10 Mr. J. Gill on the Dynamical Theory of Heat. 



If the steam supplied to the reservoir were injected at the 

 same time and at the same rate as the steam is withdrawn in 

 giving motion to the working piston, so that the density and 

 pressure in the reservoir should remain constant, it would appear 

 that the whole work would be done directly by the injecting 

 piston, and merely transferred to the working piston by means 

 of a moveable plug of steam at constant pressure and tempera- 

 ture. In this case no heat should disappear either from the 

 reservoir or from the cylinder; and the contents of the cylinder 

 at the end of the working stroke should be identical with an 

 equal mass of steam of corresponding pressure and temperature 

 from any common source. 



The formation of steam from the water in a boiler may be 

 considered equivalent to the injection of steam ready formed from 

 some other source ; and during the whole stroke, if at full pres- 

 sure, or the full-pressure part of the stroke if working expan- 

 sively, only a small fraction of the heat assumed to be converted 

 into work would disappear from the cylinder itself, being in the 

 proportion of the contents of the cylinder to the whole contents 

 of the steam-space in the boiler, and smaller as the fluctuations 

 of pressure in the boiler due to the intermittence of supply to the 

 cylinder are bounded by narrower limits. On the other hand, 

 all the loss of heat which corresponds to the work done by ex- 

 pansion in the cylinder after the cut-off should take place in the 

 cylinder itself. Hence, in experimental researches on the assumed 

 disappearance of heat from the steam in the act of doing external 

 work, we should not expect to find tangible proof of this pheno- 

 menon by comparing the quantity of heat in the steam working 

 at full pressure in the cylinder during the whole stroke with the 

 amount of heat found in the condenser ; for theoretically these 

 quantities may be nearly equal, if, as supposed above, the work 

 given out by the piston is virtually done in the boiler ; and the 

 steam which fills the cylinder at the end of a working stroke 

 may differ but slightly from an equal weight of common satu- 

 rated steam of the same pressure taken directly from a boiler. 

 If such be the case, it were useless to search in this part of the 

 process for a loss of heat equivalent to the work done ; for the 

 exhaust-steam dashing tumultuously into the vacuous condenser 

 should produce there as much heat as there is cold produced in 

 the cylinder by the expansion which drives out the steam — as 

 in Joule's experiment of causing compressed air to expand into 

 a vacuous receiver, by which process the total quantity of heat 

 remains unaltered. This view of the case may help to account 

 for the apparently anomalous results obtained by Seguin and by 

 1 Iirn in their researches on this subject ; and as in my own ex- 

 periments sufficient account was not taken of the influence of 



