404 Mr, llerapath on the CauseSy Laws, and principal [Junje^ 



( V_ilLliLl!_ .) — 448. Hence if equal portions of the 



same gas be brought to the temperatures of water freezing and 

 boiUng, and mixed together, F = 32, F, = 212, and F,, will be 

 found = 118a*^. Between these temperatures, the variations in 

 volume of mercury and gas have been found to be perfectly- 

 equal ; therefore, if, instead of gas, equal portions of mercury^ 

 water, or any otiier fluid be mixed together, the temperature of 

 the mixture by Fahrenheit's mercurial thermometer should be 

 118^°. De Luc, as I have read in Dr. lire's paper published in 

 the Transactions of the Royal Society for 1818, by mixing equal 

 portions of water at those temperatures, obtained a mean of 1 19°y 

 which so nearly agrees with what has been computed from 

 theory, that the difference may reasonably be ascribed to the 

 observation. It must, however, be confessed, that the experi- 

 ments of Dr. Crawford do not confirm rhose of M. De Luc. 

 According to that philosopher, the resulting temperature is 

 122°; that is, the arithmetical mean of 212° and 32°. It is, 

 however, a question, whether the experiments of De Luc are not 

 entitled to considerably more confidence than those of Craw- 

 ford. The object of the two philosophers was the same ; 

 namely, to determine, whether equal increments of temperature 

 are accompanied with equal expansions of volume ; and hence I 

 should think that De Luc, as soon as he perceived the anomaly^ 

 left no means unessayed, which repeated and careful experi- 

 ments could afford him, to satisfy himself of its existence, before 

 he ventured to publish a circumstance so contrary to the general 

 opinion of philosophers. But with respect to Dr. Crawford (an 

 account of whose experiments also on this subject I have unfor- 

 tunately not seen), he appears, from M. Biot'sTraite de Physique, 

 tom. iv. p. 700, not to have followed one of the most likely 

 methods to ensure accuracy : *' II acherche a le faire (that is, a 

 correction for the loss of temperature) en observant la loi du 

 refroidissement progressive du melange, et s'en servant pour 

 restituer, par le calcul, le chaleure qu'il avait du perdre dans les 



Eremiers momens.'* It would, in my opinion, have been much 

 etter to prevent, than to have had to account for, this loss. 

 But without discussing the merits of the methods of either phi- 

 losopher, the subject of determining v.hether the increments of 

 temperature and volume are proportional, as it respects future 

 discoveries, is of the first consequence ; and I, therefore, beg to 

 submit the settling of so important a point to the consideration 

 of the Royal Society ; and shall feel myself honoured if it appear 

 worthy of their attention. I should not trouble the Royal 

 Society if I had apparatus upon which I could depend for trying 

 an experiment of such delicacy. Unhappily I have not only no 

 thermometers myself sufficiently good, but I know of no one ia 



