Mechanical Theory of Heat to the Steam Engine. 183 



nally filled with liquid ether and is then completely pumped 

 back into the boiler. In like manner the ether vapor is con- 

 densed in metallic tubes which are only externally surrounded 

 by cold water, and is then pumped back into the first mentioned 

 space which serves for the evaporation of the ether. In order 

 to keep up a uniform action, therefore, it is only necessary to 

 add as much water or ether as escapes through the joints from 

 imperfections in the construction. 



6. In a machine of this kind in which the same mass is always 

 employed anew, the different changes which the mass undergoes 

 during a period, must, as mentioned above, form a closed cycle, 

 or according to the nomenclature which I have chosen in my 

 former paper, a circular process (kreisprocess). Those machines, 

 on the other hand, in which a periodical taking up and throwing 

 out of masses occurs, are not necessarily subject to this condition. 

 They may however also fulfill it when they separate the masses 

 again in the same condition in which they have taken them up. 



This is the case with steam engines with condensers, in which 

 the water is thrown out from the condenser in the liquid state, 

 and with the same temperature with which it passed from the 

 condenser into the boiler.* 



In other machines the condition at the exit is different from 

 that at the entrance. The caloric air machines, for instance, even 

 when they are provided with regenerators, force the air into the 

 atmosphere with a higher temperature than it previously had, 

 and the steam engines without condensers take up the water as a 

 liquid and let it pass out again as a vapor. In these cases, no 

 complete circular process takes place, it is true ; nevertheless we 

 may always imagine a second machine joined to that which is 

 really present, which takes up the mass from the first machine, 

 brings it in any way into the initial condition, and then first lets 

 it escape. The two machines together may then be regarded as 

 a single machine which again satisfies the above condition. In 

 many cases this completion may be performed without producing 

 thereby too great a complication of the investigation. Thus for 

 instance we may imagine a steam engine without condenser, re- 

 placed by one with a condenser whose temperature is 100°, if 

 we only assume that the first is fed with water at 100°. 



Hence it appears that, upon the supposition that the machines 

 which do not in themselves fulfill the condition, may in this way 

 be completed for the purpose of investigation, we may apply to 

 all thermo-dynamic machines the theorems which hold good for 

 the circular process, and in this way we arrive at some conclu- 

 sions which are quite independent of the particular nature of the 

 processes taking place in the several machines themselves. 



* The cooling water which passes into the condenser cold and out of it warm, 

 is not here taken into consideration, since it does not belong to the heat-utilizing 

 substance, but serves only as a negative source of heat. 



