I 3^5 1 



The difEculty is, to account for the non-produdion of air In 

 the fecond cafe above-mentioned, and for its apparent extinc- 

 tion in the third. As to the former, there was indeed then 

 almoft no air nor humidity in the barrel, and it was feparated 

 from whatever might be in the little rec"^ ; and though there 

 was oil, yet I found by the ointment put upon the cocks, which 

 was ever moft and fooneft turned green in colour near the 

 dudls and outer edges where the air had accefs to it, than in 

 the other parts, where it had not, that air was neceflary to 

 make the oil ad on the metal ; (nay, perhaps it is not the acid 

 in the oil, but in the air, which is the folvent;) whence in 

 this cafe, as there was no air in the barrel, fo there was no 

 fuch adion on the brafs of it. As to the phasnomena in the 

 third cafe, in which I evidently perceived a diminution of air, 

 though I cannot explain it, being ignorant of the chymical ana- 

 lyfis of elaftic fluids, and of thofe combinations of airs, which 

 occafion fuch diminution, and what might be the effed of the 

 fridion of the pifton in this cafe, from which refulted all the 

 phasnomena ; yet what I have ftated will, as I think, authorize 

 me to fay, that the perfedion of the air-pump does not depend 

 only on its mcchanifm, but alfo on the materials of which it 

 is made, and that thofe materials are the beft, on which the 

 oleaginous and aerial acid has leaft adion. Accordingly the 

 pewter-metal pump, which I firfl; made, though inferior in the 

 lize of the barrel, which was a great difadvantage, yet performed 



better 



