On the transposition of AIRS.' 17 



of any kind of air was confined In an earthen tube clofed 

 at one end, while the open end Hood in a bafon of water 

 or mercury. After this I expofed the clofed end of the 

 tube to a red heat, which I contrived to do by means of 

 a hole purpofely made in the grate of a fmall furnace. 

 In this cafe whatever kind of air was contained in the 

 tube before the procefs, it was in a very fliort time of 

 the fame quality with that on the outfide, which, being 

 in the fire, was fomething worfe than the external air. 

 It made no difference alfo whether the tubes were glazed 

 or not ; and yet that they were air-tight appeared from 

 their containing only a certain portion of air after their 

 procefs, as well as before. There was always, however, 

 fome change in the quantity, but on what principle this 

 change was made I could not fatisfy myfelf. 



Three and an half ounce meafures of inflammable air 

 treated in this manner came out two ounce meafures, 

 nearly common air, with nothing inflammable in it. 



The fame quantity of nitrous air was reduced to 2^, 

 ounce meafures and to the fame ftate. A candle burned 

 very well it. The fame quantity of phlogifl;icated air 

 came out 24-, of the fame quality with the preceding; 

 but the dephlogifticated air was increafed to 4 ounce 

 meafures, of a ftandard a little better than the reft. 



In the preceding experiments the air was confined by 

 water j but the refult was the fame with thofe kinds of 

 air that required to be confined by mercury. 



Marine acid air treated in this manner was much in- 

 creafed, but came out very nearly common air. Vitriolic 

 acid air was neither increafed nor diminifhed, but was 

 not to be diftinguifhed from common air after the pro- 

 cefs. Alkaline air alfo was unchanged in quantity, but 

 in quality it was the fame as the reft. 



In order to repeat my former experiment on the tranf- 

 miffion of fteam in this eafy procefs, 1 filled one of thefe 



Vol. V. ' G^ tubes 



