Rotatory Steam-Engine. 61 



and a second valve of emission on the opposite side of the pis- 

 ton closed. At one and the same instant, therefore, the motion 

 of the piston has to be stopped in one direction and commenced 

 in the opposite direction, one steam communication closed, a 

 second opened, a third of eduction cut off, and a fourth re- 

 newed and all this (for the perfection of the engine) must be 

 done with the most absolute precision. 



But these processes, which produce the change of state from 

 rest to motion, and fi-om motion to rest, require time. Matter 

 acquires momentum which must be gradually removed, other- 

 wise that matter is subjected to concussion, as if by the stroke 

 of a hammer, and either suffers or produces injury. And, 

 on the other hand, when in motion, matter requires a force to 

 stop it equal to the force that gives it that motion. These ef- 

 fects, therefore, cannot be instantaneous ;* and it is necessary 

 that while the motion which the steam gives off be uniform and 

 continuous, the parts of the engine itself shall be allowed time 

 to be brought to a state of rest, without shock, concussion, or 

 jolt, and as gradually and gently be again urged to their great- 

 est velocity in the opposite direction. All these with exquisite 

 adjustment the craiik effects ; it stops the piston as gently and 

 softly as if it placed beneath it a cushion of eider down, and af- 

 terwards as gradually begins and accelerates its motion to its 

 highest velocity in the opposite direction. The valves, too, are 

 opened with the same perfect adjustment, being performed with 

 that gradual motion which proportions the largeness of the aper- 

 ture to the supply of fluid required to be transmitted. An ad- 

 justment so complete could only take place by such a relation 

 as subsists between the crank and piston, the one describing 

 uniformly the circumference of a circle, while the other moves 

 by simultaneous gradations of alternately increasing and dimi- 

 nishing extent. But this is not all that distinguishes the crank. 



2d, It is one of the highest recommendations of a piece of me- 

 chanism, that any very slight error in its construction shall not 

 very materially prevent its usefulness, nor any slight derange- 

 ment of its adjustment be attended with immediate destruction, 

 but that, on the other hand, the efficiency of the mechanism shall 



* " On sail que pression ne peut pas produire tout-i-coup une vltesse finie." 

 — Lagrange, Mech. Analyt., p. ii. sec. x. 



