VOL. LXXXI.3 PHILOSOPHICAL TRANSACTIONS. 57 



which I at present use being so pure, that it contains no sensible quantity of 

 phlogisticated air. I also make use of no air-pump, but first fill the copper vessel 

 with water, and then displace it by the mixture of the 2 kinds of air; yet in 

 these circumstances, in which all phlogisticated air is excluded, I procure even a 

 stronger acid than before. 



The paper that I send along with this article contains the dry residuum of the 

 turbid green liquor, produced by a single explosion of a mixture of 2 parts inflam- 

 mable and something more than 1 part of dephlogisticated air, in a copper vessel 

 which holds 37 ounces of water; and a little more must have remained in the 

 vessel, which I could not get out by draining or shaking it. It is most evident 

 therefore, that the acid necessary to dissolve so much copper must have come 

 from the union of the dephlogisticated and inflammable air, because there was 

 nothing else in the vessel. The inflammable air was procured from iron by means 

 of steam. 



This very pure dephlogisticated air I first imagined could only be got by the 

 process in which I observed (Experi. on Air, v. 2, p. 170) that I once before 

 procured it, though I then supposed the extraordinary result to be accidental; 

 because in other circumstances I have sometimes had it very pure when I could not 

 succeed in a 2d attempt of the same kind. It was by heating the yellow product 

 of the solution of mercury in spirit of nitre, without suffering the red precipi- 

 tate into which it is converted by heat to come into contact with the external 

 air, from which I thought it probable that it might attract some phlogiston. 

 Afterwards however I found that this circumstance makes no difference whatever; 

 and that the air so procured appeared to be purer, arose from the greater purity 

 of the nitrous air which I made use of as a test, and which I got from mercury, 

 and not from copper, the nitrous air from which I find to be much less pure. 

 For trying the dephlogisticated air yielded by some red precipitate, which had 

 been prepared many months by the nitrous air from mercury, it appeared to be as 

 pure as that which was procured in the manner above described. 



That the dephlogisticated air which I now made use of was sufficiently pure 

 for my purpose, appeared from mixing l measure of it with 2 of nitrous air, 

 when the whole quantity was reduced to less than -^^ parts of 1 measure; so 

 that it is probable that, by a more accurate proportion of the 2 kinds of air, and 

 greater address in mixing them, they might have almost entirely disappeared. 

 There is besides some reason to think, from the great variety in nitrous air, that 

 the great part of this very small residuum comes from the nitrous air, and not 

 from the dephlogisticated. 



It will be said, how is it possible to reconcile the result of this experiment 

 with that of Lavoisier and his friends? which I was by no means disposed to 

 question after the publication of the Extract from the Register of the Academy 



VOL. XVII. I 



