216 FERGUSON'S LECTURES. 



LECT. Now, although the tube be exhausted of air, none of the 

 ^-v^j quicksilver will rise into it, because there is no air left 

 in the receiver to press upon its surface in the jar. But 

 let the air into the receiver by the cock k, and the quick- 

 silver will immediately rise in the tube ; and stand as 

 high in it, as it was pumped up in the last experiment. 



Both these experiments shew, that the quicksilver is 

 supported in the barometer by the pressure of the air 

 on its surface in the box, in which the open end of the 

 tube is placed. And that the more dense and heavy the 

 air is, the higher does the quicksilver rise ; and, on the 

 contrary, the thinner and lighter the air is, the more 

 will the quicksilver fall. For if the handle .Fbe turned 

 ever so little, it takes some air out of the receiver, by 

 raising one or other of the pistons in its barrel ; and 

 consequently that which remains in the receiver is so 

 much the rarer, and has so much the less spring and 

 weight ; and thereupon, the quicksilver, falls a little in 

 the tube : but upon turning the cock, and re-admitting 

 the air into the receiver, it becomes as weighty as be- 

 fore, and the quicksilver rises again to the same 

 height. Thus we seethe reason why the quicksilver in 

 the barometer falls before rain or snow, and rises before 

 fair weather ; for, in the former case, the air is too thin 

 and light to bear up the vapours, and in the latter, too 

 dense and heavy to let them fall. 



N. B. In all mercurial experiments with the air-pump, 

 a short pipe must be screwed into the hole i, so as to 

 rise about an inch above the plate, to prevent the quick- 

 silver from getting into the air-pump and barrels, in 

 case any of it should be accidentally spilt over the jar : 

 for if it once gets into the pipes or barrels, it spoils 

 them, by loosening the solder, and corroding the brass. 



8. Take the tube out of the receiver, and put one 

 end of a bit of dry hazel branch, about an inch long, 

 tight into the hole, and the other end tight into a hole 

 quite through the bottom of a small wooden cup : then 



