146 Royal Society : — 



Such redetermination was not superfluous, as appeared from a 

 collation of the statements to be found in various standard works. 

 In the tables of specific gravities compiled by Prof. F. W. Clarke, 

 and published by the Smithsonian Institution*, there are four 

 authorities quoted, with the numbers given by these, as follows : — 



Sp. gr. of solid mercury. 



Schulze 14-391 



Biddle , 14'485t at —60° C. 



Kupffer and Cavallo 14 (approx.) 



Joule 15*19 



The last of these numbers, on reference to the original paper t, 

 turns out to represent no actual experiment with mercury itself, 

 but is the density calculated for this metal from the examination 

 of a number of amalgams. Kupffer and Cavallo do not profess to 

 give the exact density, but merely state it as about 14, the number 

 apparently resting on no special experiment, though I have not 

 been able to verify this by reference to their paper §. The only 

 other apparently independent statement I have met with occurs in 

 the ' Annuaire du Bureau des Longitudes ' for 1876 (p. 385), where 

 the density 14*39 is given on the authority of Eivot ; but I have not 

 been able to find any reference to a paper by him bearing on this 

 or any analogous point, and it seems probable that we have here 

 only a reproduction of Schulze's result. In different handbooks of 

 chemistry and physics numbers between 14 and 15 are given as 

 approximations, but with no other authority than some of the above. 

 Some of the best and most recent works simply state that mercury 

 undergoes considerable contraction in freezing. Hence our know- 

 ledge on this subject appears hitherto to have rested on the experi- 

 ments of Schulze and Biddle, both of which date back to the 

 early years of the present century. Schulze's paper was published 

 in ' Gehlen's Journal,' vol. iv. p. 434, and therefore about 1807 or 

 1808; and Biddle'sjl belongs to the year 1805. I have had access 

 to neither ; but the character of the instrumental means (balances, 

 thermometers, &c.) generally available at the time the experiments 

 were made, and the then imperfect knowledge of the constants 

 needed for corrections to be applied, make it unlikely that very 

 exact results could have been obtained. Biddle alone seems to have 

 noted the temperature of the frozen mercury, and Branded ex- 

 presses doubt that this was determined with much accuracy. The 

 temperature — 60° C, if correctly quoted, is in itself somewhat 

 improbable. 



* •' The Constants of Nature.— Part I." Smithsonian Miscell. Coll. 255, 

 Washington, 1873, p. 24. 



t 14-465 as quoted by Brando in his ' Manual of Chemistry.' 



X Chem. Soc. Journ. [2] i. p. 387. 



§ Quoted at second hand from Playfair and Joule, " On Atomic Volume and 

 Specific Gravity," Chem. Soc. Mem. 2 (1845), p. 401, and 3 (1848), p. 57. 



II Nicholson's Journal, vol. x. p. 253, and Tilloch's Philos. Mag. vol. xxx. 

 p. 137. 



^ Manual of Chemistry, vol. i. p. 970. 



