Fluid Density of certain Metals. 363 



The diagram No. 2 is reduced from the oncosimeter diagram, 

 experiment 48. 



A mean of three experiments on tin, made on Mallet's 

 method, gave 6*974 as the liquid density — the results being 

 6-903, 6-982, and 6-976 respectively. 



Zinc. 



Three experiments were made with the oncosimeter on this 

 metal, the results being tabulated below: — 



No. of 



expt. 



Weight, in L •<»„_„ 



° bpecincgra- 



Diameter ^T^^y of told 



of ball, i ^f 111 ^ iballindu- 

 . . , ' tbe stem ! -,.' ,, 

 m inches. <?„.,. „t, ; cung the 

 | for attach- ^ 



ment. 



! 



Sinking 

 effect on 

 first im- 

 mersion, 

 in troy 

 ounces. 



Deduced 

 specific gra- 

 vity of fluid 

 [metal. 



Remarks. 



49. 

 50. 

 51. 



3 

 3 

 3 



52-52 

 52-58 

 52-32 



6-893 

 6-901 

 6-903 



2-4 

 3-3 

 3-75 



6-57 [ 



6-47 



64 



Zinc ball in 

 fluid zinc. 



do. 

 do. 



Specific gravity of solid zinc 6 - 8 to 



Mean 648 



7-2 (KanMne). 



The crucible was kept in the furnace during the three 

 experiments, and maintained as nearly as possible at the same 

 temperature throughout. 



Only one determination of this metal has been made by 

 Mallet's method, giving a liquid density of 6'55. 



Silver. 

 We looked with much interest to the experiments on the 

 density of molten silver for several reasons — mainly from the 

 fact that it melts at a high temperature, which has been deter- 

 mined with great care by the late Henri Ste Claire Deville, 

 who in his later experiments on the subject fixed the tempe- 

 rature at 940° C* "We operated on a thousand ounces of 

 silver, which proved on assay to contain 998 parts of pure 

 silver in 1000. We have as yet only secured one trustworthy 

 result, as we found it very difficult so to adjust the tempera- 

 ture of the molten silver as to prevent the silver balls from 

 melting either too rapidly or too slowly. In the experiment 

 No. 52, in which we did secure a result, the metal must have 

 been just about its melting-point, as at the end of four 

 * Comptes Rendus, t. xc. (1880) p. 773. 



