VOL. LXXVIII.] PHILOSOPHICAL TRANSACTIONS. 463 



Besides this, Mr. Gilpin had at different times introduced 872 measures of coip- 

 mon air, which makes in all 101 62 of air, consisting of 6968 of dephlogisti- 

 cated air, and 3 1 94 of common air. But as there were 900 measures of air 

 remaining in the tube when the accident happened, the quantity absorbed was 

 only 9262; but this is a much greater quantity than what from my own experi- 

 ments seemed necessary for this quantity of soap-lees. The soap-lees were 

 poured into a small glass cup, and the tube washed with a little distilled water, 

 in order that as little as possible might be lost. As they were by this means con- 

 siderably diluted, they were evaporated to dryness; but it was difficult to estimate 

 the quantity of the saline residuum, as it was mixed with a few particles of 

 mercury. 



Some vitriolic acid, dropped on a little of this residuum, yielded a smell of 

 nitrous acid, the same as when dropped on nitre phlogisticated by exposure to 

 the fire in a covered crucible; but it was thought less strong. The remainder 

 was dissolved in a small quantity of distilled water, and the following experiments 

 were tried with the solution. It did not at all discolour paper tinged with the 

 juice of blue flowers. It left a nauseous taste in the mouth like solutions of 

 mercury, and most other metallic substances. Paper dipped into it, and dried, 

 burnt with some appearance of deflagration, but not so strongly or uniformly as 

 when dipped in a solution of nitre. The marks of deflagration ho\yever were 

 stronger than when the paper was dipped into a solution of mercury in spirit of 

 nitre, but not so strong as when equal parts of this solution and solution of 

 nitre were used. A solution of fixed vegetable alkali, dropped into some of it 

 diluted, produced a slight reddish-brown precipitate, which afterwards assumed 

 a greenish colour. A bit of bright copper being dipped into it, acquired an evi- 

 dent whitish colour, though not so white as when dipped into the solution of 

 mercury in spirit of nitre. 



From these experiments it appears^ that the mixture of the two airs was ac- 

 tually converted into nitrous acid, only the experiment was continued too long, 

 so that the quantity of air absorbed was greater than in my experiments, and the 

 acid produced was sufficient, not only to saturate the soap-lees, but also to dis- 

 solve some of the mercury. The truth of the latter part is proved by the me- 

 tallic taste of the residuum, its not discolouring the blue paper, the precipitate 

 formed by the addition of fixed alkali, and the white colour given to the copper; 

 and the nitrous fumes produced by the addition of oil of vitriol, as well as the 

 manner in which paper impregnated with the residuum burnt, show as plainly, 

 that the acid produced was of the nitrous kind. It is remarkable however, that 

 during this experiment there were no signs which showed when the soap-lees 

 became saturated. The only time when the diminution proceeded much slower 

 than usual was on Jan. 4, It then seemed to go on very slowly; but as the air 



