in the Quarterly Journal of Science. 325 



which denote nothing that is true in nature. It is therefore 

 absurd to contend from the numerical results obtained in 

 some extreme cases, that the formulas are consonant to fact 

 in no case whatever ; I say, the formulas, for the argument ap- 

 plies equally to both. Suppose that the original volume is 

 reduced to a very small space, or to a point ; then, according 

 to the centigrade scale, 



t = -266°-7, I = -100°. 

 Now these numbers are either both true, or neither of them 

 is so ; and the latter may be affirmed, since it is very impro- 

 bable, that a law, which has been verified only for a small part 

 of the thermometrical scale, will continue to hold good without 

 limit ; it is even certain that it will not. Next let us suppose 

 that the volume of air has decreased to half its original quan- 

 tity ; then, T r= — 133°, i = - 50°. 

 There is good reason to think that these numbers are not far 

 from the truth ; but it would be rash to affirm that they agree 

 exactly with the phenomenon ; for there is no proof, at least 

 I am not aware of any, that a mass of air at the temperature 

 zero, being cooled down to half its bulk, will still preserve the 

 same capacity for heat. If we suppose a greater diminution 

 of volume than one half, we are entirely ignorant of the man- 

 ner in which the temperature and the volume vary in regard 

 to the absolute heat ; and, as the principle of the investigation 

 now ceases to be exact, the conclusions obtained, whether ex- 

 pressed in algebraic language or otherwise, must no longer 

 be applied. 



All Mr. Meikle's objections to my doctrine are derived from 

 the extreme cases just mentioned. His arguments have no 

 force ; since I have always confined my speculations to the 

 limits within which the thermometer can be reckoned an ex- 

 act measurer of heat. He finds many inconsistencies, and he 

 descants on this topic with so much politeness, that he seems 

 seriously to think, his remarks have some foundation. If such 

 be the case, it will be allowed that the same ability does not 

 attend the same person on all subjects and on all occasions ; 

 for we can here recognise very little of that acuteness and sa- 

 gacity from which we expect the thorough reform. 



M. Poisson has treated this subject in an able memoir in 

 the Conn, des Temps 1826. His equations agree with the 

 doctrine here delivered as far as the formula (7), p. 264, which 

 is derived from an integral to which I have objected. On the 



preceding page he arrives at this equation, — as k — 1, k 



standing for the same value as in this article, and co and n 

 being the variations of latent heat and temperature arising 



from 



