IN METALLIC TUBES. 47 



Notwithftanding this, in fevcral other experiments 

 inflammable air becomes phlogifticated air without any 

 addition of oxygen ; as when it is expofed to heat in 

 copper or lilver tubes, and probably, therefore, thofe of 

 other kinds of metal. Inflammable air treated in this 

 manner is generally diminifhed in quantity, though not 

 always in the fame proportion. 



Three ounce meafures of inflammable air expofed half 

 a day to a red heat in a copper tube were reduced to 

 0.52, completely phlogifticated. Two ounce meafures 

 expofed to the fame degree of heat only a few hours, 

 came out 1.25. Anoth'er equal quantity was reduced 

 to three-fourths of an ounce meafure ; and two ounce 

 meafures expofed in this manner twenty minutes came 

 out 1.5, completely phlogifticated. 



I have, however, found a remarkable difference in the 

 refult of thefe experiments made with two cafl: copper 

 tubes, in one of which the metal is much thicker than 

 the other. In the larger and thicker of thefe tubes, the 

 air was always diminiflied; and though it continual- 

 ly approached to the ftate of phlogifticated air, it was 

 very flowly ; whereas in the thinner tube the inflamma- 

 ble air was always increafed in quantity, though the 

 whole of it never failed to be phlogifticated. In this 

 tube phlogifticated air alfo was always increaled in quan- 

 tity ; whereas in the larger tube it was neither increafed 

 nor diminifhed by the fame treatment. 



When I filled the fmaller tube with water only, and 

 expofed the clofed end to a red heat, I always found 

 much more phlogifticated air in it than when I ufcd the 

 larger tube in the fame mianner. Having filled the 

 fmaller tube with water, and only kept it in an inclined 

 pofition over the fire, fo that the heat to which it was 

 expofed did not much exceed that of boiling water, I 

 found in it the next morning 4 ounce meafures com- 

 pletely phlogifticated. 



In 



