300 Dr. Henry on the Analysis of some of [Oct. 



a property discovered by Dr. Priestley, but first applied to the 

 purpose of its analysis by Sir H. Davy, in the course of his 

 researches into the compounds of nitrogen.* In the experi- 

 ments of that philosopher, the results, approaching most nearly 

 to precision, were obtained by detonating nitrous oxide with 

 rather more than an equal volume of hydrogen, viz. 39 measures 

 of the former to 40 of the latter. Both gases were in this case 

 decomposed ; water was produced ; no nitrous acid was formed ; 

 and a volume of nitrogen remained, which always a little ex- 

 ceeded that of the nitrous oxide decomposed, viz. in the propor- 

 tion of about 41 to 39. 



In the repetitions which I have frequently made of this ex- 

 periment, a similar excess of the accruing nitrogen, over the 

 volume of the nitrous oxide employed, has always been ob- 

 served, and generally in about the same proportion. But ac- 

 cording to the law regulating the combination of gaseous 

 bodies with each other, which has been deduced by M. Gay 

 Lussac from a great variety of examples,f all lending to show, 

 that gases unite in proportions as to volume, which are either 

 equal, or simple multiples of eacli other, nitrous oxide ought 

 to be constituted of exactly one volume of nitrogen and half a 

 volume of oxygen condensed into the space of one volume ; and 

 those products should result from every careful decomposition 

 of the gas in question. That they are not correctly obtained 

 by the method which I have just alluded to, appears to be 

 owing to sources of inaccuracy, necessarily connected with 

 that mode of analysis. I was induced, therefore, to try various 

 other processes, among which there is one that may deserve to 

 be made known, since it exhibits, in a very summary way, and 

 by a single operation, the quantities of nitrogen and oxygen 

 that enter into the constitution of nitrous oxide, with as much 

 precision as, I believe, is attainable in the present state of gase- 

 ous analysis. This method consists in firing, by the electric 

 spark, a mixture of nitrous oxide and carbonic oxide in due 

 proportions. The nitrous oxide, which I employed, was ob- 

 tained by the careful decomposition of nitrate of ammonia, and 

 did not contain in 100 parts more than 3 parts of gas unab- 

 sorbable by well boiled water. The carbonic oxide was gene- 

 rated from recently ignited chalk and iron filings, and after 

 having been washed with caustic potash, appeared, from the 

 results of its combustion with oxygen, to be contaminated with 

 not more than 3 per cent, of foreign gas, having the properties 

 of nitrogen. Some nicety was found to be necessary in adjust- 

 ing the proportions of the gases to each other, in order to ob- 

 tain a perfect decomposition. When an excess of nitrous oxide 



* Davy's Researches, London 1800, p. 286. 

 + Memoires de la Soc. d'Arcueil, ii. 207. 



