50 On some Thermodynamical Relations. 



and somewhat rapidly. Again, the absolute temperatures of 

 mercury cannot be calculated from those of water by means 

 of the equation ~R! = 1& + c(t' — t), for if Regnault's data are 

 correct, c is not a constant. There appeared, however, to be 

 good reason to believe that Regnault's values were not reliable, 

 and he himself admits that great difficulties were experienced 

 in the investigation of this substance. 



It was, therefore, considered desirable to attempt a re- 

 determination of the vapour-pressures of mercury with the 

 greatest possible accuracy at a few temperatures. This has 

 been accomplished, and the results are fully described in 

 Trans. Chem. fSoc. Jan. 1886. It will be sufficient here to 

 mention, that new determinations were made at the boiling- 

 points of methyl salicylate and of sulphur under atmospheric 

 pressure (220° and 450° approximately). Two other determi- 

 nations at the boiling-point of bromonaphthalene under atmo- 

 spheric and reduced pressure (280° and 270°), described in 

 the Trans. Chem. Soc. xlvii. p. 640, and four determinations of 

 the boiling-point of mercury under atmospheric pressure by 

 Regnault (Mdmoires de V Academie, xxi. p. 230), were also 

 made use of in the calculations. 



On comparing the absolute temperatures of water with 

 those of mercury at these five pressures, it was found that the 

 value of c in the equation 'R' = 'R + c(t , —t) was constant, and 

 that mercury did not form an exception to the law, which 

 has been proved to hold good for all the other substances 

 examined. 



By making the absolute temperatures of mercury ordinates, 

 and the ratios of the absolute temperatures of mercury to 

 those of water at the same pressures abscissae, and drawing 

 a straight line through the five experimentally-determined 

 points, the value of c was ascertained, and vapour-pressures of 

 mercury were calculated for each 5°, from 270° to 520° C. 

 It should be mentioned that slightly different results were 

 obtained when the absolute temperatures of water were made 

 the ordinates; but within the limits of temperature just men- 

 tioned, the difference was unimportant. 



With the absolute temperatures of mercury as ordinates, 

 the value of c is -0004788. 



The values of ~ . t for mercury were then determined; for 



the most part by the method of tangents, but between 150 

 and 700 mm. from the vapour-pressures calculated for each 

 degree of temperature, and were compared with those of water 

 and of carbon bisulphide. The results are given in the 

 following table : — 



