Mr Dunn's Description of an Improved Air- Pump. 385 



ly, or altogether as well, divested of all the other peculiarities of 

 his instrument, and possessing the decided advantages of be- 

 ing cheaper and much more easily kept in order. 



Fig. 3. is a section of the barrels of my pump, in which I em- 

 ploy metallic valves vi/ ni the bottom of the barrels, and waxed 

 silk ones S S' in the pistons, laying aside Cuthbertson's metal- 

 lie valves in the pistons, removing all his apparatus from the 

 top of the barrels, and leaving tlie pistons eacposed to the atmo^ 

 sphere, as I consider all those contrivances to be unnecessary, aU 

 though it has been uniformly held essential to a good air-pump, 

 since the time of Smeaton's invention, that the pressure of the 

 atmosphere should be taken off the piston-valves ; and my rea- 

 son for doing so is, that the air will be always so compressed in 

 the barrels, by the descent of the pistons, as of itself to have 

 sufficient elastic force to open the silk valves in the pistons, the 

 capacity of the barrels being each several thousand times greater 

 than the space betwixt the two valves, when the piston is at the 

 bottom. In fact, by making the under side of the piston and 

 the bottom of the barrel fit each other, which, with the assist- 

 ance of the oil employed in the barrels may be done perfectly, 

 there will be no space left but the small hole in the piston to its 

 valve. 



For illustration, let us suppose the stroke to be 12 inches, 

 and the diameter of the barrels 2 J inches, or 25 tenths (as is 

 the case in Mr Lees' one)^ the diameter of the hole e one-tenth 

 of an inch, and its length 1 inch, their circles being to each 

 other as the squares of their diameters, we have 1 x 1 = 1 for 

 the capacity of the hole, and 25 x 25 x 12 = 7500 for the ca- 

 pacity of the baiTels ; and consequently air, which, in the recei- 

 ver was 7000 times rarer than the atmosphere, would have suf- 

 ficient elastic force to open the valve in the piston ; but as this 

 is a degree of rarefaction far beyond what has ever been attained, 

 or even expected, it follows that any greater nicety of construc- 

 tion here is unnecessary. 



The above plan may, however, be objected to, on account of 

 its still leaving smnething to depend on the elastic force of the 

 air which, should any one consider desirable to be removed, can 

 be so done by adapting metallic valves I I^ with projecting 

 points p' p\ to strike against the bottom of the barrels, having 

 the spaces CK T^, O I, filled with oil, to exclude the external air 



