realizing the Advantages of the Air-Engine. 21 



when the pressure and temperature of the steam have fallen 

 to the degree fixed for its condensation. This completes the 

 process C. During the process D the piston returns, and a 

 portion of the steam is liquefied by contact with some cold 

 conducting substance, which abstracts the heat generated by 

 its liquefaction, so as to maintain it at the fixed temperature 

 and pressure. The process D ought to stop in time to leave a 

 portion of uncondensed steam sufficient for the process A, now 

 about to be described. The water and steam being now pre- 

 vented from receiving or discharging heat by conduction, the 

 piston continues its return stroke, and forcibly compresses the 

 remaining portion of steam into the liquid state. This con- 

 stitutes the process A ; and the portion of steam so condensed 

 ought to be just sufficient, by the heat generated by its lique- 

 faction, to elevate its own temperature, as well as that of the 

 water previously liquefied, to the original temperature of eva- 

 poration, so that the entire portion of the water employed may 

 be in every respect in the same condition as it was at the be- 

 ginning of the cycle of processes B, C, D, A, which may be 

 repeated ad infinitum. Such an engine would fulfil the con- 

 ditions of maximum theoretical efficiency ; for the elevation of 

 the temperature of the water would be effected without expen- 

 diture of heat, and the only heat expended would be the latent 

 heat of evaporation: those results being produced by the proper 

 ' arrangement of the changes of volume and pressure undergone 

 by the working substance during each stroke. 



15. Impracticability of such a perfect Steam-Engine. — It 

 is impossible to fulfil wholly in practice the conditions pre- 

 scribed in the preceding article. To show the nature of the 

 obstacles, let us begin with the process A. The forcible com- 

 pression of a certain proportion of the steam into the liquid 

 state would not only cause a very inconvenient degree of in- 

 equality in the action upon the piston at different periods of 

 the stroke, but it is difficult to conceive any mechanism by 

 which it could be effected in practice. The steam must there- 

 fore be wholly liquefied during the process D, and the tempe- 

 rature of the feed-water must be raised from the point of con- 

 densation to that of evaporation by expenditure of heat. A 

 certain amount of heat is thus wasted ; at the same time the 



