346 



nitrogen, are in a gaseous state, composing a mixture which may be 

 detonated with a fresh addition of oxygen ; and in this way all the 

 hydrogen may be saturated with oxygen, and the nitrogen may be 

 obtained as a final result of the process. 



From a set of experiments thus conducted, the proportion of hy- 

 drogen to the nitrogen in ammonia was pretty uniformly 70'6 to 

 29 '4 ; but there was some irregularity in the entire quantities of 

 both, the highest results being as much as 200*6, and the lowest only 

 182 from 100 of alkaline gas; a difference which Dr. Henry cannot 

 fully explain, but supposes it may arise from absorption of ammonia 

 by mercury, and subsequent liberation by the shock. 



The letter concludes with the mention of some experiments on the 

 electrization of carburetted hydrogen, defiant gas, carbonic oxide, 

 and carbonic acid. 



Carburetted hydrogen and olefiant gas were each expanded con- 

 siderably in bulk ; no carbonic acid was generated, but charcoal was 

 deposited on the surface of the tube. 



Carbonic acid was also partially decomposed by long electrization, 

 and was converted into carbonic oxide and oxygen ; and accordingly 

 carbonic oxide underwent no change by the same operation. 



Some Observations on the foregoing Paper of Dr. Henry. By Hum- 

 phry Davy, Esq. 



Mr. Davy, having lately had much occasion to pay particular at- 

 tention to the electrization of ammonia during his researches on the 

 decomposition of nitrogen, has observed various sources of error that 

 may occur in the process, and may account for the difference between 

 his results and those of Dr. Henry. 



The first precaution that is to be taken, is to boil the mercury over 

 which the decomposition is to be effected ; next to prepare the gas 

 in a separate vessel, and thence transfer it to the boiled mercury for 

 electrization ; for unless these precautions are taken, water contained 

 in the mercury becomes the cause of that absorption of ammonia 

 suspected by Dr. Henry, and by thus adding fresh ammonia during 

 the process, occasions a fallacious result. 



A second error may also be occasioned, when the mercury has not 

 been boiled, by common air adhering to the sides of the tube. 



In a late experiment, Mr. Davy obtained from 15 measures of am- 

 monia 27 of permanent gases, consisting of 73 hydrogen and 27 ni- 

 trogen, agreeing so nearly with his original results of 74 to 26, that 

 he conceives either of them to be more near the truth than Dr. Henry's 

 estimate of 7 1| and 28|. 



With respect to Dr. Henry's attempt to prove the existence of 

 oxygen in ammonia by the formation of water, Mr. Davy does not 

 see much probability of success, as water existing in ammonia may 

 elude any hygrometrical test. And although in his own electrization 

 of ammonia the platina wires were tarnished, at the same time that 

 the ammoniacal gas seemed to lose weight during decomposition, he 



