1821 .] Causes of Calorific Capacity, Latent Heat, fyc. 101 



notions of equal increments of heat being accompanied with 

 equal augmentations of volume. 



Thus in three experiments made by two different individuals, 

 at different times, with different instruments, widely different 

 fluids, and, perhaps, different methods of making the experi- 

 ments, the errors of a theory which I had investigated before I 

 knew that any experiments of the kind had been made, or even 

 before I had apparatus, or thought of making experiments of 

 this kind myself, are only — °T — °*6 and + °'6 ; while the errors of 

 the old theory are + 2-5° + 3° and + 10-7°. Philosophers wilL 

 know how to appreciate the merits of a theory so investi- 

 gated and so continued. Other facts I have no doubt I shall 

 be able to adduce before I get to the end of this paper equally 

 satisfactoiy and equally convincing of the soundness and the- 

 accuracy of the views 1 have taken. 



It must not be concealed, that Dr. Crawford has made some 

 experiments which have been thought to invalidate those of 

 De Luc. " That philosopher exposed a thermometer equally, it 

 is said, to the influence of air heated by watery vapour to 212°, 

 and of air cooled by snow to 32°, and found the thermometer 

 thus exposed stand something lower than 122°, but above what 

 De Luc has stated." Hence Crawford concludes, that the 

 mercurial thermometer, at least between the temperatures of 

 water freezing and water boiling, is a correct indication of the 

 temperature. Probably few men were better qualified, on 

 account of his care and ingenuity, for deciding a question of this 

 importance than Crawford, but certainly no one ever employed 

 in a subject of consequence a more inaccurate and inefficient 

 method than he has in this. Of all bodies, the aeriform are the 

 most improper he could have selected for experiments of 

 this kind. The extreme facility with which they acquire 

 the temperature of any bodies they may come in contact with ; 

 the great difficulty with which they communicate the tem- 

 perature of others ; and the almost utter impossibility of appre- 

 ciating the equality of their influence ; conspire to render every 

 thing deduced from such experiments suspicious and equivocal. 

 But in spite of all this, Crawford found that his experiments 

 favoured the general conclusion of De Luc ; namely, that the 

 true mean is beneath Fahrenheit's arithmetical. De Luc 

 undoubtedly took the most direct and simple course for settling 

 this point that could be devised ; and had he chosen a different 

 fluid, and made his experiments embrace a greater range, he 

 would most probably have discovered the true laws of temper- 

 ature. 



M. Biot, in his Tiaite de Physique, thinks, that as the coeffi- 

 cient indicating "the specific heat of a body" is constant from 

 the melting of ice to the boiling of water, " we ought to conclude 

 that the degrees of the mercurial thermometer between these 

 limits are proportional to the increments of temperature." In 

 this M. Biot does not consider that he is endeavouring to esta- 



