 but the gas disengaged was not equal to 0*09 of a cubic inch ; 

 consequently the quantity of ammonia contained in the carbonate 

 could not amount to 018 of a cubic inch. 



The residuum of the decomposition of the fulminate of silver 

 necessarily containing some carbon, it was distilled with oxide 

 of copper; 17*874 cubic inches of gas were obtained, which, 

 added to the 1 1 "929 before procured, give a total of about 29*8 

 cubic inches; the difference between this quantity and 36-81 

 which ought to have been obtained, is still very great ; but the 

 potash formed, necessarily retailed a portion of carbonic acid, and 

 on other accounts this experiment was not sufficiently correct. 



The chloride of potassium not having answered our purpose, 

 we employed sulphate of potash, calcined and reduced to a very 

 fine powder, that it might be rubbed with fulminate of silver 

 without danger. 5-32 grains of fulminate mixed with about 

 twenty times as much sulphate of potash, and dried in a vacuum, 

 yielded by distillation 14-68 cubic inches of gas ; the same 

 quantity of fulminate distilled with oxide of copper would have 

 given 32-04 cubic inches. The residuum heated with oxide of 

 copper afterwards gave 14-96 cubic inches of gas ; but we 

 remarked that it was rather red, and' that consequently nitrous 

 acid was formed : this happened unquestionably, because, in 

 order that the bulk of the mixture might not be too great, too 

 small a proportion of oxide of copper had been used. There was 

 also produced a small quantity of carbonate of ammonia, which 

 appeared to us to be smaller than that obtained in the preceding 

 experiment, and no trace of water was perceptible, which seems 



