in Exhausting Air from Tubes, ^c, 11 



the pipes, k, and m, into communication, and the cock, s, 

 turned so as to put the pipes, n, and p, into communica- 

 tion, and close the aperture, v, the cock, q, turned to put 

 pipe, p, in communication with the atmosphere through 

 the aperture, x, also the cock, r, must put o^, in commu- 

 nication with Y. The signal being given to exhaust the 

 traction-tube, the engine will be again put in motion and 

 both the pumps put in action, the engine being sufficiently 

 powerful to exhaust the tube to the lower numbers of the 

 barometer gauge, but will soon become unequal to continue 

 the exhaustion, but the cock, q, being turned one quarter 

 round, so as to put p, in communication with t^, the vacuum 

 in the first reservoir will then not only relieve the engine 

 of the resistance of pump, a, but will aid in working 

 pump, B. Afterwards, as the exhaustion of the traction- 

 tube increases, and that of the first reservoir decreases, such 

 aid will become insufficient; the cock, r, may then be moved 

 one quarter turn, putting o^, in communication with the 

 first reservoir; the power of engine being then in excess, 

 Q, may be turned back, connecting p, with the atmosphere ; 

 after a few strokes, turned again to t^, afterward to t^, 

 then R, may be turned to t^, q, back to t^, again to t^, 

 and then to t^, when the exhaustion should be complete, 

 the train being passed, and leaving a considerable degree 

 of exhaustion in the third reservoir ready for any con- 

 tingency. The train being passed, the exhaustion of the 

 reservoir, diminished by the before-described action, is to 

 be restored by the engine and pump, a, as hereinbefore 

 directed. 



Another part of my improvements consists in the con- 

 struction of the pistons of large air-pumps, regarding 

 which I will remark, that the pistons of small air-pumps 

 are generally packed with leather, sometimes with strong 

 leather cut or turned to fit the cylinder of the pump, and 

 sometimes with leather so arranged that the leather forms 

 a cup or concavity in the direction in which the air would 

 tend to pass by the piston ; but these methods do not pro- 

 vide any sufficient adjustment of the leather packing after 

 it becomes worn, nor would such packings work at all 

 times so far air-tight as desirable without considerable 

 friction being produced. To obviate, as far as possible, 

 these defects, and, at the same time, to avoid the friction 

 and expense attendant on the use of metallic pistons, I 

 pack the pistons of air-pumps with leather cut into bands. 



