SS8 Mr Meikle on the Theory of 



But AE = — ; AH = —J— — — — , and the compres- 

 sion has evidently increased the density in the ratio of unit to 



AE „ 



-^ = ,, Hence 



I _ 1 -f ar-k-ai 



1 + a 



as in equation (D). 



Again, area BDGI'] rr k log ^-| = area BCIH = log ^^ 



and the compression has augmented the pressure in the ratio of 



•. . AH „ 



unit to -— - — p. Hence 

 AB ^ 



'p-=i ^ y as in equation (C). 

 In the hyperbola, as already hinted, the areas vary arithme- 

 tically, whilst the abscissae vary geometrically. But the varia- 

 tions of the area represent variations of heat, and the varia- 

 tions of the abscissa represent the corresponding variations of 

 volume under a constant pressure, or of pressure under a con- 

 stant volume. So that, besides agreeing with the other con- 

 elusions, this construction exactly represents the former result,, 

 that the real temperatures are as the logarithms of those on the 

 common scale, reckoning from — 448'' F., or — 266°.7 cent. ; 

 and placing the new zero at — 447° F., or at — 265°.7 cent. 



The law of temperature now given, stands on a much surer 

 foundation than any other that has been proposed, and affords 

 ample ground for questioning the present graduation of our 

 thermometers. Were DG a straight line parallel to AB, as the 

 ordinary graduation supposes, then, whatever CI might be, it 

 is evident, that, unless in a few particular cases, very little 

 change of temperature could be produced by a change of den- 

 sity ; because there would not then be that inexhaustible source 

 of both heat and cold which experiment proves, and which a 

 line differing much from the hyperbola could not supply. In- 

 dependently therefore, of more elaborate proof, this considera- 

 tion alone ought to overturn the common theory. This law al- 

 so gives some countenance to the notion, that the quantity of 

 heat in bodies is infinite, compared with all the change that we 

 can effect on it. But it ill agrees with the opinion of MM. Du- 

 long and Petit, that the absolute zero on the common scale may 



