58 
THE VOYAGE OF H.M.S. CHALLENGER. 
Column VI., D — D r . 
Column VII., the sp. gr. 4 S, calculated as S c . 
Column VIII., the sp. gr. 4 S, found, as S (plain). 
Column IX., S — S c . 
Column X., the adopted sp. gr. of pure water at t°, water of 4° = 1000, as W t. 
The numbers in Column X. are taken from the table appended by Prof. Thorpe to 
his Manual of Quantitative Analysis, as being “compiled from the experiments by 
k "]>p, Pierre, Despretz, Hagen, Matthiesen, Weidner, Kremers, and Rosetti.” A com- 
pirison of these numbers with the corresponding entries in a table given in Naumann’s 
■ lit ion of the first volume of Gmelin’s chemistry, shows that up to 30° the values are 
Rosetti’s. The values r at the foot of the two difference columns were calculated by 
the well-known “probable error” formula — 
r= dbO"G745 l~ . 
sin— 1 
The calculation of the values D and S was effected according to the formula : — 
ft _ w 
— — --—^ =D=a + bt + ct 2 . 
X 
ci= 1-459 93 . . log. a = 0-164 3304 
b= -0 005 5922 . . log. 6 = 3-747 5827 
c — + "000 0G494 . . log. c = 5-8I2 5190 
Tin ■ xo llent work of the Norwegian chemists enabled me to check my results in two 
directions. 
'! . determined the relation between \ and . from the data in Columns II. and 
I. of the following table, from which I calculated the ratios * - * -- * ■ — given in 
X 
Column III. : — 
I. 
ii. 
III. 
102670 
19-47 
1-3713 
1017-39 
12-71 
1-3682 
1025-73 
18-68 
1-3774 
1026-76 
19-56 
1-3681 
1024-88 
18-09 
1-3753 
1026-69 
19-47 
1-3708 
1026-55 
19-38 
1-3699 
Mean, 
. . - . 
1-3716 
