1 6 CALORIFIC POWER OF FUELS. 



CORRECTIONS FOR THE READINGS. 



The corrections to be applied to thermometric readings, 

 besides those due to the thermometer itself, are of various 

 kinds, and naturally vary with the kind of calorimeter used. 

 Some, however, are common to all. 



The correction relative to heating and cooling concerns all 

 calorimeters. Favre and Silbermann made this correction with 

 a coefficient previously determined, once for all, by a series 

 of experiments. For example, the coefficient that they found 

 for their calorimeter ( 0.0020225) represents the influence 

 of the external temperature through the envelopes and pack- 

 ings for one minute and one degree. 



Instead of a coefficient of correction thus determined, 

 use preferably a system of correction devised by Regnault and 

 Pfaundler. This system is superior to the preceding, as it 

 allows consideration of all external conditions at the time of 

 the experiment. It is evident, for example, that the evapora- 

 tion of a liquid may vary in such proportions that a fixed 

 coefficient will not always represent it. 



The system of Regnault and Pfaundler does not need 

 previous experiments nor a determined coefficient. It rests 

 on observation of the thermometer immersed in the bath a 

 Tew minutes before and after the experiment, or at the times 

 when external influence is at its minimum or maximum. 

 Knowing the value of these two kinds of influence, it is 

 easy to calculate it for the whole duration of the test. 



It is well to continue the observations before combustion 

 for some five minutes. These five minutes should be pre- 

 ceded by at least ten minutes' immersion of the combustion 

 chamber with agitator, so as to establish equilibrium of tem- 

 perature between the cylinder and the water. 



Suppose the initial correction corresponding to the first 

 period to be zero which is rare, it is true, but simplifies the 



