64 CORRELATION OF PHYSICAL FORCES. 



entirely arrested ; any residual motion has to be deducted 

 from the producing force. Experiments to verify this are 

 difficult of execution, and even of conception. If a weight, 

 the descent of which produces friction, fall very rapidly with 

 slight resistance, the heat from friction is very slight, and the 

 impact when the weight reaches the ground and force thence 

 dispersed is great. If it fall with extreme slowness, the heat is 

 also experimentally slight, for the cooling effects of surround- 

 ing matter go on nearly pari passu with the heating ; there is, 

 therefore, a practical degree of velocity at which the measured 

 heat is a maximum ; but how can it, in any given experimental 

 investigation, be affirmed with certainty that this point has 

 been attained, and if so, that the heat so measured is the real 

 mechanical equivalent ; motion not being actually arrested, 

 and the dispersion of force, by the residual impact on the 

 weight reaching its destination, being almost incapable of 

 measurement in terms of heat ? 



I have endeavoured to give a proof (by showing the 

 anomaly to which the contrary conclusion would lead) that, 

 whatever amount of mechanical power is produced by one 

 mode of application of heat, the same should, in theory, be 

 equally produced by any other mode. But in practice the 

 difference is immense ; and therefore it becomes a question of 

 great interest practically to ascertain what is the most con- 

 venient medium on which to apply the heat employed, and 

 the best machinery for economising it. One great problem 

 to be solved is the saving of the heat which the steam in 

 ordinary engines, after having done its work, carries into the 

 condenser, or, in the high-pressure engine, into the air. It is 

 argued you have a large amount of fuel consumed to raise 

 water to the boiling-point, at which its efficiency as a motive 

 agent commences. After it has done a small portion of work, 

 and while it still retains a very large portion of the heat 

 originally communicated to it, you reject it, and have to start 

 again with a fresh portion of steam which has similarly 

 exhausted fuel in other words, you throw away all, and 

 more than all the heat which has been employed in raising 

 the water to the boiling-point. Various plans have been 

 devised to remedy this. Using again the warm water of the 



