60 PHILOSOPHICAL TRANSACTIONS. [aNNO l/Ql. 



be acid, or pure water, the 2 kinds of air unite in nearly the same proportions. 

 But since water has an affinity to almost every substance in nature, and a pecu- 

 liarly strong one to the acid and alkaline principles, it may be impossible that it 

 should be wholly free from them ; and if they be in proper proportions to saturate 

 one another, and in the same quantities, their presence may never appear. 



As the reason why, in my former experiments, I always produced an acid liquor, 

 and never pure water, was my using too great a proportion of dephlogisticated 

 air; so the reason why M. Lavoisier and his friends generally produced but little 

 acid, and at last not at all, must have been, that the slow combustion which 

 they made use of gave the principle of acidity in the dephlogisticated air, and 

 the phlogiston in the inflammable air, a better opportunity of escaping, and 

 forming the phlogisticated air in their residuum, of which they have not pub- 

 lished any satisfactory account*; and it is probable, that the weight of these 

 elements compared with that of the water which forms the basis of the 2 kinds 

 of air, may be very small. That excellent philosopher M. de Luc supposes that 

 they have even no weight at all. 



M. Lavoisier himself, I observe, lays particular stress, (p. 262) on the slow- 

 ness of the combustion, as if he suspected it to be necessary to his result. This 

 circumstance may also account for my want of success in the attempts that I 

 made to repeat his experiment: for whenever I made a stream of inflammable air 

 to burn in a vessel of dephlogisticated air (which I contrived to do by means of 

 a less expensive, but I own a less accurate, apparatus than his) I always got some 

 acid, though less than in my own process; but I made a larger and stronger 

 flame than I imagine M. Lavoisier chose to produce. 



In the course of these experiments, I found, that when the inflammable air I 

 made use of was from turnings of cast iron, there was always a considerable 

 quantity of fixed air in the residuum, not less than -jL of a measure, after the 

 explosion of 2 measures of inflammable air and 1 of dephlogisticated; whereas 

 there was either no fixed air at all, or the slightest appearance of it imaginable, 

 when I made use of inflammable air from malleable iron, extracted either by 

 means of steam or acids. The principal of these experiments, as well as those 

 in my former papers on this subject, will be found to confirm the similar ones of 

 Mr. Cavendish; but they prove the source of the acid in the results not to be 

 what he imagined, viz. phlogisticated air, but the union of the dephlogisticated 

 and inflammable air; and they also make it at least doubtful, whether these 2 

 kinds of air compose pure water. 



* Since this was written. Mess. Fourcroy, Vauquelin, and Seguin, have published a very parti- 

 ciUar account of their experiment ; from which it appears, that, after the combustion of the '.? kinds 

 of air, there was a pretty large residuum of phlogisticated air, more tlian was contained in tlie airs 

 before combustion. See Annales de Chimie, for April 179', p. 35. — Orig. 



