44 On air HEATED 



2. Of the tranfmtjjlon of air through the fubfimKe rf 

 fame metallic tubes. 



When I had difcovered the rendy pafTage of air 

 through bladders and earthen tubes, I thought the fadl a 

 very extraordinary one, and ftill more, that the internal 

 and external air fliould change places, as I obferved in 

 my laft communication to the fociety. But 1 have fince 

 that obferved that even fome metallic tubes, though per- 

 fedlly air-tight, admit the tranfmifllon of air through 

 them when they are heated. Of this I had no fufpicion 

 til! after heating air in the experiments above mentioned, 

 I fometimes let them remain a coniiderable time before I 

 examined the air^ey contained ; not doubting but: that 

 whenever it fhould be convenient for me to do this, I 

 Ihould always find the air in the iame quantity, and of 

 the fame quality. But I frequently found that it was 

 much increaled, and that in thefe cafes there was always 

 a confiderable proportion of atmofpherical air in them. 

 This, however, was never the cafe with iron tubes, but 

 with thofe of copper, lilver, and gold. As the firfl 

 copper tubes I made ufe of were made of fheet copper 

 fodered, I had one cafl. folid ; and though I found it to 

 be pcrfeftly air-tight, (as appeared by fetting a I'yringe to 

 it, and being unable by that means to force any air through 

 it) it was evident that it was fufficiently porous for the 

 tranfmifhon of air. 



Having put 4^ ounce meafures of inflammable air into 

 this copper tube, fodered to a piece of a gun barrel, the 

 end of which was immerfed in a bafon of mercury, 1 

 found that two ounce meafures were expelled by the 

 heat when the clofed end was furrounded with hot coals. 

 After continuing fome time in this fituation, 1 found in it 

 1.45 ounce meafures, partially phlogifticated, fo that 25 

 meafures were reduced to 1.45. Afterwards, though the 

 tube continued perfectly air-tight, after a repetition of the 



fame 



